A University of Sussex PhD thesis 

Available online via Sussex Research Online: 

http://sro.sussex.ac.uk/   

This thesis is protected by copyright which belongs to the author.   

This thesis cannot be reproduced or quoted extensively from without first 
obtaining permission in writing from the Author   

The content must not be changed in any way or sold commercially in any 
format or medium without the formal permission of the Author   

When referring to this work, full bibliographic details including the 
author, title, awarding institution and date of the thesis must be given 

Please visit Sussex Research Online for more information and further details   
 Classification of Arcs in Finite
Geometry and Applications to
Operational Research

by

Salam Abdulqader Falih Alabdullah

A thesis submitted in partial fulfilment of the requirements for

the degree of Doctor of Philosophy

in the

School of Mathematical and Physical Sciences

University of Sussex

June, 2018
 Declaration of Authorship
I hereby declare that the work represented in this thesis is entirely my own unless
otherwise stated, and has been not presented for examination, in whole or in part, to
this or any other university.

Signed:

Date:

i
To The Souls Of My Parents
 Abstract
In PG(2, q), the projective plane over the field Fq of q elements, a (k, n)-arc is a set K
of k points with at most n points on any line of the plane. When n = 2, a (k, 2)-arc is
called a k-arc. A fundamental question is to determine the values of k for which K is
complete, that is, not contained in a (k + 1, n)-arc. In particular, what is the largest
value of k for a complete K, denoted by mn (2, q)?

This thesis focusses on using some algorithms in Fortran and GAP to find large com-
plete (k, n)-arcs in PG(2, q). A blocking set B is a set of points such that each line
contains at least t points of B and some line contains exactly t points of B. Here, B is
the complement of a (k, n)-arc K with t = q + 1 − n. Non-existence of some (k, n)-arcs
is proved for q = 19, 23, 43. Also, a new largest bound of complete (k, n)-arcs for prime
q and n > (q − 3)/2 is found. A new lower bound is proved for smallest size of complete
(k, n)-arcs in PG(2, q). Five algorithms are explained and the classification of (k, n)-
arcs is found for some values of n and q. High performance computing is an important
part of this thesis, where Algorithm Five is used with OpenMP that reduces the time
of implementation. Also, a (k, n)-arc K corresponds to a projective [k, n, d]q -code of
length k, dimension n, and minimum distance d = k − n. Some applications of finite
geometry to operational research are also explained.
 Introduction
A projective plane of order q consists of a set of q 2 + q + 1 points and a set of q 2 + q + 1
lines, where each line contains exactly q +1 points and two distinct points lie on exactly
one line. It follows from the definition that each point is contained in exactly q + 1
lines and two distinct lines have exactly one common point.

A (k, n)-arc is a set K of k points, such that there is some n but no n + 1 are collinear,
where n ≥ 2; a (k, n)-arc is complete if there is no (k + 1, n)-arc containing it.

Many studies have been done to find all complete (k, n)-arcs in PG(2, q) for some values
of q. The size of the largest and the second largest complete (k, n)-arc are denoted by
mn (2, q) and m0n (2, q); also the minimum size of a complete (k, n)-arc is denoted by
tn (2, q). The value of m2 (2, q) is q + 1 for odd q and q + 2 for even q.

In this thesis, a t-fold blocking set B in PG(2, q) is a set of points such that each line
contains at least t points of B and some line contains exactly t points of B. A 1-fold
blocking set is called a blocking set. A 2-fold blocking set is called a double and a
3-fold blocking set is called a triple blocking set. A blocking set is the complement of a
(k, n)-arc K in PG(2, q) with t = q + 1 − n. The smallest blocking sets are just the lines
and any blocking sets containing a line will be called trivial. A blocking set is said to
be minimal, when no proper subset of it is a blocking set. A new largest bound for a
(k, n)-arc in PG(2, q), for n > (q − 3)/2 and prime q, has been proved and applied to
PG(2, 47) in Chapter 2.

A new lower bound for the smallest complete (k, n)-arc in PG(2, q) has been found
and applied for t2 (2, q) and t3 (2, q). Also, constructions of complete k-arcs from a
quadrangle and the configuration of the union of two conics are found. The classification
of (k, n)-arcs in PG(2, q) for some q and n is done using four different methodologies.
Also, a comparison among these methodologies is done to show which method is best
according to the time of implementation to get the final results. All these results are
presented in Chapter 3. High Performance Computing (HPC) technique is used in
Chapter 4 to accelerate the calculations without affecting the accuracy of the results.
Here HPC depends on dividing the big problem into smaller problems and solve each
of them individually.
The relationship between coding theory and finite projective spaces is presented in
Chapter 5. A linear [k, n, d]-code is an n-dimensional subspace of the k-dimensional
vector space V (k, q) with non-zero vectors having weight at least d. An important
problem in coding theory is that to optimise one of the parameters k, n, d for given
values of the other two and fixed q. So, e errors can be corrected for a code with mini-
mum distance at least 2e + 1. Chapter 6 provides general definitions and the historical
development of operational research. In addition, applications of finite geometry to
operational research and a generalisation of Kirkman’s problem and the golf problem
are considered in Chapter 7.

Aim and Objectives

The aim of this research study is investigate structures in finite geometry and apply
the results to operational research.

The objectives are the following.

1. To find the largest and the smallest complete (k, n)-arcs in PG(2, q).

2. To establish some good algorithms to find the classification of (k, n)-arcs in

PG(2, q) in a short time.

3. To apply the results of (k, n)-arcs to coding theory.

4. To apply the results to operational research.

Thesis Structure
This thesis is organised into seven chapters.

Chapter 1: Background

This chapter includes the general definitions of finite geometry. It also includes some
important theorems and lemmas which are used to prove new lower bounds of the
smallest and the largest complete (k, n)-arcs in PG(2, q). The stabiliser groups and
methods of the classification of the largest size of (k, n)-arcs are also given in this
chapter.
Chapter 2: Blocking Sets

This chapter provides some basic equations about blocking sets, and includes an overview
of previous studies in this area. A significant part of this chapter focusses on the proof
of non-existence of some arcs in PG(2, q) and finding new largest bounds of (k, n)-arcs.

Chapter 3: Classification of (k, n)-arcs in PG(2, q)

This chapter includes a new lower bound for the smallest complete (k, n)-arc in PG(2, q)
with a comparison to previous bounds. Four algorithms have been used in this chapter
to find large sizes of (k, n)-arcs with respect to the time of implementation.

Chapter 4: High Performance Computing (HPC)

This chapter includes some details of the High Performance Computing (HPC) tech-
niques used to accelerate the calculations without affecting the accuracy of the results.
In this chapter the paralleling computing systems can be classified as shared memory
(OpenMP) and distributed memory (MPI).

Chapter 5: Coding Theory

This chapter presents the relationship between projective geometry and coding theory.
MDS codes of dimension three and codes of dimension five are considered.

Chapter 6: Operational Research

This chapter provides general definitions and the historical development of operational
research. Also, the concept and the approach of operational research have been ex-
plained.

Chapter 7: Application of Finite Geometry to Operational Research

This chapter highlights new applications of finite geometry to operational research. A

generalisation of Kirkman’s Problem is also provided.
 Acknowledgements
First of all, I would like to thank Allah for support. Also, I am much indebted to
Professor James Hirschfeld, who supervised the work in this thesis. He has give me a
lot of advice, particularly in the logic of mathematics and LaTeX. He has led me to do
my best and put me in optimal working conditions. To be honest, during my studies,
he has been the most important source of academic inspiration, technical advisor and
mentor of the research methodology for this study. This thesis could not have been
completed without his support. I am so proud because he is my supervisor.

Also, I am very grateful to the Iraqi government represented by the Ministry of Higher
Education and Scientific Research which allowed me to complete my Ph.D. in the
United Kingdom. Furthermore, many thanks also go to the staff of the Iraqi Cultural
Attache in London for their support during my Ph.D. study. I also, wish to express
sincere thanks to the University of Basra and colleagues for their help and support.

I would like to express my gratitude to my lovely family, especially my wife Dina and
my beautiful daughters Fatimah, Faten and Sama, who made my life attractive. Their
support made me strong and has helped me to finish my thesis in the identified time.
Also, I want to thank my brothers and sisters in Iraq who prayed for me and supported
me when I was young.

Gratitude also extended to the members of staff within the School of Mathematical and
Physical Sciences, especially Ms. Rebecca Foster and Mr. Richard Chambers who gave
me tremendous support, device, and help. Furthermore, many thanks to Mr. Jeremy
Maris and Mr. Leonardo Rojas for helping me to use the cluster. Many thanks is also
due to my colleague Mr. Lam Duong who did not hesitate to help me. Many thanks
are also due to the University of Salford to give me permission to use the library and
provide me with all necessary books and theses which I needed. Moreover, I would
like to thank my best friend Mr. Ahmad Alsahlani and his wife, Mr. Will Bailey, Mr.
Muhammad Abdullah and Mr. Taha Al-Obaidi for their help. Also, I would like to
thank Mr. Buraq Abdul-Qader for his help to understand the HPC. Finally, I am so
grateful to my colleague Dr. Mohannad Al-Temimi for his help.

vii
Contents

Declaration of Authorship i

Acknowledgements vii

Contents vii

List of Figures xi

List of Tables xii

1 Background 1
1.1 Group Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Finite Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Projective Spaces Over a Finite Field . . . . . . . . . . . . . . . . . . . 5
1.4 Affine Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.5 Projective Planes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.6 Affine Planes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.7 Projective Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.8 Conics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.8.1 Conics in Planes of Even Order . . . . . . . . . . . . . . . . . . 12
1.8.2 Conics in Planes of Odd Order . . . . . . . . . . . . . . . . . . . 13
1.9 Arcs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.10 Stabiliser Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.11 Methodology to Find mn (2, q) . . . . . . . . . . . . . . . . . . . . . . . 24
1.11.1 Method 1 [60], [2] . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1.11.2 Method 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

2 Blocking Sets 27
2.1 Some Basic Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.2 Known Lower Bounds for B . . . . . . . . . . . . . . . . . . . . . . . . 30
2.3 Non-existence of some (k, n)-Arcs in PG(2, q) . . . . . . . . . . . . . . 32
2.3.1 Non-existence of some arcs in PG(2, 19) . . . . . . . . . . . . . 32
2.3.2 Non-existence of some arcs in PG(2, 23) . . . . . . . . . . . . . 33
2.3.3 Non-existence of some arcs in PG(2, 43) . . . . . . . . . . . . . 35

viii
Contents ix
2.4 New Largest Bound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.5 Application of Theorem 2.15 . . . . . . . . . . . . . . . . . . . . . . . . 40
2.6 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

3 Classification of (k, n)-arcs in PG(2, q) 44

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.2 Known Results for mn (2, q) and tn (2, q) in PG(2, q) . . . . . . . . . . . 45
3.2.1 The classification of k-arcs in PG(2, q) for q ≤ 29 . . . . . . . . 45
3.2.2 The classification of (k, 3)-arcs in PG(2, q) for q ≤ 13 . . . . . . 48
3.3 New Lower Bounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.3.1 Comparison with known results . . . . . . . . . . . . . . . . . . 52
3.4 Construction of Complete k-Arcs in PG(2, q) from a Quadrangle . . . . 56
3.5 Algorithms for the Classification of (k, n)-Arcs in PG(2, q) . . . . . . . 58
3.5.1 Algorithm One . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.5.2 Algorithm Two . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.5.3 Algorithm Three . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.5.4 Classification of (k, n)-arcs in PG(2, q) using Algorithm Three . 67
3.5.5 Algorithm Four to find the classification of (k, n)-arcs . . . . . . 78
3.6 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

4 High Performance Computing (HPC) 88

4.1 Parallel Computing on a Shared Memory Architecture - OpenMP . . . 88
4.2 Parallel Computing on a Distributed Memory Architecture - MPI . . . 91
4.3 Algorithm Five to Find the Classification of (k, n)-Arcs . . . . . . . . . 93
4.3.1 Application of Algorithm Five Using OpenMP . . . . . . . . . . 94
4.4 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

5 Coding Theory 95
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
5.2 Linear Codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
5.3 The Main Coding Problem . . . . . . . . . . . . . . . . . . . . . . . . . 97
5.4 MDS Codes of Dimension Three . . . . . . . . . . . . . . . . . . . . . . 98
5.5 Codes of Dimension Five . . . . . . . . . . . . . . . . . . . . . . . . . . 98
5.6 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

6 Operational Research 100

6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
6.2 Development of Operational Research . . . . . . . . . . . . . . . . . . . 100
6.3 The Concept of Operational Research . . . . . . . . . . . . . . . . . . . 101
6.4 The Operational Research Approach . . . . . . . . . . . . . . . . . . . 102
6.5 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

7 Application of Finite Geometry to Operational Research 105

7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
7.2 Application to Probability . . . . . . . . . . . . . . . . . . . . . . . . . 105
Contents x

7.3 Application of Projective Planes to Scheduling Problems . . . . . . . . 110

7.3.1 A simple Schedule Obtained via a Projective Plane . . . . . . . 110
7.3.2 Constraint Satisfaction Problem (CSP ) . . . . . . . . . . . . . . 111
7.4 Generalisation of Kirkman’s Problem . . . . . . . . . . . . . . . . . . . 116
7.5 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

A Points and Lines 120

A.1 Projective Plane of Order Five . . . . . . . . . . . . . . . . . . . . . . . 120
A.2 Projective Plane of Order Seven . . . . . . . . . . . . . . . . . . . . . . 122
A.3 Projective Plane of Order Eight . . . . . . . . . . . . . . . . . . . . . . 124
A.4 Projective Plane of Order Nine . . . . . . . . . . . . . . . . . . . . . . 125
A.5 Projective Plane of Order Eleven . . . . . . . . . . . . . . . . . . . . . 126
A.6 Projective Plane of Order Thirteen . . . . . . . . . . . . . . . . . . . . 127
A.7 Projective Plane of Order Sixteen . . . . . . . . . . . . . . . . . . . . . 129
A.8 Projective Plane of Order Seventeen . . . . . . . . . . . . . . . . . . . . 132

B GAP Program 135

B.1 GAP Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

C Fortran Program 140

C.1 Fortran Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
C.1.1 Algorithm Two . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
C.1.2 Algorithm Five . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

Bibliography 158
List of Figures

1.1 Fano plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

3.1 Time required for implementation of Algorithm Three and Algorithm Four 87

4.1 Shared memory architecture . . . . . . . . . . . . . . . . . . . . . . . . 89

4.2 Threads diagram in OpenMP program . . . . . . . . . . . . . . . . . . 90
4.3 Distributed memory architecture . . . . . . . . . . . . . . . . . . . . . 91
4.4 Example of MPI data fraction . . . . . . . . . . . . . . . . . . . . . . . 92
4.5 Time required for implementation of Algorithm One, Algorithm Three,
Algorithm Four and Algorithm Five . . . . . . . . . . . . . . . . . . . . 93
4.6 Time required for implementation of Algorithm Five . . . . . . . . . . 94

7.1 Coordinates of the points in AG(2, 3) . . . . . . . . . . . . . . . . . . . 107

7.2 Lines in AG(2, 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
7.3 Class 1 of AG(2, 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
7.4 Class 2 of AG(2, 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
7.5 Class 3 of AG(2, 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
7.6 Class 4 of AG(2, 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

xi
List of Tables

1.1 Primitive cubic polynomials [37] . . . . . . . . . . . . . . . . . . . . . . 4

1.2 Type of elements of PG(n, q) . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 Groups of order 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.4 Groups of order 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.5 Groups of order 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.6 Groups of order 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.7 Groups of order 10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.8 Groups of order 12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.9 Groups of order 14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.10 Groups of order 16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.11 Groups of order 18 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.12 Groups of order 20 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.13 Groups of order 21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.14 Groups of order 22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.15 Groups of order 24 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.16 The i-secant distribution of (k, 4)-arcs in PG(2, 8) . . . . . . . . . . . . 26

2.1 Lower bounds for t-blocking sets without condition . . . . . . . . . . . 31

2.2 Lower bounds for t-blocking sets with condition . . . . . . . . . . . . . 31
2.3 The values of τ6 for i ≤ 8 . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.4 The values of τ6 for i ≤ 10 . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.5 The values of n that make µ a non-integer . . . . . . . . . . . . . . . . 35
2.6 The values of n that make µ an integer . . . . . . . . . . . . . . . . . . 35
2.7 The values of τ14 for i ≤ 22 . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.8 The values of n when µ is not an integer . . . . . . . . . . . . . . . . . 41
2.9 The values of n when µ is integer . . . . . . . . . . . . . . . . . . . . . 42
2.10 The values of 14 ≤ r ≤ 36 . . . . . . . . . . . . . . . . . . . . . . . . . 42

3.1 The classification of k-arcs in PG(2, q) for 4 ≤ k ≤ 18 and 5 ≤ q ≤ 17 . 46

3.2 The classification of k-arcs in PG(2, q) for 4 ≤ k ≤ 7 and 19 ≤ q ≤ 29 . 46
3.3 The classification of k-arcs in PG(2, q) for 8 ≤ k ≤ 30 and 19 ≤ q ≤ 29 47
3.4 Size and number of the smallest complete k-arc, q ≤ 29 . . . . . . . . . 48
3.5 The classification of (k, 3)-arcs in PG(2, q) for 4 ≤ k ≤ 15 and 5 ≤ q ≤ 13 48
3.6 The classification of (k, 3)-arcs in PG(2, q) for 16 ≤ k ≤ 23, 5 ≤ q ≤ 13 49
3.7 Number of t3 (2, q), 5 ≤ q ≤ 13 . . . . . . . . . . . . . . . . . . . . . . . 49
3.8 Bounds for complete k-arcs for 4 ≤ q ≤ 23 . . . . . . . . . . . . . . . . 52
xii
List of Tables xiii

3.9 Bounds for complete (k, 3)-arcs for 4 ≤ q ≤ 16 . . . . . . . . . . . . . . 52

3.10 Lower bounds for complete (k, 3)-arcs for 4 ≤ q ≤ 16 . . . . . . . . . . 53
3.11 Application of Proposition (3.5) for some values of n . . . . . . . . . . 55
3.12 Application of Equation (3.27) in PG(2, 5) . . . . . . . . . . . . . . . . 57
3.13 The classification of (k, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . 60
3.14 The i-secant distribution of (7, 4)-arcs in PG(2, 5) . . . . . . . . . . . . 62
3.15 The i-secant distribution of (8, 4)-arcs in PG(2, 5) . . . . . . . . . . . . 62
3.16 The i-secant distribution of (9, 4)-arcs in PG(2, 5) . . . . . . . . . . . . 63
3.17 The i-secant distribution of (10, 4)-arcs in PG(2, 5) . . . . . . . . . . . 63
3.18 The i-secant distribution of (11, 4)-arcs in PG(2, 5) . . . . . . . . . . . 64
3.19 The i-secant distribution of (12, 4)-arcs in PG(2, 5) . . . . . . . . . . . 64
3.20 The i-secant distribution of (13, 4)-arcs in PG(2, 5) . . . . . . . . . . . 65
3.21 The i-secant distribution of (14, 4)-arcs in PG(2, 5) . . . . . . . . . . . 65
3.22 The i-secant distribution of (15, 4)-arcs in PG(2, 5) . . . . . . . . . . . 66
3.23 The i-secant distribution of (16, 4)-arcs in PG(2, 5) . . . . . . . . . . . 66
3.24 The classification of (k, 4)-arcs in PG(2, 5) according to their i-secant
distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.25 (6, 2)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.26 (11, 3)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.27 (16, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.28 (16, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.29 (8, 2)-arcs in PG(2, 7) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.30 (15, 3)-arcs in PG(2, 7) . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.31 (22, 4)-arcs in PG(2, 7) . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.32 (29, 5)-arcs in PG(2, 7) . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.33 (10, 2)-arcs in PG(2, 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
3.34 (15, 3)-arcs in PG(2, 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
3.35 (28, 4)-arcs in PG(2, 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
3.36 (33, 5)-arcs in PG(2, 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
3.37 (42, 6)-arcs in PG(2, 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
3.38 (42, 6)-arcs in PG(2, 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
3.39 (10, 2)-arcs in PG(2, 9) . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
3.40 (17, 3)-arcs in PG(2, 9) . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
3.41 (28, 4)-arcs in PG(2, 9) . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
3.42 (37, 5)-arcs in PG(2, 9) . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
3.43 (12, 2)-arcs in PG(2, 11) . . . . . . . . . . . . . . . . . . . . . . . . . . 73
3.44 (21, 3)-arcs in PG(2, 11) . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.45 (32, 4)-arcs in PG(2, 11) . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.46 (43, 5)-arcs in PG(2, 11) . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.47 (14, 2)-arcs in PG(2, 13) . . . . . . . . . . . . . . . . . . . . . . . . . . 75
3.48 (23, 3)-arcs in PG(2, 13) . . . . . . . . . . . . . . . . . . . . . . . . . . 75
3.49 (49, 5)-arcs in PG(2, 13) . . . . . . . . . . . . . . . . . . . . . . . . . . 76
3.50 (18, 2)-arcs in PG(2, 16) . . . . . . . . . . . . . . . . . . . . . . . . . . 76
3.51 (52, 4)-arcs in PG(2, 16) . . . . . . . . . . . . . . . . . . . . . . . . . . 77
List of Tables xiv

3.52 (18, 2)-arcs in PG(2, 17) . . . . . . . . . . . . . . . . . . . . . . . . . . 77

3.53 (28, 3)-arcs in PG(2, 17) . . . . . . . . . . . . . . . . . . . . . . . . . . 77
3.54 (28, 3)-arcs in PG(2, 17) . . . . . . . . . . . . . . . . . . . . . . . . . . 78
3.55 Stabilisers of (7, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . 79
3.56 Stabilisers of (8, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . 79
3.57 Stabilisers of (9, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . 80
3.58 Stabilisers of (10, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . 81
3.59 Stabilisers of (11, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . 82
3.60 Stabilisers of (12, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . 83
3.61 Stabilisers of (13, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . 84
3.62 Stabilisers of (14, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . 85
3.63 Stabilisers of (15, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . 86
3.64 Stabilisers of (16, 4)-arcs in PG(2, 5) . . . . . . . . . . . . . . . . . . . . 86

5.1 MDS Codes over F13 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

5.2 Codes of dimension 5 over F13 . . . . . . . . . . . . . . . . . . . . . . . 99

7.1 Points of PG(2, 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

7.2 Lines of PG(2, 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
7.3 3-secants in AG(2, 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
7.4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
7.5 3-secants in PG(2, 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
7.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
7.7 4-secants in PG(2, 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
7.8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
7.9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
7.10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
7.11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
7.12 8-secants in PG(2, 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7.13 8-secants in PG(2, 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
7.14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
7.15 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
7.16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

A.1 Points of PG(2, 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

A.2 Lines of PG(2, 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
A.3 Points of PG(2, 7) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
A.4 Lines of PG(2, 7) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
A.5 Points of PG(2, 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
A.6 Lines of PG(2, 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
A.7 Points of PG(2, 9) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
A.8 Lines of PG(2, 9) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
A.9 Points of PG(2, 11) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
A.10 Lines of PG(2, 11) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
A.11 P1 , . . . P84 of PG(2, 13) . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
List of Tables xv

A.12 P85 , . . . P183 of PG(2, 13) . . . . . . . . . . . . . . . . . . . . . . . . . . 128

A.13 Lines of PG(2, 13) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
A.14 P1 , . . . P85 of PG(2, 16) . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
A.15 P86 , . . . P200 of PG(2, 16) . . . . . . . . . . . . . . . . . . . . . . . . . . 130
A.16 P201 , . . . P273 of PG(2, 16) . . . . . . . . . . . . . . . . . . . . . . . . . . 131
A.17 Lines of PG(2, 16) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
A.18 P1 , . . . P114 of PG(2, 17) . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
A.19 P115 , . . . P252 of PG(2, 17) . . . . . . . . . . . . . . . . . . . . . . . . . . 133
A.20 P253 , . . . P307 of PG(2, 17) . . . . . . . . . . . . . . . . . . . . . . . . . . 134
A.21 Lines of PG(2, 17) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
Chapter 1

Background

1.1 Group Theory

Definition 1.1. [40] A group is a set G together with an operation ◦ satisfying the
following requirements:

(a) for each pair x, y of elements of G, x ◦ y is an element of G;

(b) for all elements x, y, z of G, (x ◦ y) ◦ z = x ◦ (y ◦ z);

(c) there is an element e ∈ G such that e ◦ g = g = g ◦ e for all g ∈ G;

(d) given an element g ∈ G, there is an element g ∗ ∈ G such that

g ◦ g ∗ = e = g ∗ ◦ g.

A group G is abelian if, for all x, y ∈ G, x ◦ y = y ◦ x. A cyclic group is a group

generated by a single element; that is, a group consisting of all powers of one of its
elements. A finite group G of order n is cyclic if and only if it contains an element of
order n. Also, a group of prime order is cyclic.

Definition 1.2. [40] An action of a group G on a set Ω is a function µ : Ω × G −→ Ω

with the following properties:

(a) ((x, g)µ, h)µ = (x, g ◦ h)µ for all x ∈ Ω and g, h ∈ G;

1
Chapter 1. Background 2

(b) (x, 1)µ = x for all x ∈ Ω, where 1 is the identity of G;

(c) ((x, g)µ, g −1 )µ = ((x, g −1 )µ, g)µ = x for all x ∈ Ω, g ∈ G.

A relationship ∼ on Ω by the rule that x ∼ y if there exist g ∈ G with (x, g)µ = y. So

∼ is an equivalence relation.

Definition 1.3. Let G be a group. A homomorphism θ : G −→ H is a function θ from

G to H that satisfies the condition, for all g1 , g2 ∈ G,

(g1 g2 )θ = (g1 θ)(g2 θ).

A homomorphism that is one-to-one and onto is an isomorphism; then G and H are

isomorphic. A bijective homomorphism θ group to itself is an automorphism.

Definition 1.4. A bijection f : X −→ X is a permutation on X, and the set S(X) of

permutations f is a group under composition of functions.

Definition 1.5. [18]

(i) A permutation ρ is κ-cycle if there exists a positive κ and an integer i such that

(a) κ is the smallest positive integer such that ρκ (i) = i;

(b) ρ fixes each j not in {i, ρ(i), . . . , ρκ−1 (i)}.

(ii) A permutation group is a group whose elements are permutations; that is, a
subgroup of a symmetric group.

(iii) The order of a permutation is equal to the least common multiple of the lengths
of its cycles.

Example 1.1. There are 27 possible maps from the set X = {1, 2, 3} to itself. Most
of these are not injective, only six of these maps are permutations denoted as follows:
       
1 2 3 1 2 3 1 2 3 1 2 3
= , α =  , β =  , γ =  ,
1 2 3 2 3 1 3 1 2 1 3 2
   
1 2 3 1 2 3
σ= , φ =  ,
3 2 1 2 1 3
Chapter 1. Background 3

where the values of each map are given in the second row. Thus the map β takes 1 to
3, β(2) = 1 and β(3) = 2. Since the maps are injective, the entries in the second row
are all distinct. The five non-identity elements of S(3) are cycles, and may be written
as
(123), (132), (23), (13), (12).

Definition 1.6. A group G is a semidirect product of a subgroup N by a subgroup H

if the following conditions are satisfied:

1. G = N H;

2. N is a normal subgroup of G;

3. H ∩ N = {e}.

1.2 Finite Fields

A field is a set F closed under two operations +, × such that

(i) (F, +) is an abelian group with identity 0;

(ii) (F \ {0}, ×) is an abelian group with identity 1;

(iii) a(b + c) = ab + ac, (a + b)c = ac + bc, for all a, b, c ∈ F.

Let Fq denoted a field of q elements. Note that q must always be an integral power
ph of a prime p. Here, p is characteristic of the finite field. For q prime, Fq = Z/qZ.
Here, for all q = ph , there exists α ∈ Fq such that Fq = {0, 1, α, . . . , αq−2 | αq−1 = 1}.
Here, α is the root of a primitive polynomial of degree h over Fp . In Table 1.1, some
primitive polynomials f (x) are given for the fields of prime power order up to 17.
Chapter 1. Background 4

Table 1.1: Primitive cubic polynomials [37]

q f (x) q f (x)

2 x3 + x + 1 9 x3 − βx2 − 1

3 x3 − 2x2 − x − 2 11 x3 − x2 − x − 3

4 x3 − x2 − x − 2 13 x3 − x2 − 2

5 x3 − x2 − 1 16 x3 − γx2 − γ

7 x3 − x2 − 2x − 4 17 x3 − 8x2 − 1

8 x3 − x − α 4

There exists a primitive element s in Fq such that

Fq = {0, 1, s, . . . , sq−2 | sq−1 = 1}.

When q is not prime, the root α of f (x) is such an element. If q = ph , then Fp is the
prime subfield of Fq .

Definition 1.7. An automorphism σ of Fq is a permutation of Fq such that

(x + y)σ = xσ + yσ, (xy)σ = (xσ)(yσ) for all x, y ∈ Fq .

The group Aut(Fq ) of automorphisms of Fq , q = ph , is isomorphic to Zh . It is generated

i
by the Frobenius automorphism φ, where xφ = xp for all x ∈ Fq ; so xφi = xp .
Equivalently, if σ is any automorphism of Fq , then σ = φi for some i. It is occasionally
convenient to write xσ instead of xσ. Thus

Aut(Fq ) = {1, σ, σ 2 , . . . , σ h−1 }.

Since φ is an automorphism, (x + y)p = xp + y p for all x, y ∈ Fq .

Chapter 1. Background 5

1.3 Projective Spaces Over a Finite Field

Let V = V (n + 1, q) be an (n + 1)−dimensional vector space over a field Fq with zero

element 0. Consider the equivalence relation on the elements of V0 = V \ {0} whose
equivalence classes are the one-dimensional subspaces of V with zero removed. Thus,
if X, Y ∈ V0 , then X is equivalent to Y if Y = tX for some t in Fq \ {0}; that is,
yi = txi for all i. Then the set of equivalence classes is the n-dimensional projective
space over Fq and is denoted by PG(n, q). The elements of PG(n, q) are points; the
equivalence class of the vector X is the point P (X). It will also be said that X is a
coordinate vector for P (X) or that X is a vector representing P (X). In this case, tX
with t in Fq \ {0} also represents P (X); that is, by definition, P (tX) = P (X). So,
the points of PG(n, q) can be described in terms of coordinates as in Table 1.2, where
x0 , x1 , . . . , xn−1 ∈ Fq . So

Table 1.2: Type of elements of PG(n, q)

Type of elements Number of elements

P (x0 , . . . , xn−1 , 1) qn
P (x0 , . . . , xn−2 , 1, 0) q n−1
.. ..
. .
P (x0 , 1, 0, . . . , 0) q
P (1, 0, . . . , 0) 1
θ(n, q)

|PG(n, q)| = θ(n, q) = (q n+1 − 1)/(q − 1).

Definition 1.8. Let F be a subfield of the field K.

(i) The dimension of K as a vector space over F is the degree of K over F , and is
denoted by [K : F ].

(ii) K is a finite extension of F if [K : F ] is finite.

(iii) When [K : F ] = 2, . . . , n, then K is a quadratic, cubic, . . . , n-ic extension of F .

Chapter 1. Background 6

Definition 1.9. A collineation of projective plane Π is a map ψ of Π onto Π such that

1. ψ is bijection;

2. ψ maps points onto points and lines onto lines;

3. if P and ` are an incident point and line in Π, then ψ(P ) and ψ(`) are incident.

Theorem 1.10. The Fundamental Theorem of Projective Geometry [37]

(i) If {P0 , . . . , Pn+1 }, {P00 , . . . , Pn+1

0
} are two sets of n+2 points of PG(n, K) such that
no n+1 points chosen from the same set lie in a subspace of dimension n−1, then
there exists a unique projectivity T such that Pi0 = Pi T , for all i ∈ {0, . . . , n + 1}.

(ii) Let S = PG(2, K), and ψ : S −→ S be a collineation, then ψ = σT , where

σ is an automorphism and T is a projectivity. This means that if K = Fq ,
and P (X 0 ) = P (X)ψ, then there exists % ∈ Nh , tij ∈ Fq for (i, j) ∈ Nh2 and
% % % %
tij ∈ Fq \ {0} such that tX ∗ = X p T , where X p = (xp0 , . . . , xpn ) and T =
(tij ), i, j ∈ Nh = {0, 1, . . . , h}.

Definition 1.11. For any positive integer n, the general linear group, GL(n, q), is the
set of all invertible n × n matrices over Fq under matrix multiplication. The order of
the group GL(n, q) is

(q n − 1)(q n − q)(q n − q 2 ) · · · (q n − q n−1 ).

Definition 1.12. The special linear group SL(n, q) is the subgroup of the group
GL(n, q) consisting of those matrices of determinant 1.

Definition 1.13. [37]

(i) The projective general linear group PGL(n, q) is the group of projectivities of
PG(n − 1, q) with respect to the operation of composition of maps.

(ii) The collineation group PΓL(n, q) is the group of collineations of PG(n − 1, q) with
respect to the operation of composition of maps.

Definition 1.14. The projective special linear group PSL(n, q) is the quotient group
SL(n, q)/Z, where Z is the subgroup of scalar matrices in SL(n, q).
Chapter 1. Background 7

1.4 Affine Spaces

If H∞ is any hyperplane in PG(n, K) that is, a subspace of dimension n − 1, then

AG(n, K) = PG(n, K)\H∞ is an affine space of n dimensions over field K. When
K = Fq , write AG(n, K) = AG(n, q). The subspaces of AG(n, K) are the subspaces of
PG(n, K), apart from H∞ , with the points of H∞ deleted in each case. Here H∞ is
referred to as the hyperplane at infinity of AG(n, q).

1.5 Projective Planes

Consider the projective plane PG(2, q) over Fq . So, PG(2, q) contains q 2 + q + 1 points
and q 2 + q + 1 lines. There are exactly q + 1 points on each line, and q + 1 lines through
each point. The points and lines of PG(2, q) satisfy the axioms of a projective plane:

1. every two distinct points are on a unique common line;

2. every two distinct lines contain a unique common point;

3. there are four distinct points, no three of which are on a common line.

The constants k, l, m determine the line `. A point P (x, y, z) is incident with a line
`(k, l, m) if and only if
 
k
 
t
XU = (x y z)  l  = kx + ly + mz = 0.
 
 
m

With every non-singular matrix T = (tij ) ∈ Fq associate a bijection from the point
t
P (X) = P (x, y, z) to P (X 0 ) = P (x0 , y 0 , z 0 ) and the line `(U ) = `(k, l, m) to `(U 0 ) =
t
`(k 0 , l0 , m0 )t with X 0 = XT and U 0 = T −1 U T . In other words,
 
t t t
 00 01 02 
(x0 y0 z 0 ) = (x y z)  t10 t11 t12 ,
 
 
t20 t21 t22
Chapter 1. Background 8

   −1  
0
k t t t k
   00 01 02   
 0  
 l  =  t10 t11 t12  l .
  

     
0
m t20 t21 t22 m

Such a bijection is a projectivity or projective linear transformation. A projectivity

preserves the incidence between points and lines: a point P (X) lies on a line `(U t )
t
if and only if the image P (X 0 ) of P (X) lies on the image `(U 0 ) of `(U t ). Indeed,
t
X 0 U 0 = XT T −1 U t = 0. As with coordinate triples, also the matrix T of a projectivity
is only determined up to a scalar factor.

The group of all projectivities of PG(2, q) is the projective general linear group PGL(3, q)
and has order q 3 (q 3 − 1)(q 2 − 1). From the Fundamental Theorem of Projective Ge-
ometry, a projectivity is uniquely determined by the four images of the vertices of a
quadrangle. In other words, PGL(3, q) acts transitively on ordered quadrangles.

A collineation in PG(2, q) is a bijection mapping points to points and lines to lines,

which preserves incidence. Each projectivity is a collineation. However, in general not
all collineations are projectivities. The Frobenius automorphism mapping the point
P (x, y, z) to P (xp , y p , z p ) is a collineation, but not a projectivity. From the Funda-
mental Theorem of Projective Geometry, let ψ be a collineation, then there exists a
non-singular matrix T and a field automorphism σ, such that
 
t t t
 00 01 02 
ψ : (x y z) 7→ (x y z)σ  t10 t11 t12 ,
 
 
t20 t21 t22

   −1  σ
k t00 t01 t02 k
     
ψ :  l  7→  t10 t11 t12  l  .
     

     
m t20 t21 t22 m

The group of all collineations of PG(2, q), q = ph , is the collineation group PΓL(3, q)
and has order hq 3 (q 3 − 1)(q 2 − 1). So, PGL(3, q) is a subgroup of PΓL(3, q) and

[PΓL(3, q) : PGL(3, q)] = h.

When q is prime, then PGL(3, q) = PΓL(3, q).

Chapter 1. Background 9

Example 1.2. The simplest finite projective plane is PG(2, 2). It consists of seven
points and seven lines with three points on every line and three lines through every
point. This projective plane is the Fano plane and may be illustrated as in Figure 1.1.
The points are A, B, C, D, E, F, G and the lines are ABC, AGE, ADF , CEF , CDG,
BDE, BF G.
Figure 1.1: Fano plane

B D
G

C E F

In all known examples the order q of projective plane is a prime power. It is a fact
that, for any given prime power, there exists at least one projective plane of that order,
namely PG(2, q). It is not known whether planes of non-prime power order exist,
although it is commonly believed that any finite projective plane must have prime
power order.

1.6 Affine Planes

An affine plane AG(2, q) of order q is an incidence structure of points and lines with
the following properties:

1. every two points are incident with a unique line;

2. given a point P and a line ` such that P ∈

/ ` then there exist a unique line m such
that P ∈ m and m ∩ ` = ∅;

3. there are three points that are not collinear;

4. every line is incident with a constant q points and every point is incident with q + 1
lines;

5. an affine plane of order q has q 2 points and q 2 + q lines.

Chapter 1. Background 10

1.7 Projective Lines

The projective line, PG(1, q), has q + 1 points:

PG(1, q) = {(1, 0)} ∪ {(x, 1) | x ∈ Fq }.

Each point P (x, 1) of PG(1, q) can be represented by the non-homogeneous coordinate

x in Fq and the coordinate for P (1, 0) is ∞. These points are often referred as the
parameters t ∈ Fq ∪ {∞}.

A projectivity T of PG(1, q) is given by Y = XT , where X = (x0 , x1 ), Y = (y0 , y1 ) and

 
a b
T = .
c d

From the Fundamental Theorem of Projective Geometry, any three points of PG(1, q)
can be mapped to any other three points by a projectivity. A simple example of how
a projectivity of the line maps points to points is presented below.

Example 1.3. In PG(1, 5) there are six points (0, 1), (1, 1), (2, 1), (3, 1), (4, 1), (1, 0).
Then the projectivity  
2 1
T = 
1 1

maps these points of PG(1, 5) to the points (1, 1), (4, 1), (0, 4), (3, 1), (1, 0), (2, 1).

1.8 Conics

An irreducible conic of PG(2, q) is a set of points C whose coordinates (x, y, z) satisfy

the following equation:

Q(x, y, z) = ax2 + by 2 + cz 2 + dxy + exz + f yz,

with condition that ( ∂F , ∂F , ∂F ) 6= (0, 0, 0) at any points of C and a, b, c, d, e, f ∈ Fq .

∂x ∂y ∂z

Let C = v(Q) = {P (x, y, z) | Q(x, y, z) = 0}.

Chapter 1. Background 11

When q is odd,
Q(x, y, z) = XAX T ,

where X = (x, y, z) and  

a d/2 e/2
 
A =  d/2 b f /2 .
 
 
e/2 f /2 c

Consider a projectivity T ∈ PGL(3, q). Let C denote the conic represented by the
matrix A. Then the image C 0 of C by T is represented by the matrix A0 = T −1 A(T −1 )t .
Indeed,

t
X 0 A0 X 0 = (XT )(T −1 A(T −1 )t )(XT )t

= XT T −1 A(T −1 )t T t X t

= XAX t .

An oval O is a set of q+1 points in PG(2, q), with the property that every line is incident
with at most two points of O. A conic is an example of an oval in PG(2, q). Here the
classification of lines and points of the plane with respect to a conic is considered from
a geometric viewpoint, and then an algebraic method for classifying points and lines
with respect to conic is given. Since no three points of conic C are collinear, any line l in
Π = PG(2, q) can intersect C in at most two points, Hence, l is geometrically classified
in the following way:

1. l is a tangent line to C when |l ∩ C| = 1;

2. l is a secant line to C when |l ∩ C| = 2;

3. l is an exterior line to C when |l ∩ C| = 0.

Note 1.15. [37] For q odd, there are q + 1 tangent lines, 21 q(q + 1) secant lines, and
1
2
q(q − 1) exterior lines.
Chapter 1. Background 12

1.8.1 Conics in Planes of Even Order

The points of the plane are geometrically classified with respect to a conic. If P is a
point in a plane of even order, then P is classified in three ways with respect to C:

1. P is a conic point, when P is on C;

2. P is the nucleus, if P lies on all the tangent lines to C;

3. P is a regular point, if P is not on C and not the nucleus.

Any conic point has one tangent and q secants through it, while the nucleus has q + 1
tangents through it. The remaining points have exactly one tangent, 21 q secants and 12 q
exterior lines through them. The set of q + 2 points obtained from adding the nucleus
to the oval is a hyperoval ; it is a maximal arc of degree 2.

Lemma 1.16 ([37]). In PG(2, q) for q even, the nucleus of the conic

v(ax2 + by 2 + cz 2 + dxy + exz + f yz)

is the point P (f, e, d).

Example 1.4. Let C be a conic in the projective plane PG(2, q), q > 4 and even. Let
N be the nucleus of C, and let H denote the hyperoval which consist of the q + 1 points
of C and the nucleus of C. So, H is a set of q + 2 points with the property that no
three are collinear. By construction, there are no tangent lines to H. Any line in the
plane must be either a secant line or an exterior line. Let O be the oval which consists
of N and any q points of C. Recall that five points, no three collinear, determine a
unique conic. The sets C and O have q points in common and these q points satisfy
a quadratic equation. But, the point N of O does not satisfy this quadratic equation.
Hence, there is no quadratic equation which every point of O satisfies; so O is not a
conic.
Chapter 1. Background 13

1.8.2 Conics in Planes of Odd Order

In planes of odd order, every oval is a conic. Here, every conic is an oval and an oval has
q + 1 points on it, every conic consist of q + 1 points that satisfy some non-degenerate
quadratic equation.

If P is a point in the plane, then P is classified in three ways with respect to C:

1. when P is on C, then P is a conic point;

2. when P is not on C, but on a tangent line to C, then P is exterior point;

3. when P is neither on C nor a tangent line to C, then P is an interior point.

Any conic point has one tangent and q secants through it. An interior point has 12 (q +1)
exterior lines through it, an exterior point has exactly two tangents, 21 (q − 1) secant
and 12 (q − 1) exterior lines through it.

Example 1.5. Consider the conic C whose points satisfy −xy + y 2 + 2yz + z 2 = 0.
The lines x = 0 and y = 0 are tangent lines to C. Since (0, 0, 1) lies on both of these
tangent lines, then (0, 0, 1) is an external point with respect to C. The lines through
(0, 0, 1) are x = 0 and y = mx, for m ∈ Fq . Then y = mx is a secant when m is
non-zero square and is a tangent when m is a non-square element.

Lemma 1.17. A conic in PG(2, q) has the following canonical forms:

(i) v(X02 + X1 X2 ), all q;

(ii) v(X02 − X1 X2 ), all q;

(iii) v(a0 X02 + a1 X12 + a2 X22 ), with a0 a1 a2 6= 0, q odd;

(iv) v(X02 + X12 + X22 ), q odd.

Corollary 1.18 ([37]). In PG(2, q) for q even, the q + 1 tangents to a conic are
concurrent.
P
Theorem 1.19 ([19]). If F = i≤j aij xi xj , then F = v(F ) is singular if and only if
δ = 0, where

δ = 4a00 a11 a22 + a01 a02 a12 − a00 a212 − a11 a202 − a22 a201 .
Chapter 1. Background 14

Theorem 1.20 ([19]). A conic of PG(2, q) is reducible if and only if it is singular.

(1) Suppose a conic C of PG(2, q) is reducible. If it reduces to two distinct lines `

and m, then the point ` ∩ m is a singular point of the conic. If C reduces to
two coincident lines, then every point of C is singular. In either case, the conic is
singular.

(2) Let F = ax2 +by 2 +cz 2 +dxy+f xz+eyz be the equation of a singular conic. Without
loss of generality, let (1, 0, 0) be a singular point of C. Then (1, 0, 0) satisfies F = 0
and also
∂F
= 2ax + dy + f z = 0,
∂x
∂F
= dx + 2by + ez = 0,
∂y
∂F
= f x + ey + 2cz = 0.
∂z
Therefore, a = d = f = 0. The equation of C becomes ψ = by 2 + cz 2 + eyz = 0.
The quadratic ψ is reducible, and therefore the conic C is reducible.

Theorem 1.21 (Segre’s Theorem [57]). If q is odd, every (q + 1)-arc of PG(2, q) is an

irreducible conic.

Example 1.6 ([19]). Let C be a conic of PG(2, q) with equation

x2 + 2xy + y 2 + 4yz + 4z 2 = 0.

(i) Find the equation of the tangent to C at (1, −1, 0).

(ii) Find the tangent of C passing through (1, 0, 0).

Solution From the given equation of the conic, a = 1, b = 1, c = 4, d = 2, e = 0, f = 4.

The matrix A of the given conic is
 
1 1 0
 
A =  1 1 2 .
 
 
0 2 4
Chapter 1. Background 15

(i) The tangent at the point (1, −1, 0) is

  
1 1 0 x
  
(1, −1, 0)  1 1 2 = 0x + 0y − 2z = 0,
  
 y 
  
0 2 4 z

which is the line of equation z = 0.

(ii) For P = (1, 0, 0), then P T AX = x + y and P T AP = 1. So the equation of the

tangents passing through (1, 0, 0) is

(P T AX)2 − (P T AP )(X T AX) = 0,

that is (x + y)2 − (x2 + 2xy + y 2 + 4yz + 4z 2 ) = −4(yz + z 2 ) = −4z(y + z) = 0.

The two tangents passing through (1, 0, 0) are the lines z = 0 and y + z = 0.

1.9 Arcs

Definition 1.22. [37] A (k, n)-arc in PG(2, q) is a set K of k points, no n + 1 of which

are collinear, but with at least one set of n points collinear. When n = 2, a (k, 2)-arc
is called a k-arc.

Definition 1.23. [37] A (k, n)-arc is complete if it is not contained in a (k + 1, n)-arc.

Definition 1.24. [37] A set of lines concurrent at a point P of a projective plane Π is

a pencil of lines. A set of points incident with a fixed line ` in Π is a range of points.
Chapter 1. Background 16

Notation 1.25. For a (k, n)-arc K in PG(2, q), let

τi = the total number of i-secants of K,

ρi = ρi (P ) = the number of i-secants through a point P of K,

σi = σi (Q) = the number of i-secants through a point Q of PG(2, q) \ K,

mn (2, q) = the maximum size of a (k, n)-arc in PG(2, q),

tn (2, q) = the minimum size of a (k, n)-arc in PG(2, q).

Definition 1.26. [37]

(i) The type of a point P on a (k, n)-arc K is the (n + 1)-tuple (ρ0 , ρ1 , . . . , ρn ).

(ii) The i-secant distribution of K is the (n + 1)-tuple (τn , τn−1 , . . . , τ0 ).

Theorem 1.27 ([37]).


q + 2, for q even;

m2 (2, q) =
q + 1, for q odd.


Theorem 1.28 ([37]). (1)


= (n − 1)q + n,

for q even and n | q;
mn (2, q)
< (n − 1)q + n,

for q odd.

(2) A (k, n)-arc K is maximal if and only if every line in PG(2, q) is either an n-secant
or an external line.
Chapter 1. Background 17

Lemma 1.29 ([37]). For a (k, n)-arc K, the following equations hold:

n
X
τi = q 2 + q + 1; (1.1)
i=0
n
X
iτi = k(q + 1); (1.2)
i=1
n
X
1 1
2
i(i − 1)τi = 2
k(k − 1). (1.3)
i=2

Lemma 1.30 ([37]). For a (k, n)-arc K, the following equations also hold:

n
X
ρi = q + 1; (1.4)
i=1
n
X
(i − 1)ρi = k − 1; (1.5)
i=2
n
X
σi = q + 1; (1.6)
i=0
n
X
iσi = k; (1.7)
i=1
X
ρi = iτi ; (1.8)
P ∈K
X
σi = (q + 1 − i)τi . (1.9)
Q∈Π\K

Notation 1.31. (i) Let the Equations (1.4) and (1.5) of Lemma 1.30 have M distinct
solutions
Bj = (ρ1j , . . . , ρnj ), j = 1, . . . , M.

(ii) Let bj be the number of points on the (k, n)-arc K with solution Bj .

(iii) Let the Equations (1.6) and (1.7) have L distinct solutions

Mj = (σ0j , . . . , σnj ), j = 1, . . . , L.

(iv) Let mj be the number of points in PG(2, q) \ K with solution Mj .

Chapter 1. Background 18

Lemma 1.32 ([36]). For a (k, n)-arc K in PG(2, q), the following equations hold:

M
X
bj ρij = iτi ; (1.10)
j=1
M
X
bj = k; (1.11)
j=1
L
X
mj σij = (q + 1 − i)τi ; (1.12)
j=1
L
X
mj = q 2 + q + 1 − k. (1.13)
j=1

Lemma 1.33 ([37]). If K is a complete (k, n)-arc, then (q + 1 − n)τn ≥ q 2 + q + 1 − k,

with equality if and only if σn = 1 for all Q in PG(2, q) \ K.

1.10 Stabiliser Group

If Q is a point of PG(2, q), then Qg denotes the image of Q through the right action of
the non-singular matrix g that induces the projectivity T; that is, T ∈ PGL(3, q). Let
S be a set of points of PG(2, q), possibly an (k, n)-arc. Then S g = {Qg | Q ∈ S}.

Definition 1.34 (Orbit). For G ≤ PGL(3, q), if T ∈ PGL(3, q), then the orbit of a
point Q ∈ PG(2, q) under G is the set QG = {Qg | g ∈ G}. The orbit of S ⊂ PG(2, q)
under G is the set {QG | Q ∈ S}.

Definition 1.35 (Stabiliser). A set S ⊆ PG(2, q) is said to be stabilised by the pro-

jectivity T ∈ PGL(3, q) if and only if S g = S.

To determine the automorphism or stabiliser group of a (k, n)-arc S, first calculate every
projectivity T ∈ PGL(3, q) that maps S to itself; that is S g = S. This is achieved
by finding every g that maps a 4-arc of S onto the frame points, where the frame
points are P (1, 0, 0), P (0, 1, 0), P (0, 0, 1), P (1, 1, 1) and then determining if S g = S.
The stabiliser group of S consists of all such projectivities. If this group has order
i = 1, 2, 3, 5, 7, 11, 13, 15, 17, 19, 23, then it is isomorphic to Zi . If the group has order
i = 4, 6, 8, 9, 10, 12, 14, 18, 20, 21, 22, 24, then the group with which it is isomorphic is
determined by the orders of its elements.
Chapter 1. Background 19

Z =
group of integers;
Zn =
cyclic group of order n;
Sn =
symmetric group of degree n;
An =
alternating group of degree n;
Dn dihedral group of order 2n = hr, s | rn = s2 = (rs)2 = 1i;
=
Qn dicyclic group of order 2n = hr, s | rn = 1, rn/2 = s2 = (sr−1 )2 i;
=
V4 =
Klein 4-group which is the direct product of two copies of the cyclic
group of order 2;
G × H = the direct product of G and H;
G o H = the semi-direct product of G with H, where G is normal subgroup.

In the tables, N is the number of elements of order O in the group.

Table 1.3: Groups of order 4

Groups O N O N O N

Z4 1 1 2 1 4 2

Z2 × Z2 1 1 2 3

Table 1.4: Groups of order 6

Groups O N O N O N O N

Z6 1 1 2 1 3 2 6 2

S3 1 1 2 3 3 2
Chapter 1. Background 20

Table 1.5: Groups of order 8

Groups O N O N O N O N

Z8 1 1 2 1 4 2 8 4

Z2 × Z4 1 1 2 3 4 4

(Z2 )3 1 1 2 7

D4 1 1 2 5 4 2

Q4 1 1 2 1 4 6

Table 1.6: Groups of order 9

Groups O N O N O N

Z9 1 1 3 2 9 6

Z3 × Z3 1 1 3 8

Table 1.7: Groups of order 10

Groups O N O N O N O N

Z10 1 1 2 1 5 4 10 4

D5 1 1 2 5 5 4

Table 1.8: Groups of order 12

Groups O N O N O N O N O N O N

Z12 1 1 2 1 3 2 4 2 6 2 12 4

Z6 × Z2 1 1 2 3 3 2 6 6

D6 1 1 2 7 3 2 6 2

Q6 1 1 2 1 3 2 4 6 6 2

A4 1 1 2 3 3 8
Chapter 1. Background 21

Table 1.9: Groups of order 14

Groups O N O N O N O N

Z14 1 1 2 1 7 6 14 6

D7 1 1 2 7 7 6

Table 1.10: Groups of order 16

Groups O N O N O N O N O N

Z16 1 1 2 1 4 2 8 4 16 8

Z8 × Z2 1 1 2 3 4 4 8 8 abelian

Z4 × Z4 1 1 2 3 4 12 abelian

Z4 × (Z2 )2 1 1 2 7 4 8 abelian

(Z2 )4 1 1 2 15

D8 1 1 2 9 4 2 8 4

Q8 1 1 2 1 4 10 8 4

D4 × Z2 1 1 2 11 4 4

Q4 × Z2 1 1 2 3 4 12 non-abelian

Z8 o Z2 , H1 1 1 2 3 4 4 8 8 non-abelian

Z8 o Z2 , H2 1 1 2 5 4 6 8 4

Z4 o Z2 1 1 2 3 4 12 non-abelian

(Z4 × Z2 ) o Z2 , H3 1 1 2 7 4 8 non-abelian

(Z4 × Z2 ) o Z2 , H4 1 1 2 7 4 8 non-abelian
Chapter 1. Background 22

Table 1.11: Groups of order 18

Groups O N O N O N O N O N O N

Z18 1 1 2 1 3 2 6 2 9 6 18 6

Z6 × Z3 1 1 2 1 3 8 6 8

D9 1 1 2 9 3 2 9 6

S3 × Z3 1 1 2 3 3 8 6 6

(Z3 × Z3 ) o Z2 1 2 9 3 8

Table 1.12: Groups of order 20

Groups O N O N O N O N O N O N

Z20 1 1 2 1 4 2 5 4 10 5 20 7

Z10 × Z2 1 1 2 3 5 4 10 12

D10 1 1 2 11 5 4 10 4

Q10 1 1 2 1 4 10 5 4 10 4

Z5 o Z4 1 1 2 5 4 10 5 4

Table 1.13: Groups of order 21

Groups O N O N O N O N

Z21 1 1 3 2 7 6 21 12

Z7 o Z3 1 1 3 14 7 6

Table 1.14: Groups of order 22

Groups O N O N O N O N

Z22 1 1 2 1 11 10 22 10

D11 1 1 2 11 11 10
Chapter 1. Background 23

Table 1.15: Groups of order 24

Groups O N O N O N O N O N O N O N O N

Z24 1 1 2 1 3 2 4 2 6 2 8 4 12 4 24 8

Z12 × Z2 1 1 2 3 3 2 4 4 6 6 12 8

Z6 × (Z2 )2 1 1 2 7 3 2 6 14

S4 1 1 2 9 3 8 4 6

D12 1 1 2 13 3 2 4 2 6 2 12 4

Q12 1 1 2 1 3 2 4 14 6 2 12 4

D6 × Z2 1 1 2 15 3 2 6 6

A4 × Z2 1 1 2 7 3 8 6 8

Q6 × Z3 1 1 2 3 3 2 4 12 6 6

D4 × Z3 1 1 2 5 3 2 4 2 6 10 12 4

Q4 × Z3 1 1 2 1 3 2 4 6 6 2 12 12

S3 × Z4 1 1 2 7 3 2 4 8 6 2 12 4

SL(2, 3) 1 1 2 1 3 8 4 6 6 8

Z3 o Z8 1 1 2 1 3 2 4 2 6 2 8 12 12 4

Z4 o D4 1 1 2 9 3 2 4 6 6 6
Chapter 1. Background 24

Definition 1.36 (Companion matrix). [18] Let f (x) be a monic polynomial in F [x]:

f (x) = xn + an−1 xn−1 + · · · + a1 x + a0 .

The companion matrix C(f ) is the n × n matrix given by

 
0 1 0 ... 0 0
 
 0 0 1 ... 0 0
 

 
C(f ) =  . . . . . . . . . ..
.
..
.
..
. .
 
 
 
 0 0 0 ... 0 1 
 
−a0 −a1 −a2 . . . −an−2 −an−1

In PG(2, q), let

f (x) = x3 + a2 x2 + a1 x + a0 .

The companion matrix C(f ) is the 3 × 3 matrix given by

 
0 1 0
 
C(f ) =  .
 
0 0 1
 
−a0 −a1 −a2

1.11 Methodology to Find mn(2, q)

There are many methods used to find the classification of (k, n)-arcs in PG(2, q). Some
of these methods are now described.

1.11.1 Method 1 [60], [2]

For any (k, n)-arc K, k ≥ n + 2, there are at least four points in K no three of which
are collinear. Let A1 = {Pi | i = 1, 2, . . . , k} and A2 = {Pi0 | i = 1, 2, . . . , k} be two
(k, n)-arcs in PG(2, q), where the coordinates of the points Pi and Pi0 are the following:

Pi = P (xi (1), xi (2), xi (3)) and Pi0 = P (yi (1), yi (2), yi (3)).
Chapter 1. Background 25

By the Fundamental Theorem of Projective Geometry, there exists a unique projectivity

which takes any set of four points of the (k, n)-arc A1 no three collinear to any set of
four points of A2 no three collinear.

Consider the (3 × 3) matrix Z = (zi,j ), i, j = 1, 2, 3. Map a fixed set of four points

of A1 no three collinear, say {P1 , P2 , P3 , P4 }, to any set of four points of A2 no three
collinear, say {P10 , P20 , P30 , P40 }. The (k, n)-arcs A1 and A2 are said to be projectively
equivalent if ZXi = λYj , i, j = n + 2, . . . , k, where Xi and Yj are the column vectors
that represent the other points Pi of A1 and Pj0 of A2 .

To determine Z, fix a quadrangle Q of A1 . Then find Z such that Q maps to a

quadrangle Q0 of A2 . Do this for every Q0 . Here Z ∈ G if it takes the remaining points
of A1 to the remaining points of A2 .

1.11.2 Method 2

This algorithm depends on the type of i-secant distribution and is used to find the
large complete (k, n)-arcs in PG(2, q). To explain this method, the classification of
(k, 4)-arcs in PG(2, 8) is used. The Equations (1.1), (1.2) and (1.3) of Lemma 1.29 are
used here.

1 The construction of the distinct (4, 4)-arcs

Let A = {1, 2, 4, 37} be a (4, 4)-arc in PG(2, 8). A (4, 4)-arc has the same type
of i-secant distribution as A. Therefore there is only one (4, 4)-arc in PG(2, 8) with
respect to the type of i-secant distribution. This can be calculated from the following
equations:

τ0 + τ1 + τ2 + τ3 + τ4 = 73,

τ1 + 2τ2 + 3τ3 + 4τ4 = 36,

τ2 + 3τ3 + 6τ4 = 6.

Since τ4 = 1, τ3 = 0, τ2 = 0, so the only type of (4, 4)-arc is (1, 0, 0, 32, 40).

Chapter 1. Background 26

2 The construction of the distinct (5, 4)-arcs

From Step 1, there is only one (4, 4)-arc A, and there are 64 points of index zero
which do not lie on 4-secant of A. So by adding one point of the points of index zero
to A, then there is only one type of (5, 4)-arc denoted by B, satisfying the following:

τ0 + τ1 + τ2 + τ3 + τ4 = 73,

τ1 + 2τ2 + 3τ3 + 4τ4 = 45,

τ2 + 3τ3 + 6τ4 = 10.

Since τ4 = 1, τ3 = 0, so the only type of (5, 4)-arc is (1, 0, 4, 33, 35).

3 The construction of the distinct (6, 4)-arcs

From Step 2, there is only one (5, 4)-arc B, and there are 63 points of index zero. So
by adding one point of the points of index zero to B, two distinct (6, 4)-arcs C1 and
C2 are obtained. Then C1 is of type (1, 0, 9, 14, 7) and C2 is of type (1, 1, 6, 17, 6).

4 The construction of the distinct (k, 4)-arcs

Table 1.16 shows the number of i-secant distribution of (k, 4)-arcs in PG(2, 8). Here, ξ
is the number of distinct (k, 4)-arcs according to i-secant distribution.

Table 1.16: The i-secant distribution of (k, 4)-arcs in PG(2, 8)

ξ (k, 4) − arcs ξ (k, 4) − arcs ξ (k, 4) − arcs

9 (7, 4) − arcs 128 (15, 4) − arcs 15 (23, 4) − arcs

20 (8, 4) − arcs 130 (16, 4) − arcs 6 (24, 4) − arcs

32 (9, 4) − arcs 127 (17, 4) − arcs 2 (25, 4) − arcs

52 (10, 4) − arcs 119 (18, 4) − arcs 1 (26, 4) − arcs

75 (11, 4) − arcs 94 (19, 4) − arcs 1 (27, 4) − arcs

95 (12, 4) − arcs 71 (20, 4) − arcs 1 (28, 4) − arcs

108 (13, 4) − arcs 45 (21, 4) − arcs

118 (14, 4) − arcs 32 (22, 4) − arcs

Chapter 2

Blocking Sets

A t-fold blocking set B in PG(2, q) or AG(2, q) is a set of points such that each line
contains at least t points of B and some lines contain exactly t points of B. A 1-fold
blocking set is called a blocking set. A blocking set in PG(2, q) should contain no
line. A 2-fold and a 3-fold blocking sets are called a double and a triple blocking set,
respectively.

However, a blocking set can be considered as the complement of a (k, n)-arc K in

PG(2, q) with t = q + 1 − n. The smallest blocking sets are just the lines, and any
blocking set containing a line will be called trivial. A blocking set is said to be minimal,
when no proper subset of it is a blocking set, and some authors call them irreducible
instead of minimal. A minimal blocking set in a projective plane is either a line, or
does not contain a line. The terminology is not standard; sometimes it is supposed
that a blocking set contains no line. For K with an external line l, so K is contained
in the affine plane AG(2, q) = PG(2, q) \ l. The complement of K within AG(2, q) is a
t-blocking set with t = q − n.

Blocking sets have been first studied in 1969 by Di Paola [51], where the author has
calculated the minimum size of a non-trivial blocking set in PG(2, q), having orders
of 4, 5, 7, 8, 9. The major challenge was finding the minimum size of blocking set in
√
PG(2, q). In 1970, Bruen [13, 14] proved that |B| ≥ q + q + 1 for any non-trivial
√ √
blocking set. A Baer subplane of PG(2, q) is a (q + q + 1)-set B of type (1, q + 1).
An alternative approach is to consider a blocking set which does not contain a Baer

27
Chapter 2. Blocking Sets 28

√
subplane. Bruen and Thas showed in 1977 [15] that a blocking set with q+ q+2 points
necessarily contains a Baer subplane. Therefore, for a non-trivial minimal blocking set
√
which does not contain a Baer subplane, B ≥ q + q + 3 will hold. Bruen and Thas in
[15] have introduced another term for these blocking sets: Rédei type. A blocking set
of Rédei type is a blocking set of size q + m that has an m-secant. In Theorem 13.4.1
and Theorem 13.4.2 [36], it is shown that, when q is odd, then there is a blocking set,
the projective triangle, with size 3(q + 1)/2, while for even q there is a blocking set, the
projective triad, with size (3q + 2)/2. Hill and Mason studied multiple blocking sets in
PG(2, q). In 1981 [35], it is shown that, for even q, there are 2-blocking sets of sizes 3q
and 3-blocking sets of sizes 4q. Brouwer and Wilbrink [12] have shown that the size of
the smallest example is between q + q/pe + 1 and q + (q − 1)/(pe − 1) for a divisor e of n.
√
Bruen and Silverman improved the latter bound to q + 2q + 1 − a/(2q) in 1987 [16],
and to q + q 2/3 + 1 for p > 3. In the special case of q = p2 the bound is improved to
√
q + q 3/4 / 2 + 1 by Blokhuis and Storme in 1995 [11]. After that it has been improved
√
in 1996 to q + 2 q + 1 by Ball and Blokhuis [6]. One of the most important questions
raised by Szőnyi in 1977, is the following: “What are the possible sizes of minimal
blocking sets in the interval (q + 1, 3(q + 1)/2)?”

According to Ball in 1994 [3], the arcs (78, 8) and (90, 9) are the largest complete arcs
in PG(2, 11), while for PG(2, 13), there exist no arcs of size (106, 9), (110, 10), (134, 11).
Furthermore, for a triple blocking set in PG(2, q) a new lower bound has been found for
q < 11 by Ball [5] in 1996. For a double blocking set in PG(2, q), Ball and Blokhuis [6]
have introduced a new lower bound for q ≥ 11. In addition, Daskalov [30] investigated
PG(2, 17), and found the largest complete (k, n)-arc for n = 11, . . . , 16 in 2004.

Here, a new upper bound is found for the size of a (k, n)-arc in PG(2, q), for all values
of n > (q + 3)/2 and prime q. It may be noted that a (k, n)-arc is the complement of
a {q 2 + q + 1 − k, q + 1 − n}-blocking set and conversely.
Chapter 2. Blocking Sets 29

2.1 Some Basic Equations

According to Richardson [53], “The name blocking set originates from game theory,
where we have a set of individuals, and certain subsets called coalitions, with the prop-
erty that a coalition can force a particular decision. A blocking set then is a subset that
is not a coalition, but contains at least one member of each coalition, so that it can
block any decision without being able to force one”.

Theorem 2.1. A t-blocking set in AG(2, q) has at least (t + 1)(q − 1) + 1 points.

Theorem 2.2. A t-blocking set in AG(2, q), (t, q) = 1, has at least (t + 1)(q − 1) + t
points.

Theorem 2.3 (Ball [3]). Let K be (k, r)-arc in PG(2, p), where p is prime.

(1) If r ≤ (p + 1)/2, then k ≤ (r − 1)p + 1.

(2) If r ≥ (p + 3)/2, then k ≤ (r − 1)p + r − (p + 1)/2.

Theorem 2.4 (Ball [4]). Let B be t-fold blocking set in PG(2, p), p prime and p > 3.

(1) If t < p/2, then |B| ≥ (t + 21 )(p + 1).

(2) If t > p/2, then |B| ≥ (t + 1)p.

Theorem 2.5 (Ball [4]). Let B be a t-fold blocking set in PG(2, q) that contains a line.

(1) If (t − 1, q) = 1, then |B| ≥ q(t + 1).

(2) If (t − 1, q) > 1 and t ≤ q/2 + 1, then |B| ≥ tq + q − t + 2.

(3) If (t − 1, q) > 1 and t ≥ q/2 + 1, then |B| ≥ t(q + 1).

Definition 2.6. [3] A polynomial in Fq is fully reducible if it factors completely into

linear factors over Fq . If in the sequence of coefficients of a polynomial a long run of
zeros occurs, this polynomial is lacunary.

Theorem 2.7 (Ball [4]). Let f ∈ Fq [x] be fully reducible, and suppose that f has the
form f (x) = xq v(x) + w(x), where v and w have no common factor. Let m < q be the
maximum of the degrees of v and w. Let e be maximal such that f and hence also v
and w are a pe -th power. Then one of the following holds:
Chapter 2. Blocking Sets 30

(1) e = h and m = 0;

(2) e ≥ h/2 and m ≥ pe ;

(3) e < h/2 and m ≥ pe [(ph−e + 1)/(pe + 1)];

(4) e = 0, m = 1 and f (x) = a(xq − x).

Theorem 2.8 gives a slight improvement of Theorem 2.7 as follows.

Theorem 2.8 (Daskalov [30]). Let B be an {l, t}-blocking set in PG(2, p), p prime.

(1) If t < p/2, and p > 3, then l ≥ n(p + 1) + (p + 1)/2.

(2) If l = t(p + 1) + (p + 1)/2, then

(a) through each point of B there are exactly (p + 3)/2 lines that are not t-secants;

(b) through each point of B there are exactly (p − 1)/2 lines that are t-secants;

(c) the total number of t-secants is µ = l(p − 1)/(2t).

Lemma 2.9 ([37], Chapter 12 ). For any set of k points in PG(2, q), the following
hold:

q+1
X
τi = q 2 + q + 1; (2.1)
i=0
q+1
X
iτi = |B|(q + 1); (2.2)
i=1
q+1
X
i(i − 1)τi = |B|(|B| − 1). (2.3)
i=2

Here, τi is the number of i-secants to B.

2.2 Known Lower Bounds for B

According to [38], Table 2.1 gives lower bounds on the number of points in a t-blocking
set B of PG(2, q) without conditions on the t-blocking sets. Also, Table 2.2 gives lower
bounds on the number of points in a t-blocking set B of PG(2, q) with conditions on
the t-blocking sets.
Chapter 2. Blocking Sets 31

Table 2.1: Lower bounds for t-blocking sets without condition

q t |B| ≥ Sharp for References

q = p prime, p > 3 t < p/2 (2t + 1)(p + 1)/2 t = 1, (p − 1)/2 [4]

q = p prime, p > 3 t > p/2 (t + 1)p t = (p + 1)/2 [4]

q<9 2 3q t=2 [9], [25], [42]

q = 11, 13, 17, 19 2 (5q + 7)/2 [4], [6]

√
q square, q > 4 2 2q + 2 q + 2 t=2 [4], [6]
e+1
q = p2e+1 , q > 19 2 2q + pe b p pe +1
+1
c+2 [4], [6]

q = 5, 7, 9 3 4q t=3 [5], [25], [42]

q=8 3 31 t=3 [35],

q = 11, 13, 17 3 (7q + 9)/2 [4], [5]

√
q odd square, q > 21 3 3q + 3 q + 3 t=3 [4], [5]
e+1
q = p2e+1 , q > 17 2 3q + pe b p pe +1
+1
c+3 [4], [5]
√
q even square, q > 4, 3 3q + 2 q + 3 [4], [5]

or q ∈ {25, 49, 81, 121}

Table 2.2: Lower bounds for t-blocking sets with condition

q t Condition |B| ≥ References

√
q t B does not contain a line tq + tq + 1 [4]

q t B contains a line and (t − 1, q) = 1 q(t + 1) [10]

q t B contains a line and (t − 1, q) > 1, t ≤ q/2 + 1 tq + q − t + 2 [17]

q t B contains a line and (t − 1, q) > 1, t ≥ q/2 + 1 t(q + 1) [4]

Chapter 2. Blocking Sets 32

2.3 Non-existence of some (k, n)-Arcs in PG(2, q)

In this section, the non-existence of (k, n)-arcs in PG(2, q) is proved for q=19, 23, 43
and n > 21 (q + 3).

2.3.1 Non-existence of some arcs in PG(2, 19)

Theorem 2.10. (1) There exists no (211, 12)-arc in PG(2, 19); so m12 (2, 19) ≤ 210.

(2) There exists no (231, 13)-arc in PG(2, 19); so m13 (2, 19) ≤ 230.

(3) There exists no (291, 16)-arc in PG(2, 19); so m16 (2, 19) ≤ 290.

Proof. (1) Finding a maximum (k, 12)-arc in PG(2, 19) is equivalent to finding a min-
imum 8-fold blocking set. Theorem 2.4 implies that B must have at least 170 points.
Theorem 2.8 gives that the total number of 8-secants is (170 ∗ 18)/16 which is not an
integer number. Therefore a (211, 12)-arc does not exist and

m12 (2, 19) ≤ 210.

The other bounds are proved in the same way.

Theorem 2.11. (1) There exists no (251, 14)-arc in PG(2, 19); so m14 (2, 19) ≤ 250.

(2) There exists no (271, 15)-arc in PG(2, 19); so m15 (2, 19) ≤ 270.

(3) There exists no (311, 17)-arc in PG(2, 19); so m17 (2, 19) ≤ 310.

(4) There exists no (331, 18)-arc in PG(2, 19); so m18 (2, 19) ≤ 330.

Proof. (1) Finding a maximum (251, 14)-arc is equivalent to finding a {130, 6}-blocking
set B. Theorem 2.8 implies that the total number of 6-secants is 195. Let r be the
length of the longest secant. If r = 20, then B contains a line and Theorem 2.5 implies
that |B| ≥ 133, a contradiction. If 16 < r ≤ 19 then considering lines through a point
on the longest secant but not in B, so B must have at least 6 ∗ 19 + r points. This
contradicts that |B| = 130.
Chapter 2. Blocking Sets 33

Consider the intersection of the 6-secants through P ∈

/ B with the longest secant. So,

τ6 ≥ 9r + (20 − r)(19 − i). (2.4)

The values of τ6 are calculated from (2.4) for i ≤ 8, and give Table 2.3.

Table 2.3: The values of τ6 for i ≤ 8

r 16 15 14 13 12 11 10 9 8

i 0 1 2 3 4 5 6 7 8

τ6 ≥ 220 225 228 229 228 225 220 213 204

This shows that all values of τ6 for r = 8, . . . , 16 give contradictions. This is because
the total number of 6-secants is 195. For r = 6, 7, Lemma 2.9 gives the following:

τ6 + τ7 = 381,

6τ6 + 7τ7 = 2600,

3τ6 + 42τ7 = 16770.

There is no solution for this system. Therefore, no 130 point 6-blocking set exists and
hence no (251, 14)-arc exists.

The remaining cases are proved similarly.

2.3.2 Non-existence of some arcs in PG(2, 23)

Theorem 2.12. (1) There exists no (301, 14)-arc in PG(2, 23); so m14 (2, 23) ≤ 300.

(2) There exists no (325, 15)-arc in PG(2, 23); so m15 (2, 23) ≤ 324.

(3) There exists no (349, 16)-arc in PG(2, 23); so m16 (2, 23) ≤ 348.

(4) There exists no (373, 17)-arc in PG(2, 23); so m17 (2, 23) ≤ 372.

(5) There exists no (421, 19)-arc in PG(2, 23); so m19 (2, 23) ≤ 420.
Chapter 2. Blocking Sets 34

Proof. (1) Finding a maximum (k, 14)-arc in PG(2, 23) is equivalent to finding a min-
imum 10-fold blocking set. Theorem 2.4 implies, since 23 is prime, that such a set
must have at least 252 points. Theorem 2.8 shows that the total number of 10-secants
is (252 ∗ 22)/20 which is not an integer. Therefore there exists no (301, 14)-arc in
PG(2, 23) and m14 (2, 23) ≤ 300. In the same way, the other bounds are shown.

Theorem 2.13. (1) There exists no (397, 18)-arc in PG(2, 23); so m18 (2, 23) ≤ 396.

(2) There exists no (445, 20)-arc in PG(2, 23); so m20 (2, 23) ≤ 444.

(3) There exists no (469, 21)-arc in PG(2, 23); so m21 (2, 23) ≤ 468.

(4) There exists no (493, 22)-arc in PG(2, 23); so m22 (2, 23) ≤ 492.

Proof. (1) Finding a maximum (397, 18)-arc is equivalent to finding a {156, 6}-blocking
set B. Theorem 2.8 implies that the total number of 6-secants is 286. Let r be the
length of the longest secant. If r = 24, then B contains a line and Theorem 2.5 can be
applied. It follows from Theorem 2.5 that |B| ≥ 161, a contradiction. If 18 < r ≤ 23
then consider lines through a point on the longest secant but not in B. Since B must
have at least 6 ∗ 23 + r points, then the values 18 < r ≤ 23 do not give |B|.

Consider the intersection of the 6-secants through P ∈

/ B with the longest secant. So,

τ6 ≥ 11r + (24 − r)(23 − i). (2.5)

The values of τ6 are calculated from (2.5) for i = 0, . . . , 10, and give Table 2.4.

Table 2.4: The values of τ6 for i ≤ 10

r 18 17 16 15 14 13 12 11 10 9 8

i 0 1 2 3 4 5 6 7 8 9 10

τ6 ≥ 336 341 344 345 344 341 336 329 320 309 296

This shows that all values of τ6 for r = 8, . . . , 18 give contradictions. This is because
the total number of 6-secants is 286. For r = 6, 7, Lemma 2.9 gives
Chapter 2. Blocking Sets 35

τ6 + τ7 = 553,

6τ6 + 7τ7 = 3744,

3τ6 + 42τ7 = 24180.

There is no solution for this system. So, no 156 point 6-blocking set exists and hence
no (397, 18)-arc exists.

The proof of the remaining cases is similar.

2.3.3 Non-existence of some arcs in PG(2, 43)

Theorem 2.14. In PG(2, 43) there exists no (k, n)-arc for the following cases.

Case I:
Table 2.5: The values of n that make µ a non-integer

k 991 1035 1079 1123 1167 1211 1299 1343 1431 1475

n 24 25 26 27 28 29 31 32 34 35

mn (2, 43) ≤ 990 1034 1078 1122 1166 1210 1298 1342 1430 1474

k 1519 1651 1695

n 36 39 40

mn (2, 43) ≤ 1518 1650 1694

Case II:
Table 2.6: The values of n that make µ an integer

k 1255 1387 1563 1605 1739 1783

n 30 33 37 38 41 42

mn (2, 43) ≤ 1254 1386 1562 1606 1738 1782

Chapter 2. Blocking Sets 36

Proof. (1) The largest (k, 24)-arc in PG(2, 43) is equivalent to the smallest 20-blocking
set. According to Theorem 2.4, a 20-blocking set should have at least 920 points. From
Theorem 2.8, there are 23 lines not on 20-secants while there are 21 lines which are
20-secants. Then µ is not n integer. Therefore there exist no (991, 24)-arc in PG(2, 43)
and m24 (2, 43) ≤ 990.

Similarly, the remaining cases are proved.

(2) Finding a maximum (1255, 30)-arc is equivalent to finding a {638, 14}-blocking set
B. Theorem 2.8 implies that the total number of 14-secants is 957. Let r represent the
length of the longest secant. If r = 44, then B contains a line and Theorem 2.5 can be
applied. It follows from Theorem 2.5 that |B| ≥ 645, a contradiction. If 37 ≤ r ≤ 43
then consider lines through a point on the longest secant but not in B. So, B must
have at least 14 ∗ 43 + r points, a contradiction.

Consider the intersection of the 14-secants through P ∈

/ B with the longest secant. So,

τ14 ≥ 21r + (44 − r)(43 − i). (2.6)

The values of τ14 are calculated according to (2.6) as shown in Table 2.7.

Table 2.7: The values of τ14 for i ≤ 22

r 36 35 34 33 32 31 30 29 28 27 26 25

i 0 1 2 3 4 5 6 7 8 9 10 11

τ14 ≥ 1100 1113 1124 1133 1140 1145 1148 1149 1148 1145 1140 1133

r 24 23 22 21 20 19 18 17 16 15 14

i 12 13 14 15 16 17 18 19 20 21 22

τ14 ≥ 1124 1113 1100 1085 1068 1049 1028 1005 980 953 924

This shows that all values of τ14 for r = 16, ..., 36 and i = 0, . . . , 20 give a contradiction.
This is because the total number of 14-secants is 957.
Chapter 2. Blocking Sets 37

For r = 14 and r = 15 then from Lemma 2.9 the standard equations for the set B are
the following:

τ14 + τ15 = 1893,

14τ14 + 15τ15 = 28072,

182τ14 + 210τ15 = 406406.

There is no solution of this system. So no 638-point 14-blocking set exists and hence
no (1255, 30)-arc exists.

The remaining cases are proved similarly.

2.4 New Largest Bound

Theorem 2.15. For 21 (q + 3) < n < q, with q prime,

(q + 1)(2n − 3)
mn (2, q) ≤ .
2

Proof. From Theorem 2.3, a (k, n)-arc satisfies

k ≤ (q + 1)(n − 23 ) + 1.

Suppose that there exists a ((q + 1)(n − 23 ) + 1, n)-arc K.

Let B be an {l, t}-blocking set that is the complement of K. Since l = q 2 + q + 1 − k

and t = q + 1 − n, so
l ≥ (q + 1)(q − n + 23 ).

This implies that B is a (q + 1)(q − n + 32 ), q + 1 − n-blocking set, and

|B| = t(q + 1) + 12 (q + 1)

= (q + 1)(q − n + 32 ). (2.7)

Let T be the total number of t-secants of B. From Theorem 2.7, f (x) = xq v(x) + w(x).
Since |B| = (q+1)( 21 +t), then the lacunary polynomial from a point of B is xv(x)∗qw(x)
Chapter 2. Blocking Sets 38

which satisfies f (x) = (vx + w)(v1 w − w1 v). This implies that the number of different
factors in f (x) is precisely (q + 3)/2. So, each point of B is incident with precisely
(q − 1)/2 t-secants. Then count {(x, L)}, where x is in B and L is a t-secant. So
T t = |B|(q − 1)/2; this implies that T = |B|(q − 1)/(2t). Then

T = (q + 1)(t + 12 )(q − 1)/(2t) (2.8)

= (q 2 − 1)(2t + 1)/(4t). (2.9)

If µ = (q 2 − 1)(2t + 1)/(4t) is not an integer, this implies that there exists no

3
((q + 1)(n − ) + 1, n) − arc.
2

So
3
mn (2, q) ≤ (q + 1)(n − ).
2
Suppose that µ is an integer. Let L be an r-secant and P ∈ L\B, where r is the largest
number of points of B on any line through P . If there are s lines that are t-secants to
B, and si through P that are (t + i)-secants, for 1 ≤ i ≤ m, then

|B| = st + s1 (t + 1) + s2 (t + 2) + · · · + sm (t + m) + r

= st + s1 (t + 1) + s2 (t + 1) + s2 + · · · + sm (t + 1) + (m − 1)sm + r

= st + (s1 + s2 + · · · + sm )(t + 1) + s2 + 2s3 + · · · + (m − 1)sm + r

= st + s0 (t + 1) + s00 + r,

where s + s0 = q and s0 = s1 + s2 + · · · + sm , s00 = s2 + 2s3 + · · · + (m − 1)sm . So,

st + (q − s)(t + 1) + r ≤ |B|.

This implies that

s ≥ q(t + 1) + r − |B|.
Chapter 2. Blocking Sets 39

Since |B| = t(q + 1) + (q + 1)/2 = (t + 12 )(q + 1), so

s ≥ r + q(t + 1) − (t + 12 )(q + 1)

≥ (q − 1)/2 + r − t. (2.10)

Now, the number of t-secants is at least s(q + 1 − r) + r(q − 1)/2. So

l(q − 1)/(2t) ≥ r(q − 1)/2 + ( 21 (q − 1) + r − t)(q + 1 − r) (2.11)

≥ (q 2 − 1)/2 + (r − t)(q + 1 − r). (2.12)

So, the Inequality (2.12) becomes

r2 − r(q + 1 + t) + t(q + 1) − (q 2 − 1)/(4t) ≥ 0. (2.13)

To solve (2.13) for r = t, t+1, . . . , q +1, the values of r can be divided into the following
cases.

(1) When r = q+1, then B contains a line, and Theorem 2.5 implies that |B| ≥ q(t+1),
a contradiction.

(2) When 21 (q + 1) + t < r ≤ q, then consider lines through a point on the longest
secant but not on B. So B must have at least qt + r points. This contradicts that
|B| = (t + 12 )(q + 1).

(3) When t + 2 ≤ r ≤ 21 (q + 1) + t, then, since t = q + 1 − n, n > 21 (q + 3), so q > 2t + 1.

Let f (r) = r2 − (q + 1 + t)r + t(q + 1) + (q 2 − 1)/(4t), so

f (r) > r2 − (3t + 2)r + (2t + 1)(t + 1),

> r2 − (3t + 2)r + ( 21 (3t + 2))2 − ( 12 (3t + 2))2 + (2t + 1)(t + 1),

> (r − 21 (3t + 2))2 − t2 /4.

Here, f (r) is positive for some values of t and negative for others.
Chapter 2. Blocking Sets 40

Therefore (2.13) is not true for t + 2 ≤ r ≤ 21 (q + 1) + t.

(4) When t ≤ r ≤ t + 1, according to Lemma 2.9,

τt + τt+1 = q 2 + q + 1; (2.14)

tτt + (t + 1)τt+1 = |B|(q + 1); (2.15)

t(t − 1)τt + t(t + 1)τt+1 = |B|(|B| − 1). (2.16)

Multiplying Equation (2.14) by t and subtracting Equation (2.15) gives

τt+1 = |B|(q + 1) − t(q 2 + q + 1)

1 2
= 2
(q + 2q + 2qt + 1); (2.17)

τt = q 2 + q + 1 + t(q 2 + q + 1) − |B|(q + 1)
1 2
= 2
(q − 2qt + 1). (2.18)

Substituting the values of τt and τt+1 in (2.16) implies that

t(t − 1)τt + t(t + 1)τ + t + 1 = tq + t2 (q 2 + 3q + 1)

6= |B|(|B| − 1).

Therefore, there exists no ((q + 1)(n − 23 ) + 1, n)-arc in PG(2, q) for n > (q + 3)/2.
Hence
mn (2, q) ≤ (q + 1)(n − 32 ) for n > 12 (q + 3).

2.5 Application of Theorem 2.15

Case I : Bounds for complete (k, n)-arcs when µ is a non-integer.

Theorem 2.16. In PG(2, 47), there exist no (k, n)-arc for the following values of k,
giving corresponding upper bounds for mn (2, 47).
Chapter 2. Blocking Sets 41

Table 2.8: The values of n when µ is not an integer

k 1177 1225 1273 1321 1369 1417 1465 1513

n 26 27 28 29 30 31 32 33

mn (2, 47) ≤ 1176 1224 1272 1320 1368 1416 1464 1512

k 1561 1609 1705 1753 1801 1897 1993

n 34 35 37 38 39 41 43

mn (2, 47) ≤ 1560 1608 1704 1752 1800 1896 1992

Proof. For k = 1177 and n = 26, then l=1080, t=22;

|B| = t(q + 1) + 12 (q + 1)

= 22 ∗ 48 + 44

= 1080.

This implies that |B| = l.

Assume that the total number of t-secants is T . Then, from (2.9),

T = |B|(q − 1)/2t

= (1080 ∗ 46)/44.

As µ = (1080 ∗ 46)/44 is not an integer, then there exists no (1177, 26)-arc. So

m26 (2, 47) ≤ 1176.

The remaining cases are proved similarly.

Case II: Bounds for complete (k, n)-arcs when T is integer.

Theorem 2.17. In PG(2, 47) there exists no (k, n)-arc for the following values of k.
Hence the upper bound for mn (2, 47) is established in the corresponding cases.
Chapter 2. Blocking Sets 42

Table 2.9: The values of n when µ is integer

k 1657 1849 1945 2041 2089 2137

n 36 40 42 44 45 46

mn (2, 47) ≤ 1656 1848 1944 2040 2088 2136

Proof. Finding a (1657, 36)-arc is equivalent to finding a {600, 12}-blocking set B. The
total number of 12-secants is 1150. Let r be the length of the longest secant. If r = 48,
then B contains a line and |B| ≥ 611, a contradiction. If 35 ≤ r ≤ 47, considering
lines through a point on the longest secant but not in B, then B must have at least
12 ∗ 47 + r points. This contradicts that |B| = 600.

Now, consider the intersection of the 12-secants through P ∈

/ B with the longest secant.
Then (2.13) becomes
τ12 ≥ (r − 12)(48 − r). (2.19)

The lower bounds for τ12 are calculated according to (2.19) as shown in Table 2.10.

Table 2.10: The values of 14 ≤ r ≤ 36

r 36 35 34 33 32 31 30 29 28 27 26 25

τ12 ≥ 288 299 308 315 320 323 324 323 320 315 308 299

r 24 23 22 21 20 19 18 17 16 15 14

τ12 ≥ 288 275 260 243 224 203 180 155 128 99 68

This shows that all values of r for r = 14, . . . , 36 give a contradiction. This is because
the total number of 12-secants is 46.

However, for r = 12 and r = 13, Equations (2.1), (2.2), (2.3) of Lemma 2.9 become the
following:

τ12 + τ13 = 2257,

12τ12 + 13τ13 = 28800,

132τ12 + 156τ13 = 359400.

Chapter 2. Blocking Sets 43

As there is no solution of this system, so no {600, 12}-blocking set exists and hence no
(1657, 36)-arc exists.

The other bounds are established similarly.

Here, all of the largest bounds which were found by Theorem 2.15 are the same as
known bounds as in [7].

2.6 Chapter Summary

This chapter explains some basic equations for (k, n)-arcs. Also, the non-existence of
(k, n)-arcs in PG(2, q) is proved for q = 19, 23, 43 and n > 21 (q + 3). Finally, a new
upper bound for k is proved and applied to PG(2, 47).
Chapter 3

Classification of (k, n)-arcs in PG(2, q)

3.1 Introduction

Many studies have been done to find all complete (k, 2) and (k, 3)-arcs in PG(2, q) for
some values of q. The size of the largest and the second largest complete (k, 2)-arc are
denoted by m(2, q) and m0 (2, q) as in Notation 1.25. Theorem 1.27 shows the value of
m(2, q), but the value of m0 (2, q) is not known in general.

The full classification of k-arcs in PG(2, q) for q ≤ 19 is shown in [37]. Sticker [26], [27]
obtained the full classification of k-arcs in PG(2, 23), PG(2, 25), PG(2, 27). Coolsaet
[29] obtained the classification of k-arcs in PG(2, 31), in 2014. The values of m0 (2, q) in
PG(2, q) have been shown by Chao and Kaneta [23] to be 21, 22, 24 for q = 25, 27, 29.
The classification of all arcs of size k ≥ q − 8 has been calculated by Kéri [41] for values
of q ≤ 32.

The classification of (k, 3)-arcs in PG(2, q) for q ≤ 9 has been done in 2001 by Marcugini
et al. [45]. Here the largest size of complete arc has been found to be 21 in PG(2, 11)
and 23 in PG(2, 13), while the smallest size is found to be 15 in PG(2, 13). They also
found that there is a complete (k, 3)-arc for each k, where 15 ≤ k ≤ 23, [44], [47].
Bartoli [8] showed that the smallest and largest (k, 3)-arc in PG(2, 16) have sizes 15
and 28. The size of the smallest complete (k, n)-arc in PG(2, q) is denoted by tn (2, q).

Many attempts have been done to find a general lower bound of t2 (2, q) in PG(2, q).
√
In 1959 Segre [56] showed that t2 (2, q) < 2q + 1. Furthermore, Ball [3] found that
44
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 45

√
t2 (2, q) < 3q + 21 . Regarding (k, 3)-arcs, a bound for the smallest size t3 (2, q) of a
complete (k, 3)-arc is
p
q−8+ 24q 3 − 23q 2 − 40q + 16
t3 (2, q) ≤ ,
2(q − 2)

as indicated by Marcugini et al. [47]. Hirschfeld and Pichanick [39] found that

p
tn (2, q) ≤ n(n − 1)(q + 1).

A new lower bound for the smallest complete (k, n)-arc in PG(2, q) has been found in
Theorem 3.3. This general bound can be applied for t2 (2, q) and t3 (2, q).

In this chapter, the classification of (k, n)-arcs in PG(2, q) for some q has been done
using four different methodologies. Also a comparison among these methodologies has
been done to show which method is best according to the time of implementation to
get the final result.

3.2 Known Results for mn(2, q) and tn(2, q) in PG(2, q)

Many studies have been done to find the largest and the smallest complete (k, n)-arcs in
PG(2, q). Some of these studies focused on the full classification of k-arcs in PG(2, q).

3.2.1 The classification of k-arcs in PG(2, q) for q ≤ 29

The full classification of k-arcs in PG(2, q) are given in Table 3.1, Table 3.2 and Table
3.3 for q ≤ 29. For q ≤ 8, see [37]. For q = 9, see [49]. For q = 11, see [54]. For q = 13,
see [2]. For q = 17, 19, see [21]. For q = 23, 25, see [26]. For q = 27, see [27]. For
q = 29, see [22]. For q = 31, see [29]. For q = 32, see [48]. Table 3.4 gives the values
of t2 (2, q) for 2 ≤ q ≤ 29.

Here η is the number of projectively-distinct complete arcs of this size. In each table
the entry gives the number of projectively distinct arcs for the corresponding k and q.
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 46

Table 3.1: The classification of k-arcs in PG(2, q) for 4 ≤ k ≤ 18 and 5 ≤ q ≤ 17

q = 5 q = 7 q = 8 q = 9 q = 11 q = 13 q = 16 q = 17

k=4 1 1 1 1 1 1 1 1

k=5 1 1 1 2 2 3 4 4

k=6 1 3 5 7 15 26 61 74

k=7 1 2 4 21 80 454 733

k=8 1 2 2 21 181 2633 5441

k=9 2 1 5 110 6014 17633

k = 10 1 1 2 27 4899 21064

k = 11 1 2 1171 6814
k = 12 1 2 587 629

k = 13 1 260 15

k = 14 1 100 4

k = 15 30 1

k = 16 9 1

k = 17 3 1

k = 18 2 1

Table 3.2: The classification of k-arcs in PG(2, q) for 4 ≤ k ≤ 7 and 19 ≤ q ≤ 29

q = 19 q = 23 q = 25 q = 27 q = 29

k=4 1 1 1 1 1

k=5 5 6 8 4 10

k=6 117 257 365 174 682

k=7 1768 7613 14114 8261 41301

Chapter 3. Classification of (k, n)-arcs in PG(2, q) 47

Table 3.3: The classification of k-arcs in PG(2, q) for 8 ≤ k ≤ 30 and 19 ≤ q ≤ 29

q = 19 q = 23 q = 25 q = 27 q = 29

k=8 20361 172416 419385 311313 1933469

k=9 115492 2235523 7490938 7348659 58423579

k = 10 280104 15032508 74026338 101047498 1072049736

k = 11 235320 46333282 366007216 744145433 11123944005

k = 12 55708 56846595 806719354 2665334400 60140705285

k = 13 2733 23362684 690593155 4145194407 153994534160

k = 14 83 2634266 195308347 2452359922 167238862321

k = 15 5 64773 15070303 472714330 67799467128

k = 16 4 692 263843 24808360 8854773945

k = 17 1 41 1492 290532 314349510

k = 18 1 22 222 1431 2540088

k = 19 1 6 58 183 7280

k = 20 1 4 29 82 1477

k = 21 1 9 32 646

k = 22 1 5 15 293

k = 23 1 1 4 98

k = 24 1 1 3 43

k = 25 1 1 10

k = 26 1 1 5

k = 27 1 1

k = 28 1 1

k = 29 1

k = 30 1
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 48

Table 3.4: Size and number of the smallest complete k-arc, q ≤ 29

q 2 3 4 5 7 8 9 11 13 16 17 19 23 25 27 29

t2 (2, q) 4 4 6 6 6 6 6 7 8 9 10 10 10 12 12 13

η 1 1 2 1 2 4 2 1 2 8 560 29 1 606 7 708

3.2.2 The classification of (k, 3)-arcs in PG(2, q) for q ≤ 13

The full classification of (k, 3)-arcs in PG(2, q) are given in Table 3.5 and Table 3.6 for
q ≤ 13. For q = 5, see [28]; for q = 7, see [46]; for q = 8, see [43], [60]; for q = 9, see
[45]; for q = 11, see [44]; for q = 13, see [47]. Table 3.7 gives the values of t3 (2, q) for
q ≤ 13.

Table 3.5: The classification of (k, 3)-arcs in PG(2, q) for 4 ≤ k ≤ 15 and 5 ≤ q ≤ 13

q=5 q=7 q=8 q=9 q = 11 q = 13

k=4 1 1 1 1 1 1

k=5 2 3 2 3 3 4

k=6 7 14 15 24 37 62

k=7 13 53 98 188 543 1349

k=8 13 180 505 1341 6743 25670

k=9 16 526 2248 8231 70550 405813

k = 10 7 907 6680 36572 574775 5175900

k = 11 2 923 12664 111833 3520994 52242281

k = 12 395 12781 209172 15291647 403124641

k = 13 65 5822 211818 44020760 2282452774

k = 14 4 871 97050 76936027 9001288812

k = 15 1 43 16386 73157838 23188169036

Chapter 3. Classification of (k, n)-arcs in PG(2, q) 49

Table 3.6: The classification of (k, 3)-arcs in PG(2, q) for 16 ≤ k ≤ 23, 5 ≤ q ≤ 13

q=5 q=7 q=8 q=9 q = 11 q = 13

k = 16 734 32916332 36058738738

k = 17 6 5884405 30742092308

k = 18 333585 12779923892

k = 19 4467 2246238494

k = 20 17 140208097

k = 21 2 2507054

k = 22 9805

k = 23 7

Table 3.7: Number of t3 (2, q), 5 ≤ q ≤ 13

q 5 7 8 9 11 13 16

t3 (2, q) 9 9 11 12 13 15 15

η 2 1 2 4 5 33 1

Theorem 3.1 ([47]). Let K be a complete (k, 3)-arc in PG(2, q). Then
p
q−8+ 24q 3 − 23q 2 − 40q + 16
k≥ .
2(q − 2)

Theorem 3.2 ([39]). Let K be a complete (k, n)-arc in PG(2, q), where n ≥ 2 and
q ≥ n. Then
p
k≥ n(n − 1)(q + 1).

3.3 New Lower Bounds

This section shows some new lower bounds for the smallest complete (k, n)-arcs K in
PG(2, q). A comparison among Theorem 3.1, Theorem 3.2 and Theorem 3.3 is given.
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 50

Theorem 3.3. In PG(2, q), a complete (k, n)-arc does not exist for k ≤ k ∗ , where
p
(q + 1 − n2 ) + (q + 1 − n2 )2 + 4(n2 − n)(q + 1 − n)(q 2 + q + 1)
k∗ = .
2(q + 1 − n)

Proof. From Lemma 1.30, Equations (1.4) and (1.5) are as follows:

ρ1 + ρ2 + · · · + ρn = q + 1,

ρ2 + 2ρ3 + · · · + (n − 1)ρn = k − 1.
 
k−1
(1) With bdc the integer part of d, let m = , n > 2.
n−1
So the maximum value that ρn can have is m.
$ %
(k − 1) − iv=1 (n − v)ρn−v+1
P
(2) Let rn−i = . Here rn−i gives the possible values
(n − i) − 1
for ρn−i .

The points of the plane are of the following types:

 
(k − 1) − (n − 1)ρn
(i) rn−1 = gives the possible values for ρn−1 ; the points are of
(n − 2)
type
(ρ0 , ρ1 , 0, . . . , 0, ρn−1 , j),

where j = 1, 2, . . . , m.
 
(k − 1) − (n − 1)ρn − (n − 2)ρn−1
(ii) rn−2 = gives the possible values for ρn−2 ;
(n − 3)
the points are of type

(ρ0 , ρ1 , 0, . . . , 0, ρn−2 , ρn−1 , j),

where j = 1, 2, . . . , m.

(iii) r2 = (k − 1) − (n − 1)ρn − (n − 2)ρn−1 , . . . , 2ρ3 gives the possible values for ρ2 ;

the points are of type
(ρ0 , ρ1 , ρ2 , . . . , ρn−1 , j),

where j = 1, 2 . . . , m.
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 51

Suppose that α denotes the number of points of PG(2, q) of type (r2 , r3 , . . . , rn−1 ).

From Lemma 1.32, Equations (1.10) and (1.11),

rn−1
m m r2
!
X X X X
j ··· α(r2 , r3 , . . . , rn−1 , j) = nτn , (3.1)
j=1 ρn−1 =0 ρn−2 =0 ρ2 =0
m rn−1 r2
X X X
··· α(r2 , r3 , . . . , rn−1 , j) = k, (3.2)
ρn−1 =0 ρn−2 =0 ρ2 =0

where τn is the total number of n-secants of a (k, n)-arc in PG(2, q).

Since m > 0 for k ≥ n, so

rn−1
m r2
!
X X X
m ··· α(r2 , r3 , . . . , rn − 1, j)
ρn−1 =0 ρn−2 =0 ρ2 =0
rn−1
m m r2
!
X X X X
≥ j ··· α(r2 , r3 , . . . , rn − 1, j) .
j=1 ρn−1 =0 ρn−2 =0 ρ2 =0

This implies that

mk
τn ≤ . (3.3)
n
Since
(k − 1)
m≤ ,
(n − 1)
so
k(k − 1)
τn ≤ . (3.4)
n(n − 1)
On the other hand, if the (k, n)-arc K is complete, Lemma 1.33 implies that

q2 + q + 1 − k
τn ≥ . (3.5)
q+1−n

Now, from Equations (3.4) and (3.5),

k2 − k q2 + q + 1 − k
> .
n2 − n q+1−n
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 52

Hence

(q + 1 − n)k 2 − (q + 1 − n)k > (n2 − n)(q 2 + q + 1) − (n2 − n)k,

(q + 1 − n)k 2 − (q + 1 − n − n2 + n)k − (n2 − n)(q 2 + q + 1) > 0,

(q + 1 − n)k 2 − (q + 1 − n2 )k − (n2 − n)(q 2 + q + 1) > 0. (3.6)

Now, Inequality (3.6) implies that k = k ∗ > 0.

This can be applied to k-arcs and (k, 3)-arcs, as in Table 3.8 and Table 3.9, with the
notation n∗ = bn (2, q) and n = 2, 3.

Table 3.8: Bounds for complete k-arcs for 4 ≤ q ≤ 23

q 4 5 7 8 9 11 13 16 17 19 23

b2 (2, q) 5 5 5 5 6 6 6 7 7 7 8

t2 (2, q) 6 6 6 6 6 7 8 9 10 10 10

m2 (2, q) 6 6 8 10 10 12 14 18 20 20 25

Table 3.9: Bounds for complete (k, 3)-arcs for 4 ≤ q ≤ 16

q 4 5 7 8 9 11 13 16

b3 (2, q) 7 8 9 9 9 10 11 12

t3 (2, q) 7 9 9 11 12 13 15 15

m3 (2, q) 9 11 15 15 17 21 23 28

3.3.1 Comparison with known results

(i) Table 3.10 gives the comparison, for (k, 3)-arcs, among Theorem 3.1, Theorem
3.2 and Theorem 3.3, for 4 ≤ q ≤ 16.
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 53

Table 3.10: Lower bounds for complete (k, 3)-arcs for 4 ≤ q ≤ 16

q Theorem 3.1 Theorem 3.2 Theorem 3.3 Exact value

k k k k

4 7 6 7 7

5 8 6 8 9

7 9 7 9 9

8 9 8 9 11

9 10 8 10 12

11 10 9 10 13

13 11 10 11 15

16 12 11 12 15

(ii) From Theorem 3.3, when n = 3, this implies the same result as Theorem 3.1.

Theorem 3.4. Let K be a (k, n)-arc in PG(2, q). When τn ≤ k, the following state-
ments hold:

(1) k ≤ q + 2 + n(n − 2);

(2) if K is complete, then q + n − 1 ≤ k ≤ n(n − 2) + q + 2.

Proof. From Lemma 1.29, Equations (1.2) and (1.3) are as follows:

n−1
X
τ1 + nτn + iτi = k(q + 1); (3.7)
i=2
n−1
X
n(n − 1)τn + i(i − 1)τi = k(k − 1). (3.8)
i=2

By subtracting Equation (3.7) from (3.8),

n−1
X
n(n − 2)τn = k(k − q − 2) + τ1 − i(i − 2)τi . (3.9)
i=2
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 54

Since k ≥ τn , so

n−1
X
n(n − 2)k ≥ k(k − q − 2) + τ1 − i(i − 2)τi . (3.10)
i=2

Pn−1
If τ1 − i=2 i(i − 2)τi ≥ 0, then

n(n − 2) ≥ k − q − 2. (3.11)

Now, Equation (3.11) implies that

k ≤ q + 2 + n(n − 2). (3.12)

On the other hand, if the (k, n)-arc K is complete, Lemma 1.33 implies that

(q + 1 − n)τn ≥ q 2 + q + 1 − k. (3.13)

Since k ≥ τn , so

(q + 1 − n)k ≥ q 2 + q + 1 − k, (3.14)

(q + 2 − n)k ≥ q 2 + q + 1, (3.15)
q2 + q + 1
k ≥ , (3.16)
q+2−n
n2 − 3n + 3
k ≥ q+n−1+ , (3.17)
q+2−n
k ≥ q + n − 1. (3.18)

Equations (3.12) and (3.18) imply that

q + n − 1 ≤ k ≤ n(n − 2) + q + 2. (3.19)

Proposition 3.5. In PG(2, q), if n | q and q even, then

(1) τn = (q + 1)2 − q(q + 1)/n.

(2) τ0 = q(q + 1 − n)/n.

Chapter 3. Classification of (k, n)-arcs in PG(2, q) 55

Proof. (1) From Theorem 1.28, Equations (1.1) and (1.2) of Lemma 1.29 be

τ0 + τn = q 2 + q + 1 (3.20)

nτn = k(q + 1) (3.21)

Since k = (n − 1)q + n, Equation (3.21) implies that

(q + 1)[(n − 1)q + n]
τn =
n
q(q + 1)
= (q + 1)2 − .
n

(2) From Equation (3.20),

q(q + 1)
τ0 = q 2 + q + 1 − (q + 1)2 +
n
q(q + 1 − n)
= .
n

Table 3.11 shows a special case.

Table 3.11: Application of Proposition (3.5) for some values of n

n 2 q/2 q

|k| q+2 q(q − 1)/2 q2

τ0 q(q − 1)/2 q+2 1

τn (q + 1)(q + 2)/2 q2 − 1 q(q + 1)

Example 3.1. (i) m2 (2, 8) = 10 and the i-secant distribution is (45, 0, 28).

(ii) m2 (2, 16) = 18 and the i-secant distribution is (153, 0, 120).

(iii) m4 (2, 8) = 28 and the i-secant distribution is (63, 0, 0, 0, 10).

(iv) m4 (2, 16) = 52 and the i-secant distribution is (255, 0, 0, 0, 18).

(v) m8 (2, 8) = 64 and the i-secant distribution is (72, 0, 0, 0, 0, 0, 0, 0, 1).

(vi) m8 (2, 16) = 256 and the i-secant distribution is (272, 0, 0, 0, 0, 0, 0, 0, 1).
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 56

3.4 Construction of Complete k-Arcs in PG(2, q) from

a Quadrangle

The equation of a conic C is v(x0 , x1 , x2 ) = 0, where

v(x0 , x1 , x2 ) = a1 x20 + a2 x21 + a3 x22 + a4 x0 x1 + a5 x0 x2 + a6 x1 x2 , (3.22)

ai ∈ Fq . With U0 , U1 , U2 ∈ C, where U0 = P (1, 0, 0), U1 = P (0, 1, 0), U2 = P (0, 0, 1)

then
a4 x0 x1 + a5 x0 x2 + a6 x1 x2 = 0. (3.23)

This is non-singular for a4 a5 a6 6= 0. By dividing Equation (3.23) by a4 then

a5 a6
x0 x1 + x0 x2 + x1 x2 = 0. (3.24)
a4 a4

a5 a6
Let α = and β = , then Equation (3.24) becomes
a4 a4

x0 x1 + αx0 x2 + βx1 x2 = 0. (3.25)

Substitute P (1, 1, 1) in Equation (3.25), then 1 + α + β = 0 implies

β = −(1 + α). (3.26)

Assume α = 0, so β = −1 substitute these values in Equation (3.25), obtains

x0 x1 − x1 x2 = 0,

which implies Equation (3.25) is degenerate, a contradiction. So, α ∈ Fq \ {0, −1}.

Then, for q odd, the form of a conic containing the frame points is

Ci = v(x0 x1 + αx0 x2 − (1 + α)x1 x2 ); (3.27)

where i = 1, . . . , q − 2.

Table 3.12 illustrates a simple application of Equation (3.27).

Chapter 3. Classification of (k, n)-arcs in PG(2, q) 57

Table 3.12: Application of Equation (3.27) in PG(2, 5)

i conic equation
1 x0 x1 + x0 x2 + 3x1 x2
2 x0 x1 + 2x0 x2 + 2x1 x2
3 x0 x1 + 3x0 x2 + 1x1 x2

If there are two conics C 0 = v(x0 x1 +α1 x0 x2 +β1 x1 x2 ), C 00 = v(x0 x1 +α2 x0 x2 +β2 x1 x2 ),
where α1 + β1 + 1 = 0, α2 + β2 + 1 = 0, then the general form of the union of them is

C 0 ∪ C 00 = v(x20 x21 + α1 α2 x20 x22 + β1 β2 x21 x22 + (α1 + α2 )x20 x1 x2 + (β1 + β2 )x0 x21 x2

+(α1 β2 + α2 β1 )x0 x1 x22 ). (3.28)

So, applying Equation (3.28) to the conics in Table 3.12,

C1 ∪ C2 = v(x20 x21 + 2x20 x22 + x21 x22 + 3x20 x1 x2 + 3x0 x1 x22 ).

C1 ∪ C3 = v(x20 x21 + 3x20 x22 + 3x21 x22 + 4x20 x1 x2 + 4x0 x21 x2 ).

C2 ∪ C3 = v(x20 x21 + x20 x22 + 3x21 x22 + 3x0 x21 x2 + 3x0 x1 x22 ).

From the general form of conic and Corollary 1.16, the final form of hyperoval contained
the frame points is the following.

Hi = {P (x0 , x1 , x2 ) | v(x0 x1 + αx0 x2 − (1 + α)x1 x2 )} ∪ {P (−(α + 1), α, 1)}, (3.29)

where i = 1, . . . , q − 2. So, the following proposition gives the number of i-secants for
the union of two hyperovals.

Proposition 3.6. In PG(2, q), q even, if |H ∩ H0 | = 4, then |H ∪ H0 | has the following

constants:

τ4 = (q 2 − 8q + 12)/4,

τ3 = 4(q − 2),

τ2 = (q 2 − 2q + 12)/2,

τ1 = 0,

τ0 = q 2 /4.
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 58

3.5 Algorithms for the Classification of (k, n)-Arcs

in PG(2, q)

Several attempts have been made to find the largest complete (k, n)-arcs in PG(2, q).
This section illustrates the algorithms which are used to find the classification of (k, n)-
arcs in PG(2, q) and make a comparison among them depending on the time to get the
results. Here, one of these methods is developed to reduce the time of implementation.
So, Algorithm One 3.5.1 and Algorithm Three 3.5.3 are first applied to PG(2, 5).

3.5.1 Algorithm One

This algorithm is used to find the maximum size of (k, n)-arc K in PG(2, q) and depends
on the Fundamental Theorem of Projective Geometry. There is a unique projectivity
of PG(2, q) transforming four points no three are collinear to any other four points no
three are collinear. Two (k, n)-arcs K1 , K2 are equivalent if K1 T = K2 for a projectivity
T.

To find a matrix T which transforms the frame points to any given 4-arc say {P (a0 , a1 , a2 ),
P (b0 , b1 , b2 ), P (c0 , c1 , c2 ), P (d0 , d1 , d2 )}, let

(1, 0, 0)T = λ(a0 , a1 , a2 ), (0, 1, 0)T = γ(b0 , b1 , b2 ), (0, 0, 1)T = ν(c0 , c1 , c2 ),

where λ, γ, ν ∈ Fq . So  
λa0 λa1 λa2
 
T =  γb0 γb1 γb2  .
 
 
νc0 νc1 νc2

Also (1, 1, 1)T = ρ(d0 , d1 , d2 ), ρ ∈ Fq , which implies that

 
  λ
a0 b 0 c 0 d 0  
   γ 


d1  .   = (0, 0, 0).
 
 a1 b 1 c 1
    ν 

a2 b 2 c 2 d2  
−ρ
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 59
       
 d0 b0 c0   a0 d0 c0   a0 b 0 d 0   a0 b 0 c 0 
       
       
A =  d1 b1 c1  , B =  a1 d1 c1  , C =  a1 b1 d1  , D =  a1 b1 c1  ,
       
       
       
 d2 b2 c2   a2 d2 c2   a2 b 2 d 2   a2 b 2 c 2 

where A, B, C, D 6= 0. Then,
 
Aa0 Aa1 Aa2
 
T =  Bb0 Bb1 Bb2  .
 
 
Cc0 Cc1 Cc2

The main steps of Algorithm One to classify (k, 4)-arcs in PG(2, 5) are the following.

1. Start with the frame points

P (1, 0, 0), P (0, 1, 0), P (0, 0, 1), P (1, 1, 1).

2. The are 31 − 4 = 27 remaining points. So some of these points are of index zero
which do not lie on any 4-secant and the others are not of index zero and lie on a
4-secant. Add one point of index zero to the frame points; the number of (5, 3)-arcs
obtained is 27.

3. This step finds which arcs are projectively distinct. When there is a projective
transformation T that maps the points of K1 to K2 , K1 and K2 are projectively
equivalent. Similarly, check if K1 is projectively equivalent to K3 , . . . , K27 . Repeat
this procedure for K2 with K3 , . . . , K27 , and similarly for the remain arcs. Therefore
there are 27
P
i=1 i = 378 checks to decide which arcs are equivalent or not equivalent.

4. From step 3, there are 4 projectively distinct (5, 3)-arcs. So, add the remaining
points of index zero to these arcs; the number of (6, 3)-arcs and (6, 4)-arcs obtained is
98. Repeat step 3; then there are 8 projectively distinct (6, 3)-arcs and 2 projectively
distinct (6, 4)-arcs. To classify (7, 4)-arcs start with the two projectively distinct
(6, 4)-arcs.

5. Table 3.13 shows the result of the classification of (k, 4)-arcs in PG(2, 5).
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 60

Here, ξ is the number of (k, 4)-arcs and η is the number of projectively distinct (k, 4)-
arcs.
Table 3.13: The classification of (k, 4)-arcs in PG(2, 5)

k 7 8 9 10 11 12 13 14 15 16

ξ 46 120 406 891 1545 1849 1594 825 250 35

η 6 24 60 123 176 207 165 82 27 6

3.5.2 Algorithm Two

A new algorithm has been obtained by modifying Algorithm One using the stabiliser
group of arcs. Algorithm Two is the following.

1. Start with an initial (k, n)-arc K.

2. Add all points of index zero to obtain (k + 1, n)-arcs. If there is no added point to
a (k, n)-arc K, then K complete.

3. Find the stabiliser group for each (k + 1, n)-arc.

4. Separate (k + 1, n)-arcs into different sets Ai , according to the group of the arcs.

5. Find the projectively distinct arcs for each set Ai and separate them into sets Bij .
Here Bij are the sets of projectively distinct (k + 1, n)-arcs.

6. Collect all projectively distinct arcs in Bij and put them in M . Here M is set of all
projectively distinct (k + 1, n)-arcs.

7. Repeat steps 2, . . . , 5, for all arcs of M .

3.5.3 Algorithm Three

This algorithm depends on the type of i-secant distribution (τn , τn−1 , . . . , τ0 ) and used
to find the largest complete (k, n)-arcs in PG(2, q). Algorithm Three is the following.

1. Start with an initial (k, n)-arc K.

Chapter 3. Classification of (k, n)-arcs in PG(2, q) 61

2. Add all points of index zero to obtain (k + 1, n)-arcs. If there is no added point to
K, then K is complete.

3. Find the i-secant distribution for each (k + 1, n)-arc.

4. Separate (k + 1, n)-arcs into different sets Ai , according to the type of n-secant

distribution.

5. Choose only one (k + 1, n)-arc of each Ai and put it in M . Here M is the set of
distinct (k + 1, n)-arcs according to n-secant distribution.

6. Repeat steps 2, . . . , 4, to all arcs of M .

As an application of Algorithm Three, (k, 4)-arcs in PG(2, 5) will be classified.

1. The construction of the distinct (4, 4)-arcs

Let A = {1, 2, 4, 21} be a (4, 4)-arc in PG(2, 5). Here, all (4, 4)-arcs have the
same type of i-secant distribution. This can be calculated from Equations (1.1),
(1.2) and (1.3) of Lemma 1.29 are as follows:

τ0 + τ1 + τ2 + τ3 + τ4 = 31,

τ1 + 2τ2 + 3τ3 + 4τ4 = 24,

τ2 + 3τ3 + 6τ4 = 6.

Since τ4 = 1, τ3 = 0, τ2 = 0, so the type of (4, 4)-arc is (1, 0, 0, 20, 10).

2. The construction of the distinct (5, 4)-arcs

From step 1, there is only one (4, 4)-arc A, and there are 25 points of index zero
for A. So by adding one point of the points of index zero to A, then there is only
one (5, 4)-arc denoted by B, for which

τ0 + τ1 + τ2 + τ3 + τ4 = 31,

τ1 + 2τ2 + 3τ3 + 4τ4 = 30,

τ2 + 3τ3 + 6τ4 = 10.

Since τ4 = 1, τ3 = 0, so the type of (5, 4)-arc is (1, 0, 4, 18, 8).

Chapter 3. Classification of (k, n)-arcs in PG(2, q) 62

3. The construction of the distinct (6, 4)-arcs

From step 2, there is only one (5, 4)-arc B, and there are 24 points of index zero.
So, by adding these points to B, two distinct (6, 4)-arcs C1 and C2 are obtained.
Then C1 is of type (1, 0, 9, 14, 7) and C2 is of type (1, 1, 6, 17, 6).

4. The construction of the distinct (k, 4)-arcs, k = 7, . . . , 16

Tables 3.14, . . . , 3.23 show the number of i-secant distribution of (k, 4)-arcs in
PG(2, 5). Here, ξ is the number of (k, 4)-arcs.

Table 3.14: The i-secant distribution of (7, 4)-arcs in PG(2, 5)

ξ τ4 τ3 τ2 τ1 τ0 ξ τ4 τ3 τ2 τ1 τ0

3 2 0 9 16 4 24 1 2 9 14 5

6 1 3 6 17 4 13 1 1 12 11 6

Table 3.15: The i-secant distribution of (8, 4)-arcs in PG(2, 5)

ξ τ4 τ3 τ2 τ1 τ0 ξ τ4 τ3 τ2 τ1 τ0

1 1 6 4 18 2 12 2 1 13 11 4

5 2 3 7 17 2 39 1 4 10 12 4

22 2 2 10 14 3 2 2 0 16 8 5

10 1 5 7 15 3 28 1 3 13 9 5

1 1 2 16 6 6
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 63

Table 3.16: The i-secant distribution of (9, 4)-arcs in PG(2, 5)

ξ τ4 τ3 τ2 τ1 τ0 ξ τ4 τ3 τ2 τ1 τ0

2 3 4 6 18 0 4 3 1 15 9 3

13 2 6 6 16 1 15 1 5 15 5 5

6 3 3 9 15 1 114 2 4 12 10 3

1 3 0 18 6 4 44 2 3 15 7 4

6 1 8 6 14 2 58 1 7 9 11 3

22 3 2 12 12 2 58 1 6 12 8 4

63 2 5 9 13 2

Table 3.17: The i-secant distribution of (10, 4)-arcs in PG(2, 5)

ξ τ4 τ3 τ2 τ1 τ0 ξ τ4 τ3 τ2 τ1 τ0

4 3 3 18 3 4 89 3 6 9 12 1

1 5 0 15 10 1 58 3 4 15 6 3

19 4 4 9 14 0 27 2 6 15 4 4

8 3 7 6 15 0 179 3 5 12 9 2

2 1 8 15 2 5 48 1 10 9 8 3

25 4 3 12 11 1 38 1 9 12 5 4

16 4 2 15 8 2 153 2 8 9 10 2

12 2 9 6 13 1 204 2 7 12 7 3

8 1 11 6 11 2
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 64

Table 3.18: The i-secant distribution of (11, 4)-arcs in PG(2, 5)

ξ τ4 τ3 τ2 τ1 τ0 ξ τ4 τ3 τ2 τ1 τ0

5 3 7 16 1 4 4 2 13 4 11 1
12 5 3 16 5 2 67 5 4 13 8 1
4 6 2 13 10 0 18 4 5 16 3 3
2 1 15 4 9 2 74 2 12 7 8 2
7 1 14 7 6 3 135 2 11 10 5 3
13 1 13 10 3 4 268 4 7 10 9 1
19 2 10 13 2 4 227 4 6 13 6 2
46 5 5 10 11 0 392 3 9 10 7 2
32 4 8 7 12 0 132 3 8 13 4 3
88 3 10 7 10 1

Table 3.19: The i-secant distribution of (12, 4)-arcs in PG(2, 5)

ξ τ4 τ3 τ2 τ1 τ0 ξ τ4 τ3 τ2 τ1 τ0
1 1 18 6 2 4 91 4 12 6 8 1

5 3 12 12 0 4 64 3 14 6 6 2
4 2 15 9 1 4 389 5 9 9 7 1
1 3 16 0 12 0 29 6 5 15 3 2
1 6 4 18 0 3 8 5 7 15 1 3
5 3 15 3 9 1 190 6 6 12 6 1
11 2 16 6 4 3 96 3 13 9 3 3
31 5 10 6 10 0 91 4 10 12 2 3
5 7 3 15 5 1 401 4 11 9 5 2
32 7 4 12 8 0 286 5 8 12 4 2
108 6 7 9 9 0
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 65

Table 3.20: The i-secant distribution of (13, 4)-arcs in PG(2, 5)

ξ τ4 τ3 τ2 τ1 τ0 ξ τ4 τ3 τ2 τ1 τ0

4 3 19 3 3 3 125 8 7 9 7 0

3 3 18 6 0 4 192 6 12 6 6 1

23 6 13 3 9 0 5 4 17 3 5 2

5 5 15 3 7 1 136 5 14 6 4 2

49 7 10 6 8 0 90 8 6 12 4 1

7 8 5 15 1 2 19 9 4 12 6 0

4 9 3 15 3 1 344 6 11 9 3 2

12 6 10 12 0 3 121 7 18 12 2 2

28 4 16 6 2 3 360 7 9 9 5 1

67 5 13 9 1 3

Table 3.21: The i-secant distribution of (14, 4)-arcs in PG(2, 5)

ξ τ4 τ3 τ2 τ1 τ0 ξ τ4 τ3 τ2 τ1 τ0

3 6 18 1 4 2 180 8 12 7 2 2

21 6 17 4 1 3 64 9 9 10 1 2

16 7 14 7 0 3 47 10 8 7 6 0

35 9 11 4 7 0 30 11 5 10 5 0

42 7 15 4 3 2 73 10 7 10 3 1

3 11 4 13 2 1 61 8 13 4 5 1

2 10 6 13 0 2 248 9 10 7 4 1
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 66

Table 3.22: The i-secant distribution of (15, 4)-arcs in PG(2, 5)

ξ τ4 τ3 τ2 τ1 τ0 ξ τ4 τ3 τ2 τ1 τ0

6 9 16 3 0 3 11 12 10 3 6 0

30 10 14 3 2 2 10 13 6 9 2 1

9 12 8 9 0 2 23 11 12 3 4 1

59 11 11 6 1 2 15 13 7 6 5 0

2 15 1 12 3 0 82 12 9 6 3 1

3 10 15 0 5 1

Table 3.23: The i-secant distribution of (16, 4)-arcs in PG(2, 5)

ξ τ4 τ3 τ2 τ1 τ0 ξ τ4 τ3 τ2 τ1 τ0

1 12 16 0 0 3 2 15 10 0 6 0

5 15 8 6 0 2 6 15 9 3 3 1

13 14 11 3 1 2 8 16 6 6 2 1

Table 3.24 shows the number of distinct type of i-secant distribution in PG(2, 5) using
Algorithm Three. Here, N is the number of distinct type of i-secant distribution.

Table 3.24: The classification of (k, 4)-arcs in PG(2, 5) according to their i-secant
distribution

k 7 8 9 10 11 12 13 14 15 16

N 4 9 13 17 19 21 19 14 11 6
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 67

3.5.4 Classification of (k, n)-arcs in PG(2, q) using Algorithm

Three

In this section (k, n)-arcs in PG(2, q) are classified.

Case I: Classification of (k, n)-arcs in PG(2, 5), n = 2, 3, 4

(i) From Note 1.15 there exists one type of i-secant distribution of (6, 2)-arcs as
shown in Table 3.25.
Table 3.25: (6, 2)-arcs in PG(2, 5)

Points τ2 τ1 τ0

1 2 3 5 7 20 15 10 1 5

(ii) There are two distinct (11, 3)-arcs according to i-secant distribution as shown in
Table 3.26.
Table 3.26: (11, 3)-arcs in PG(2, 5)

Points τ3 τ2 τ1 τ0

1 2 3 5 6 7 8 10 14 16 20 15 10 1 5

1 2 3 5 6 7 8 10 15 16 20 16 7 4 4

(iii) There are six distinct (16, 4)-arcs according to i-secant distribution as shown in
Table 3.27, Table 3.28.

Table 3.27: (16, 4)-arcs in PG(2, 5)

Points τ4 τ3 τ2 τ1 τ0

1 2 3 4 5 6 7 8 10 15 10 0 6 0

14 15 16 18 20 21 23

1 2 3 5 6 7 8 12 13 12 16 0 0 3

16 18 20 23 25 30 31
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 68

Table 3.28: (16, 4)-arcs in PG(2, 5)

Points τ4 τ3 τ2 τ1 τ0

1 2 3 5 6 7 8 12 13 15 8 6 0 2

14 16 18 20 22 25 30

1 2 3 5 6 7 8 12 13 14 11 3 1 2

14 16 18 20 25 30 31

1 2 3 4 5 6 7 8 9 15 9 3 3 1

10 11 14 15 18 20 25

1 2 3 4 5 6 7 8 10 16 6 6 2 1

11 14 15 18 20 21 29

Note that all results of (k, n)-arcs in PG(2, 5) are calculated within 1 second.

Case II: Classification of (k, n)-arcs in PG(2, 7), n = 2, 3, 4, 5

(i) From Note 1.15 there exists one type of i-secant distribution of (8, 2)-arcs as
shown in Table 3.29.
Table 3.29: (8, 2)-arcs in PG(2, 7)

Points τ2 τ1 τ0

1 2 3 4 7 12 32 33 28 8 21

(ii) There is one type of (15, 3)-arcs as shown in Table 3.30.

Table 3.30: (15, 3)-arcs in PG(2, 7)

Points τ3 τ2 τ1 τ0

1 2 3 4 5 7 8 12 13 30 15 0 12

21 32 33 34 38 52
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 69

(iii) There are two distinct (22, 4)-arcs according to i-secant distribution as shown in
Table 3.31.
Table 3.31: (22, 4)-arcs in PG(2, 7)

Points τ4 τ3 τ2 τ1 τ0

1 2 3 4 5 7 8 10 12 13 15 30 16 3 2 6

16 19 21 27 30 32 33 34 38 43 49

1 2 3 4 5 6 7 8 12 13 16 28 21 0 1 7

20 21 26 28 32 33 34 38 39 46 52

(iv) There are four distinct (29, 5)-arcs according to i-secant distribution as shown in
Tables 3.32.
Table 3.32: (29, 5)-arcs in PG(2, 7)

Points τ5 τ4 τ3 τ2 τ1 τ0

1 2 3 4 5 7 8 10 11 12 28 21 0 0 0 8

13 15 16 19 21 22 27 30 32 33

34 38 40 43 49 50 54 55 56

1 2 3 4 5 7 8 10 11 12 27 20 5 1 0 4

13 15 16 19 21 22 23 25 27 30

32 33 34 37 38 43 45 49 52

1 2 3 4 6 7 8 10 12 13 28 18 6 0 2 3

14 15 16 18 19 20 21 24 26 28

32 33 35 38 40 44 46 48 53

1 2 3 4 5 7 8 10 11 12 31 13 4 6 1 2

13 19 21 22 25 26 27 29 31 32

33 35 37 38 42 43 44 51 57

The results of (k, n)-arcs in PG(2, 7) were obtained within 12 seconds.

Chapter 3. Classification of (k, n)-arcs in PG(2, q) 70

Case III: Classification of (k, n)-arcs in PG(2, 8), n = 2, 3, 4, 5, 6

(i) From Theorem 1.28 2 there exists one type of i-secant distribution of (10, 2)-arcs
as shown in Table 3.33.
Table 3.33: (10, 2)-arcs in PG(2, 8)

Points τ2 τ1 τ0
1 2 3 10 11 40 44 48 53 61 45 0 28

(ii) There are four distinct (15, 3)-arcs according to i-secant distribution as shown in
Tables 3.34.
Table 3.34: (15, 3)-arcs in PG(2, 8)

Points τ3 τ2 τ1 τ0
1 2 3 4 10 11 12 20 38 25 30 0 18
40 44 48 50 53 61

1 2 3 4 10 11 12 20 38 27 24 6 16
40 44 48 53 59 61

1 2 3 4 10 11 12 13 21 29 18 12 14
38 40 44 53 60 61

1 2 3 4 10 11 12 40 48 31 12 18 12
52 53 56 61 69 73

(iii) There is one type of (28, 4)-arcs as shown in Table 3.35.

Table 3.35: (28, 4)-arcs in PG(2, 8)

Points τ4 τ3 τ2 τ1 τ0
1 2 3 4 10 11 12 15 20 22 63 0 0 0 10
28 30 34 35 38 40 44 46 48 50
53 59 61 63 64 65 68 71
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 71

(iv) There are three distinct (33, 5)-arcs according to i-secant distribution as shown
in Tables 3.36.
Table 3.36: (33, 5)-arcs in PG(2, 8)

Points τ5 τ4 τ3 τ2 τ1 τ0

1 2 3 4 8 10 11 12 15 17 20 39 18 10 0 0 6

24 25 26 30 34 36 38 40 41 44 48

50 51 52 53 55 58 61 62 66 70 72

1 2 3 4 8 10 11 12 15 20 22 48 0 16 0 9 0

24 30 34 35 38 40 41 44 48 50 51

52 53 58 59 61 63 64 66 69 71 73

1 2 3 4 5 10 11 12 15 20 22 36 28 0 0 5 4

23 28 30 32 34 35 38 40 42 43 44

46 48 50 53 59 61 63 64 65 68 71

(v) There are two distinct (42, 6)-arcs according to i-secant distribution as shown in
Table 3.37 and Table 3.38.
Table 3.37: (42, 6)-arcs in PG(2, 8)

Points τ6 τ5 τ4 τ3 τ2 τ1 τ0

1 2 3 5 6 7 8 9 10 11 15 50 0 18 0 3 0 2

16 20 22 23 24 26 27 28 29 30 31

32 34 37 38 39 40 41 42 43 47 48

51 52 53 59 60 65 66 72 73
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 72

Table 3.38: (42, 6)-arcs in PG(2, 8)

Points τ6 τ5 τ4 τ3 τ2 τ1 τ0

1 2 3 4 7 8 9 10 12 13 15 39 24 6 0 0 0 4

18 20 21 22 26 28 30 33 34 37 40

41 43 44 45 46 53 54 56 57 58 59

61 62 64 65 67 68 69 72 73

The results of (k, n)-arcs in PG(2, 8) were calculated within 62 seconds.

Case IV: Classification of (k, n)-arcs in PG(2, 9), n = 2, 3, 4, 5

(i) From Note 1.15 there exists one type of i-secant distribution of (10, 2)-arcs as
shown in Table 3.39.
Table 3.39: (10, 2)-arcs in PG(2, 9)

Points τ2 τ1 τ0

1 2 3 5 9 23 43 53 56 63 45 10 36

(ii) There are two distinct (17, 3)-arcs according to i-secant distribution as shown in
Table 3.40.
Table 3.40: (17, 3)-arcs in PG(2, 9)

Points τ3 τ2 τ1 τ0

1 2 3 5 7 9 12 15 23 39 43 38 22 12 19

53 56 58 63 67 79

1 2 3 5 6 7 11 26 29 30 49 39 19 15 18

52 57 61 63 86 90
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 73

(iii) There is one type of (28, 4)-arcs as shown in Table 3.41.

Table 3.41: (28, 4)-arcs in PG(2, 9)

Points τ4 τ3 τ2 τ1 τ0
1 2 3 4 5 6 7 9 11 14 19 63 0 0 28 0
20 23 25 32 43 53 55 56 57 63 64
67 76 82 85 89 90

(iv) There are two distinct (37, 5)-arcs according to i-secant distribution as shown in
Table 3.42.
Table 3.42: (37, 5)-arcs in PG(2, 9)

Points τ5 τ4 τ3 τ2 τ1 τ0
1 2 3 4 6 9 10 16 17 18 20 45 36 0 0 1 9
22 25 27 30 36 43 44 47 51 53 55
57 58 60 61 63 64 69 71 76 79 81
83 87 90 91

1 2 3 4 6 9 10 16 17 18 20 48 29 4 0 2 8
22 25 27 30 36 43 44 47 51 53 55
57 58 60 61 63 64 69 71 79 81 83
85 87 90 91

Case V: Classification of (k, n)-arcs in PG(2, 11), n = 2, 3, 4, 5

(i) From Note 1.15 there exists one type of i-secant distribution of (12, 2)-arcs as
shown in Table 3.43.
Table 3.43: (12, 2)-arcs in PG(2, 11)

Points τ2 τ1 τ0
1 2 3 5 8 17 83 84 89 98 113 128 66 12 55
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 74

(ii) There is one type of (21, 3)-arcs as shown in Tables 3.44.

Table 3.44: (21, 3)-arcs in PG(2, 11)

Points τ3 τ2 τ1 τ0
1 2 3 5 7 12 21 22 23 28 33 63 21 21 28
55 70 79 88 96 102 115 117 121 129

(iii) There is one type of (32, 4)-arcs as shown in Table 3.45.

Table 3.45: (32, 4)-arcs in PG(2, 11)

Points τ4 τ3 τ2 τ1 τ0
1 2 3 12 13 14 15 17 21 30 39 65 30 16 2 20
45 46 49 55 62 65 68 71 75 77 96
105 108 111 113 115 117 118 125 126 132

(iv) There is one type of (43, 5)-arcs as shown in Table 3.46.

Table 3.46: (43, 5)-arcs in PG(2, 11)

Points τ5 τ4 τ3 τ2 τ1 τ0
1 2 3 5 8 12 13 14 15 17 21 60 45 10 3 0 15
22 23 28 30 39 45 46 49 55 62 65
68 71 72 73 75 77 79 88 96 105 106
108 111 113 115 117 118 125 126 129 132
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 75

Case VI: Classification of (k, n)-arcs in PG(2, 13), n = 2, 3, 5

(i) From Note 1.15 there exists one type of i-secant distribution of (14, 2)-arcs as
shown in Table 3.47.
Table 3.47: (14, 2)-arcs in PG(2, 13)

Points τ2 τ1 τ0

1 2 3 8 19 21 28 30 46 48 68 125 147 163 91 14 78

(ii) There are four distinct (23, 3)-arcs according to i-secant distribution as shown in
Tables 3.48.
Table 3.48: (23, 3)-arcs in PG(2, 13)

Points τ3 τ2 τ1 τ0

1 2 3 7 14 37 39 46 51 55 66 69 71 40 29 43

84 98 100 122 125 126 143 163 171 172 180 71 40 29 43

1 2 3 12 23 25 34 35 36 39 41 46 70 43 26 44

51 56 82 85 87 96 118 142 170 171 181

1 2 3 4 7 46 50 51 66 69 72 84 72 37 32 42

94 98 101 103 106 111 114 139 163 177 183

1 2 3 13 30 46 51 66 84 86 92 94 73 34 35 41

101 103 111 126 129 137 143 163 166 172 174
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 76

(iii) There are two distinct (49, 5)-arcs according to i-secant distribution as shown in
Table 3.49.
Table 3.49: (49, 5)-arcs in PG(2, 13)

Points τ5 τ4 τ3 τ2 τ1 τ0

1 2 3 4 5 7 8 9 18 19 22 82 49 16 14 4 18

25 28 29 34 35 37 38 43 46 48 50

54 55 58 63 67 69 70 80 84 95 98

100 101 103 112 113 126 127 148 152 153 156

159 163 166 172 183

1 2 3 4 5 7 8 9 18 19 22 82 47 22 8 6 18

25 28 29 34 35 37 38 43 46 48 50

54 55 58 63 67 69 70 80 84 95 98

101 102 103 112 113 126 127 148 152 153 156

159 163 166 172 183

Case VII: Classification of (k, n)-arcs in PG(2, 16), n = 2, 4

(i) From Theorem 1.28 2 there exists one type of i-secant distribution of (18, 2)-arcs
as shown in Table 3.50.
Table 3.50: (18, 2)-arcs in PG(2, 16)

Points τ2 τ1 τ0

1 2 3 69 79 85 109 112 122 124 158 153 0 120

166 201 207 222 234 250 263

Chapter 3. Classification of (k, n)-arcs in PG(2, q) 77

(ii) There is one type of (52, 4)-arcs as shown in Table 3.51.

Table 3.51: (52, 4)-arcs in PG(2, 16)

Points τ4 τ3 τ2 τ1 τ0
1 2 3 6 7 10 11 16 19 23 24 30 32 221 0 0 0 52
38 39 53 60 64 65 67 70 85 87 90 91 106
119 120 125 132 153 156 165 169 174 186 188 192 206
209 211 212 217 221 232 234 239 244 251 255 262 263

Case VIII: Classification of (k, n)-arcs in PG(2, 17), n = 2, 3

(i) From Note 1.15 there exists one type of i-secant distribution of (18, 2)-arcs as
shown in Table 3.52.
Table 3.52: (18, 2)-arcs in PG(2, 17)

Points τ2 τ1 τ0
1 2 3 100 7 21 29 36 44 73 91 153 18 136
132 137 141 163 173 262 271

(ii) There are three distinct (28, 3)-arcs according to i-secant distribution as shown
in Table 3.53 and Table 3.54.
Table 3.53: (28, 3)-arcs in PG(2, 17)

Points τ3 τ2 τ1 τ0
1 2 3 5 7 15 33 47 52 62 68 80 90 100 78 48 81
81 100 104 184 186 189 197 206 225 232 240 243 254
287 288
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 78

Table 3.54: (28, 3)-arcs in PG(2, 17)

Points τ3 τ2 τ1 τ0
1 2 3 5 15 26 33 34 39 60 83 92 99 101 75 51 80
100 101 105 143 162 164 168 172 179 194 211 255 284
289 296

1 2 3 5 15 21 24 28 41 53 76 100 104 98 84 42 83
105 107 145 149 184 190 192 195 202 231 238 244 260
271 296

Here, all of the results which were found by Algorithm Three are the same as known
results as in [7].

3.5.5 Algorithm Four to find the classification of (k, n)-arcs

This algorithm modifies the Algorithm Two using Algorithm Three. The main steps
of Algorithm Four are the following.

1. Start with an initial (k, n)-arc K.

2. Add all points of index zero to obtain (k + 1, n)-arcs. If there is no added point to
a (k, n)-arc K, then K complete.

3. Find the n-secant distribution for each (k + 1, n)-arc.

4. Separate (k + 1, n)-arcs into different sets Ai , according to the type of n-secant

distribution.

5. Find the stabiliser group of (k + 1, n)-arc for each Ai and separate them into sets
Bij . Here Bij are the sets of distinct stabiliser (k + 1, n)-arcs.

6. Find the projectively distinct arcs for each set Bij and put them in M . Here M is
set of all projectively distinct (k + 1, n)-arcs.

7. Repeat steps 2, . . . , 5, for all arcs of M .

Chapter 3. Classification of (k, n)-arcs in PG(2, q) 79

From the classification of (k, 4)-arcs in PG(2, 5), Algorithm Four gives the following
results. Here δ represents the number of arcs with the same type of i-secant distribution
and ζ represents the number of arcs with the same group.

Table 3.55: Stabilisers of (7, 4)-arcs in PG(2, 5)

δ τ4 τ3 τ2 τ1 τ0 ζ |G| δ τ4 τ3 τ2 τ1 τ0 ζ |G|

3 2 0 9 16 4 3 8 24 1 2 9 14 5 16 1

6 1 3 6 17 4 6 2 8 2

13 1 1 12 11 6 10 2

3 8

Table 3.56: Stabilisers of (8, 4)-arcs in PG(2, 5)

δ τ4 τ3 τ2 τ1 τ0 ζ |G| δ τ4 τ3 τ2 τ1 τ0 ζ |G|

1 1 6 4 18 2 1 8 12 2 1 13 11 4 8 2

5 2 3 7 17 2 5 4 4 4

22 2 2 10 14 3 14 1 39 1 4 10 12 4 24 1

8 2 13 2

10 1 5 7 15 3 7 1 1 4

3 2 2 8

1 1 2 16 6 6 1 8 2 2 0 16 8 5 2 32

28 1 3 13 9 5 20 1

8 2
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 80

Table 3.57: Stabilisers of (9, 4)-arcs in PG(2, 5)

δ τ4 τ3 τ2 τ1 τ0 ζ |G| δ τ4 τ3 τ2 τ1 τ0 ζ |G|

2 3 4 6 18 0 2 24 15 1 5 15 5 5 10 1

6 3 3 9 15 1 6 3 5 2

13 2 6 6 16 1 13 2 114 2 4 12 10 3 91 1

1 3 0 18 6 4 1 24 18 2

22 3 2 12 12 2 22 2 5 8

6 1 8 6 14 2 4 2 44 2 3 15 7 4 12 1

2 4 29 2

63 2 5 9 13 2 32 1 3 4

26 2 58 1 7 9 11 3 49 1

4 5 9 2

4 3 1 15 9 3 4 3 58 1 6 12 8 4 40 1

17 2

1 8
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 81

Table 3.58: Stabilisers of (10, 4)-arcs in PG(2, 5)

δ τ4 τ3 τ2 τ1 τ0 ζ |G| δ τ4 τ3 τ2 τ1 τ0 ζ |G|

4 3 3 18 3 4 4 6 12 2 9 6 13 1 12 1

1 5 0 15 10 1 1 20 8 1 11 6 11 2 6 2

19 4 4 9 14 0 19 2 2 4

8 3 7 6 15 0 6 6 90 3 6 9 12 1 67 1

2 12 16 2

2 1 8 15 2 5 2 4 4 3

25 4 3 12 11 1 15 1 3 6

10 2 58 3 4 15 6 3 34 1

16 4 2 15 8 2 13 1 18 2

3 8 3 3

26 2 6 15 4 4 10 1 3 6

15 2 37 1 9 12 5 4 19 1

1 4 17 2

179 3 5 12 9 2 141 1 1 4

35 2 153 2 8 9 10 2 118 1

3 4 29 2

46 1 10 9 8 3 39 1 3 4

4 2 2 8

3 4 1 16

202 2 7 12 7 3 173 1

29 2
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 82

Table 3.59: Stabilisers of (11, 4)-arcs in PG(2, 5)

δ τ4 τ3 τ2 τ1 τ0 ζ |G| δ τ4 τ3 τ2 τ1 τ0 ζ |G|

5 3 7 16 1 4 5 4 68 5 4 13 8 1 52 1

12 5 3 16 5 2 12 2 16 2

4 6 2 13 10 0 4 8 76 2 12 7 8 2 61 1

2 1 15 4 9 2 2 4 12 2

7 1 14 7 6 3 4 1 3 4

3 2 18 4 5 16 3 3 11 1

13 1 13 10 3 4 13 2 7 2

19 2 10 13 2 4 11 1 135 2 11 10 5 3 101 1

6 2 31 2

2 8 3 4

46 5 5 10 11 0 15 1 265 4 7 10 9 1 225 1

30 2 35 2

1 20 5 4

88 3 10 7 10 1 69 1 224 4 6 13 6 2 178 1

19 2 39 2

31 4 8 7 12 0 14 1 7 4

9 2 393 3 9 10 7 2 320 1

8 4 70 2

4 2 13 4 11 1 4 8 3 4

132 3 8 13 4 3 119 1
13 2
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 83

Table 3.60: Stabilisers of (12, 4)-arcs in PG(2, 5)

δ τ4 τ3 τ2 τ1 τ0 ζ |G| δ τ4 τ3 τ2 τ1 τ0 ζ |G|

4 3 15 3 9 1 4 3 1 6 4 18 0 3 1 24

1 1 18 6 2 4 1 8 12 2 16 6 4 3 9 1

5 3 12 12 0 4 4 6 3 8

1 24 33 5 10 6 10 0 11 1

4 2 15 9 1 4 4 4 19 2

1 3 16 0 12 0 1 96 3 8

33 7 4 12 8 0 15 1 188 6 6 12 6 1 164 1

11 2 12 2

7 4 7 3

94 4 12 6 8 1 77 1 3 4

17 2 2 12

68 3 14 6 6 2 47 1 408 4 11 9 5 2 345 1

17 2 59 2

4 4 4 4

107 6 7 9 9 0 63 1 91 4 10 12 2 3 61 1

27 2 30 2

8 3 98 3 13 9 3 3 68 1

5 4 20 2

4 6 4 3

391 5 9 9 7 1 335 1 6 6

56 2 8 5 7 15 1 3 8 2

29 6 5 15 3 2 23 1 278 5 8 12 4 2 218 1

6 2 49 2

7 4

4 8
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 84

Table 3.61: Stabilisers of (13, 4)-arcs in PG(2, 5)

δ τ4 τ3 τ2 τ1 τ0 ζ |G| δ τ4 τ3 τ2 τ1 τ0 ζ |G|

13 3 18 6 0 4 3 24 32 4 16 6 2 3 21 1

5 3 19 3 3 3 5 6 11 2

23 6 13 3 9 0 5 3 11 6 10 12 0 3 5 3

18 6 6 6

4 4 17 3 5 2 4 4 67 5 13 9 1 3 33 1

6 5 15 3 7 1 6 2 34 2

48 7 10 6 8 0 34 1 88 8 6 12 4 1 74 1

10 2 14 2

4 8 145 5 4 6 4 2 113 1

7 8 5 15 1 2 7 4 32 2

4 9 3 15 3 1 4 3 362 7 9 9 5 1 310 1

122 8 7 9 7 0 76 1 52 2

40 2 120 7 8 12 2 2 93 1

6 4 18 2

195 6 12 6 6 1 170 1 6 4

13 2 3 8

6 3 356 6 11 9 3 2 293 1

4 4 63 2

2 12

21 9 4 12 6 0 9 2

5 3

4 4

3 24
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 85

Table 3.62: Stabilisers of (14, 4)-arcs in PG(2, 5)

δ τ4 τ3 τ2 τ1 τ0 ζ |G| δ τ4 τ3 τ2 τ1 τ0 ζ |G|

35 9 11 4 7 0 14 1 48 7 15 4 3 2 31 1

11 2 17 2

10 4 75 10 7 10 3 1 56 1

21 6 17 4 1 3 21 2 19 2

16 7 14 7 0 3 9 2 177 8 12 7 2 2 145 1

7 4 32 2

6 6 18 1 4 2 6 8 63 9 9 10 1 2 37 1

50 10 8 7 6 0 38 1 26 2

9 2 253 9 10 7 4 1 221 1

3 8 29 2

61 8 13 4 5 1 47 1 3 4

14 2 30 11 5 10 5 0 14 1

2 10 6 13 0 2 2 8 16 2

3 11 4 13 2 1 3 8 5 9 16 3 0 3 5 6
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 86

Table 3.63: Stabilisers of (15, 4)-arcs in PG(2, 5)

δ τ4 τ3 τ2 τ1 τ0 ζ |G| δ τ4 τ3 τ2 τ1 τ0 ζ |G|

3 10 15 0 5 1 3 20 9 13 6 9 2 1 9 1

11 12 10 3 6 0 6 3 25 11 12 3 4 1 16 1

5 12 9 2

30 10 14 3 2 2 17 1 83 12 9 6 3 1 62 1

13 2 13 2

9 12 8 9 0 2 9 2 5 3

58 11 11 6 1 2 44 1 3 6

8 2 16 13 7 6 5 0 11 2

6 4 5 4
6 4 2 15 1 12 3 0 2 24

Table 3.64: Stabilisers of (16, 4)-arcs in PG(2, 5)

δ τ4 τ3 τ2 τ1 τ0 ζ |G| δ τ4 τ3 τ2 τ1 τ0 ζ |G|

1 12 16 0 0 3 1 96 12 14 11 3 1 2 12 2

2 15 10 0 6 0 2 120 6 15 9 3 3 1 6 3

4 15 8 6 0 2 4 4 8 16 6 6 2 1 8 2
Chapter 3. Classification of (k, n)-arcs in PG(2, q) 87

Figure 3.1 shows a comparison among Algorithm One, Algorithm Three and Algorithm
Four to classify (k, 4)-arcs in PG(2, 5).

Algorithm One Algorithm Two Algorithm Four

18000 17111

16000

14000

12000

10000

8000

6000
4140
4000

2000 1170

0
Algorithm One Algorithm Two Algorithm Four

Figure 3.1: Time required for implementation of Algorithm Three and Algorithm
Four

3.6 Chapter Summary

This chapter describes a survey for known results for maximum and minimum size of
complete (k, n)-arcs in PG(2, q) is proved for k-arcs and (k, 3)-arcs. Also, a new lower
bound is proved for smallest size of complete (k, n)-arcs in PG(2, q). A comparison with
known results is shown. Furthermore, a construction of complete k-arcs in PG(2, q)
from quadrangle is proved and applied for PG(2, 5). Four algorithms are explained,
and the classification of (k, n)-arcs is shown for some values of n and q. Algorithm
Two gives the same results for large complete (k, n)-arcs which are obtained using
Algorithm One. Algorithm Two is faster than Algorithm One. Finally, the program
which is written in Fortran as mentioned in Appendix C is faster than the program
which is written in GAP as mentioned in Appendix B.
Chapter 4

High Performance Computing

(HPC)

Many complex mathematical problems deal with extremely large amounts of data that
require to be processed. In such cases, a conventional program, known as a serial
program, may require days, weeks, or even months before obtaining the results. This
requires High Performance Computing (HPC) techniques to accelerate the calculations
without affecting the accuracy of the results. The HPC relies on dividing a big problem
into smaller problems and solving each of them individually, or more precisely in a
parallel way [59]. Thus, parallel computing is the key role of the HPC. Based on
accessing the memory by the processors, the paralleling computing systems can be
divided into shared memory, distributed memory, or shared-distributed hybrid memory
[55]. Besides these techniques, Graphical Processor Units (GPUs) are utilised as an
effective tool to accelerate the calculations.

4.1 Parallel Computing on a Shared Memory Ar-

chitecture - OpenMP

In the shared-memory architecture, processors share the physical memory, that is the
processors read/write from/to any memory location, as depicted in Figure 4.1.

88
Chapter 4. High Performance Computing 89

Figure 4.1: Shared memory architecture

Global Memory

CPU 1 CPU 2 CPU 3 CPU N

The CPUs have been developed during the last decade. They started with more than
one processing unit (core). These cores are designed to work independently forming
a multi-core processor. This increases the number of tasks that can be preformed at
the same time. From the software point of view, OpenMP has been developed to
provide a standard (API) that can be utilised to develop parallel programs using the
shared-memory technique. The OpenMP matches the state of the hardware [20].

In sequential programs, the CPU provides one thread to execute the program instruc-
tions sequentially, whereas in OpenMP the program is executed serially by a master
thread, usually denoted as Thread 0, until it reaches the parallel region, which is a
group of instructions that need to be executed concurrently. In the parallel region, the
master thread creates a group of threads; depending on the nature of the program, a
master thread may or may not wait for the other threads to complete their jobs at
the end of the parallel region. Figure 4.2 shows the sequence of OpenMP threads in a
program.

The program instructions are executed sequentially in the serial programming by one
thread which is dedicated by the CPU. In OpenMP, the program is divided into two
regions: serial region and parallel region, which is a set of instructions that need to be
executed concurrently. The serial region is executed by the master thread while the
parallel region is executed by threads which are generated by the master thread, as
shown in Figure 4.2.

At the end of the parallel region, the master thread may or may nor wait for other
threads to complete executing their tasks.
Chapter 4. High Performance Computing 90

Figure 4.2: Threads diagram in OpenMP program

Serial Region Master thread

Parallel Region Team of threads

Th0 Th1 Th2 ThN

Serial Region Master thread

Th0 = thread 0

Th1 = thread 1

Th2 = thread 2

ThN = thread N

The number of instructions executed in the parallel region equals the number of spec-
ified or available threads. The OpenMP program accelerates the calculations of the
given problem; however, the acceleration factor depends on the size of the parallel re-
gion as compared with the size of the whole program. This has been formalised by
Gene Amdahl [20] as

1
S= , (4.1)
P
(1 − P ) +
SP

where S is the expected acceleration, P is the size of the program, and Sp is the
acceleration obtained from the parallel region.

The OpenMP has feature of adaptation; that is, the program parallelised using OpenMP
can be executed in a parallel or serial version. The adaptation feature allows the de-
veloper to switch between the serial and parallel versions without introducing any
changes. This is achieved by changing the compilation option(s). For instance, the
OpenMP directives in Fortran start with “!$” which are comments for non-OpenMP
compilers.
Chapter 4. High Performance Computing 91

4.2 Parallel Computing on a Distributed Memory

Architecture - MPI

As shown in Figure 4.3, the CPUs in the distributed memory architecture are physically
separated and a communication network is used to maintain the connection among
the CPUs. Each CPU has its own local memory which is non-accessible by other
CPUs. Thus, CPU2 in Figure 4.3 cannot directly fetch data in the local memory
available in CPU1. Message passing is used to perform the communication between
the CPUs. Therefore, the distributed memory architecture is known as the message
passing architecture [33]. Since 1994, the Message Passing Interface (MPI) is used for
message passing.

Figure 4.3: Distributed memory architecture

Network

CPU 1 CPU 2 CPU 3 CPU N

Memory Memory Memory Memory

Each CPU in the MPI communication world has an identifier which is a unique number
used to distinguish between the ranks. The ranks begin with 0 and end with number
of CPUs −1; therefore, the ranks of the MPI communication world with 8 CPUs are
Rank 0, . . . , Rank 7. The problem data are divided by the developer using the ranks.
Let us consider the 2-dimensional matrix shown in Figure 4.4; the matrix is divided
over a number of CPUs and can be performed in many ways. In this case, the division
is randomly chosen to be row by row. In practice, the division of data among ranks
may affect the efficiency of the parallel program. The key feature of the MPI is that
the developer can build an application-dependent topology, which is known as virtual
topology [33]. This is achieved by the aid of ranks and the MPI communication world.
The virtual topology concept overcomes the problem of selecting the efficient topology
that matches the hardware of the HPC.
Chapter 4. High Performance Computing 92

The communications among the CPUs, which are in the same MPI communication
world, are unicast, multicast, or broadcast. The unicast is the point-to-point com-
munication, whereas the multicast is the point to multi-points communications. The
broadcast is the case when a CPU sends data to all other CPUs. Although it is essen-
tial, the communication between CPUs reduces the overall computational performance.
Thus, the broadcast should be avoided as much as possible.

The MPI is a powerful tool for parallel programming. Nevertheless, the MPI library
contains many functions which may be considered a restriction for developers with
limited experience in parallel programming. Practically, this large number of functions
does not add complexity to developing a parallel program using MPI. Many of the
MPI functions are combinations of a small number of concepts [33]. Moreover, to
implement an MPI program with a good acceleration of performance, only a small
number of functions is required. Basically, the following steps are essential for any
MPI program.

• Create the MPI communication world.

• Discover the CPUs which are participating in the MPI communication world.

• Discover the CPU’s rank.

• Exchange the data with the CPU(s).

• Terminate the MPI communication world.

Figure 4.4: Example of MPI data fraction

Rank 0

Rank 1

Rank 2

Rank 3

Rank 4

Rank N-1
Chapter 4. High Performance Computing 93

4.3 Algorithm Five to Find the Classification of

(k, n)-Arcs

This algorithm has been obtained by modifying the Algorithm Four by using parallel
computing on a shared memory architecture-OpenMP, as mentioned in Appendix C.1.2.
Here a laptop is used with the following details.

1. CPU Intel (R) core (TM) i7 − 4710 HQ CPU @ 2.5 GHz.

2. RAM 16 GB.

Figure 4.5 shows a comparison between Algorithm One, Algorithm Three, Algorithm
Four and Algorithm Five.

Algorithm One Algorithm Two Algorithm Four Algorithm Five

18000 17111

16000

14000

12000

10000

8000

6000
4140
4000

2000 1170
292
0
Algorithm One Algorithm Two Algorithm Four Algorithm Five

Figure 4.5: Time required for implementation of Algorithm One, Algorithm Three,
Algorithm Four and Algorithm Five
Chapter 4. High Performance Computing 94

4.3.1 Application of Algorithm Five Using OpenMP

Figure 4.6 shows a comparison between different threads to classify k-arcs in PG(2, 13).

7000
6404.225 1 2 3 4 8 16 32

6000

5000
Time (Seconds)

4000
3374.558

3000

2277.19

2000 1753.527

966.819
1000
650.296

83.596
0
1 2 3 4 8 16 32
Number of threads

Figure 4.6: Time required for implementation of Algorithm Five

Here, the cluster of the University of Sussex is used to find the classification of k-arcs
in PG(2, 13) with different numbers of threads.

4.4 Chapter Summary

This chapter has showed some applications of HPC. A new algorithm with OpenMP
technique is applied. Also, a comparison among all five algorithms is shown.
Chapter 5

Coding Theory

5.1 Introduction

Projective geometry has applications in modern information and communication sci-

ence, more specifically, in coding theory. The aim of coding theory is to develop meth-
ods that enable the recipient of message to detect or even correct errors that occur
while transmitting data. Many aspects of coding theory can be directly translated into
geometry problems.

5.2 Linear Codes

A linear [k, n, d] code C over q = ph is an n-dimensional subspace of the k-dimensional

vector space V = V (k, q) over Fq . The minimum distance d of the code is the smallest
number of positions in which two different elements of C differ. Equivalently, d is the
smallest number of non-zero symbols in any non-zero vector of C. Here, C is MDS if
and only if any n columns of G are linearly independent; this property is preserved
under multiplication of the columns by non-zero scalars. So, consider the columns of
G as points P1 , P2 , . . . , Pk of PG(n − 1, q). It follows that C is MDS if and only if
{P1 , . . . , Pk } is a k−arc of PG(n − 1, q); so a k−arc corresponds to an MDS code. This
gives the relation between linear MDS codes and arcs. One of the main problems of
coding theory is to correct as many errors as possible.

95
Chapter 5. Coding Theory 96

Following [34], let Z+ denote the set of positive integers and F a finite set. This finite
set constitutes our alphabet, and its elements are called letters. Then

V = {x1 · · · xk | xi ∈ F, 1 ≤ i ≤ k; k ∈ Z+ }

is the set of all possible n-tuples of letters (repetition allowed) from F. A subset C of V
is referred to as a code, and the elements of C are codewords. For a fixed k ∈ Z+ , then
refer to a subset C of V as a block code of length k if each of its elements is a k-tuple of
elements of F. Here take F to be a finite field. If C has n information symbols, where
n is the dimension of C as a vector space over F, then refer to C as a (k, n) code.

Let C = {00000, 11111}. If the encoder transmits 00000 but 00101 is received instead,
then the decoder on realising that there are more 0’s than 1’s decodes the message as
00000. Such codes are repetition codes. This code detects up to 4 errors and corrects
2 of them.

In general,
 repetition
 codes can detect the presence of up to d − 1 errors and correct
d−1
up to errors in each d-tuple.
2
The space is V (k, q) = ((Fq )k , +, ×). For x ∈ V (k, q), write

x = (x1 , x2 , . . . , xk ) = x1 x2 · · · xk .

Definition 5.1. (i) A linear code is a subspace of V (k, q).

(ii) If dim C = n, then C is a [k, n]-code or [k, n]q -code,

or, if d(C) = d, it is a [k, n, d]-code or [k, n, d]q -code.

Definition 5.2. A generator matrix G of a [k, n]-code C is an n × k matrix whose

rows form a basis for C. A parity check matrix for C is a (k − n) × k matrix H such
that c ∈ C if and only if cH T = 0. Also, the dual code of C, denoted by C ⊥ , is the set
of vectors of V (k, q) which are orthogonal to every codeword in C:

C ⊥ = {v ∈ V (k, q) | v.u = 0, for all u ∈ C}.

 Chapter 5. Coding Theory 97

5.3 The Main Coding Problem

The notation (k, M, d) is used to represent a code with length k, a total of M code-
words, and minimum distance d. The main coding theory problem is to optimise a
code (k, M, d) and make a balance among k for fast transmission, large M to enable
transmission of a wide variety of messages, and large d to correct many errors.

Let Aq (k, d) be the maximum value of M for which there exists a q-ary (k, M, d)-code.

Theorem 5.3.

(i) Aq (k, 1) = q k .

(ii) Aq (k, k) = q.

Definition 5.4. An e-error-correcting code C in (Fq )k is perfect if any vector in (Fq )k

is at distance at most e from exactly one codeword; that is, every received message is
corrected.

Theorem 5.5. (i) A code C can detect up to s errors in any codeword if

d(C) ≥ s + 1.

(ii) A code C can correct up to t errors in any codeword if

d(C) ≥ 2t + 1.

From [8], the Singleton bound states that, if C is an [k, n, d]q -code, then d ≤ k − n + 1.
The Singleton defect for linear code C is defined as ∆(C) = k + 1 − n − d. Linear codes
meeting the Singleton bound are Maximum Distance Separable (MDS). A linear code
C having Singleton defect equal to 1 and such that also C ⊥ has Singleton defect 1 is
Near Maximum Distance Separable (NMDS). In general, a linear code having Singleton
defect equal to 1 is Almost Maximum Distance Separable (AMDS). Not all the AMDS-
codes are NMDS-codes.
Chapter 5. Coding Theory 98

Theorem 5.6 ([31]). If k > n + q, then every [k, n, k − n]q -code is NMDS.

Theorem 5.7 ([1]). If k > mn−2 (n − 2, q) + 2, then every [k, n, d]q -AMDS code is
NMDS.

Note 5.8. A q-ary [k, n, d]-code is a q-ary (k, q n , d)-code.

Theorem 5.9. There exists a projective [k, n, d]q -code if and only if there exists an
(k, k − d)-arc in PG(n − 1, q).

5.4 MDS Codes of Dimension Three

From Theorem 5.9, if q = 13, n = 3 then in PG(2, 13) there exists a correspondence
between (k, 2)-arcs and projective [k, 3, k − 2]13 -codes.

Table 5.1 shows the MDS codes corresponding to (k, 2)-arcs in PG(2, 13), where e is
the number of errors corrected.
Table 5.1: MDS Codes over F13

K MDS code e K MDS code e K MDS code e

(5, 2)-arc [5, 3, 3] 1 (12, 2)-arc [12, 3, 10] 4 (19, 2)-arc [19, 3, 17] 8

(6, 2)-arc [6, 3, 4] 1 (13, 2)-arc [13, 3, 11] 5 (20, 2)-arc [20, 3, 18] 8

(7, 2)-arc [7, 3, 5] 2 (14, 2)-arc [14, 3, 12] 5 (21, 2)-arc [21, 3, 19] 9

(8, 2)-arc [8, 3, 6] 2 (15, 2)-arc [15, 3, 13] 6 (22, 2)-arc [22, 3, 20] 9

(9, 2)-arc [9, 3, 7] 3 (16, 2)-arc [16, 3, 14] 6 (23, 2)-arc [23, 3, 21] 10

(10, 2)-arc [10, 3, 8] 3 (17, 2)-arc [17, 3, 15] 7

(11, 2)-arc [11, 3, 9] 4 (18, 2)-arc [18, 3, 16] 7

5.5 Codes of Dimension Five

From Theorem 5.9, if q = 13, n = 5, then in PG(2, 13) there exists a correspondence
between (k, 5)-arcs and projective [k, 5, k − 5]13 -codes.
Chapter 5. Coding Theory 99

Table 5.2 shows codes corresponding to (k, 5)-arcs in PG(2, 13) and e is the number of
errors corrected.
Table 5.2: Codes of dimension 5 over F13

K Code e K Code e K Code e

(7, 5)-arc [7, 5, 2] 0 (22, 5)-arc [22, 5, 17] 8 (37, 5)-arc [37, 5, 32] 16

(8, 5)-arc [8, 5, 3] 1 (23, 5)-arc [23, 5, 18] 9 (38, 5)-arc [38, 5, 33] 16

(9, 5)-arc [9, 5, 4] 2 (24, 5)-arc [24, 5, 19] 9 (39, 5)-arc [39, 5, 34] 17

(10, 5)-arc [10, 5, 5] 2 (25, 5)-arc [25, 5, 20] 10 (40, 5)-arc [40, 5, 35] 17

(11, 5)-arc [11, 5, 6] 3 (26, 5)-arc [26, 5, 21] 10 (41, 5)-arc [41, 5, 36] 18

(12, 5)-arc [12, 5, 7] 3 (27, 5)-arc [27, 5, 22] 11 (42, 5)-arc [42, 5, 37] 18

(13, 5)-arc [13, 5, 8] 4 (28, 5)-arc [28, 5, 23] 11 (43, 5)-arc [43, 5, 38] 19

(14, 5)-arc [14, 5, 9] 4 (29, 5)-arc [29, 5, 24] 12 (44, 5)-arc [44, 5, 39] 19

(15, 5)-arc [15, 5, 10] 5 (30, 5)-arc [30, 5, 25] 12 (45, 5)-arc [45, 5, 40] 20

(16, 5)-arc [16, 5, 11] 5 (31, 5)-arc [31, 5, 26] 13 (46, 5)-arc [46, 5, 41] 20

(17, 5)-arc [17, 5, 12] 6 (32, 5)-arc [32, 5, 27] 13 (47, 5)-arc [47, 5, 42] 21

(18, 5)-arc [18, 5, 13] 6 (33, 5)-arc [33, 5, 28] 14 (48, 5)-arc [48, 5, 43] 21

(19, 5)-arc [19, 5, 14] 7 (34, 5)-arc [34, 5, 29] 14 (49, 5)-arc [49, 5, 44] 22

(20, 5)-arc [20, 5, 15] 7 (35, 5)-arc [35, 5, 30] 15

(21, 5)-arc [21, 5, 16] 8 (36, 5)-arc [36, 5, 31] 15

5.6 Chapter Summary

This chapter has shown that arcs have applications to coding theory. Each arc can be
interpreted as a linear code.
Chapter 6

Operational Research

6.1 Introduction

This chapter aims at providing a snapshot of operational research (OR). It starts by

giving a historical overview of the emergence of this discipline of science. A number
of the definitions of operational research are presented with the aim of developing a
better understanding of its concepts. Finally, a brief description of each stage of the
operational research process is provided.

6.2 Development of Operational Research

The concept of operational research emerged in the late 1930s, in the UK, mainly in
military and industrial operations. In fact, OR can be traced back to the period of the
Second World War when it was introduced as a systematic method to deal with the
scarcity of resources and other strategic problems in the military operations, mainly
by the British army. The main motive for its development was the introduction of the
radar defence system for the Royal Air Force, and the belief that radar system would
solve many problems for fighter direction and control. Thus, OR has been used to
incorporate radar data with ground-based viewer data for fighter interruption. Due
to its success during the war, several scientists realised that the principles that they
had used to solve military-related problems can be applied to various other issues in
other fields. Therefore, OR has been introduced to other disciplines such as health
100
Chapter 6. Operational Research 101

care, manufacturing, and transportation. As an example, in 1947, George Dantzig

developed the simplex algorithm for Linear Programming (LP). This helped the growth
of OR, and it is one of the most widely used techniques of OR. Secondly, the computer
revolution made the process of dealing with complex problems much easier, particularly
the development of OR packages. As soon as the simplex method had been developed
and used, the expansion of other methods followed rapidly. The last twenty years
have witnessed the development of most of the OR techniques that are now used,
such as computer simulation, game theory, scheduling algorithms, non-linear dynamic
programming and integer dynamic programming. For more details see [58] and [32].

6.3 The Concept of Operational Research

The concept of operational research has commonly been misinterpreted by many as a

combination of mathematical tools. Although, operational research is fundamentally
based on a variety of mathematical models and techniques, however, in reality, its scope
goes much beyond that operational research, according to the Institute for Operational
Research and the Management Science (INFORMS) as cited by Monios and Bergqvist
[50], 2017, is a “discipline that deals with the application of advanced analytical methods
to help make better decisions”. Whereas Churchman, [24] in 1957, who is seen as one
of the pioneers of operations research, gives a more comprehensive definition, and
describes OR as “the application of scientific methods, techniques and tools to problems
involving the operations of a system so as to provide those in control of the system
with optimum solutions to problems”. Similarly, Rajgopal [52] in 2004 defines it as a “
systematic approach to solving problems, which uses one or more analytical tools in the
process of analysis ”. In light of the aforementioned definitions, operational research
can be viewed as an objective methodology or a complementary analytic approach
that helps managers and decision-makers make the most appropriate decisions. In
other words, it plays a consultative role through providing decision-makers with a set
of robust, scientifically derived alternative solutions for a complex problem at hand.
Nevertheless, the final decision regarding selecting the most appropriate solution is
always left to be made by a human being based on their knowledge, perspective and
experience.
Chapter 6. Operational Research 102

6.4 The Operational Research Approach

Having considered operational research as a tool to facilitate making good decisions, it

is now essential to have a better understanding of such a tool to secure its successful
application to a general problem. Accordingly, this section is dedicated to presenting
general guidelines for the implementation of operational research in practice. Accord-
ing to Rajgopal [52] 2004 the process of implementing operational research can be
decomposed into seven sequential stages as briefly described here in below.

Orientation: This is the first stage in the operational research approach. The main
objective of this stage is to establish the multifunctional team that will be responsible
for addressing the problem at hand. Each multifunctional team has, typically, a leader,
and involves members from different departments that will be influenced by or affect
the problem.

The key aim of the orientation stage is to develop a clear understanding of the problem
and its relation to various operational aspects of the system, and to reach a general
agreement on what should be the most important focus of the project. This can be
achieved through the regular meeting and discussions of the team members along with
reviewing and studying documents and literature related to the problem.

Problem Definition: This stage can be seen as the most difficult stage of the oper-
ational research approach in most cases. It aims to provide a clear definition of the
problem with regard to both the problem’s scope and the results required. This can
be met through defining three main components of the problem, namely:

(i) establishing a very clear objective of the study alongside determining its scope,
so as to set limits for the analysis;

(ii) specifying factors that can influence the defined objective. This should also be
accompanied by providing a set of all the decision alternatives that can be made
by the decision maker and which can affect the objective;

(iii) defining the constraints on the decision alternatives and setting boundaries for
the modelled system operates.
Chapter 6. Operational Research 103

Data Collection: Having defined the problem clearly, the multifunctional team at
this stage should work on collecting the data required to translating the problem into a
model that can be objectively analysed. Typically, data can be gathered from two main
sources, namely, observation and standard. In terms of observation, data is usually
derived from observing the technology of the system in operation. Whereas, in regard
to the second source of data, the company documents and records are used to provide
standard information on the problem at hand. However, it is also common to employ
other data collection techniques such as interview and questionnaires in collecting the
required data.

Model Formulation: This stage aims at translating the defined problem into math-
ematical relationships using the data collected throughout stage three. Based on the
complexity of the resulting model, a standard mathematical model such as linear pro-
gramming can be used to solve a simple model and arrive at a solution. Whereas, for a
very complex model, the team may prefer to simplify the mode and utilize a heuristic
approach, or even use simulations. However, the option to use a combination of differ-
ent mathematical, heuristic and simulation models still exists to resolve the decision
problem in many cases.

Model Solution: This is considered as the simplest stage amongst all operational
research stages. The main objective of this stage is to provide a solution to the con-
structed model of the decision problem by using one or a combination of problem-
solving techniques. During this stage, a particular attention should be paid to the
sensitivity analysis, which concerns with acquiring more information about the con-
duct of the most favourable solution when the model is subjected to some parameter
changes. This is particular critical in case the parameters of the constructed model
cannot be estimated precisely.

Validation and Analysis: This is considered an important stage that must be

promptly undertaken after obtaining the optimum solution and before developing the
final course of action for implementation. It aims to test and check whether or not the
proposed solution model can do what it is expected to do, and whether it is able to
adequately predict the behaviour of the system under study. Indeed, the operational
research team has to be sure that the outputs of the proposed solution model do not
Chapter 6. Operational Research 104

include unexpected or surprising results. In other words, the team should secure that
the proposed solution makes sense, and its results are intuitively acceptable. This can
be achieved by comparing the output of the proposed model with historical output data
under similar implementation conditions. Obtaining similar outputs gives an indication
of the validity of the proposed model. Moreover, operational research team may also
need to use simulation as a technique for verifying the results of the proposed mathe-
matical model. This is particularly important when the proposed model represents a
new system, and where the historical data and output is not available.

Implementation and Monitoring: This is the final stage of operational research

approach. The key objective of this stage is to translate the results obtained from the
solution of a validated model into operating instructions and procedures that are un-
derstandable by the people who will be responsible for implementing and administering
the recommended system. The responsibility of undertaking this task lies mainly with
the operations team. Nevertheless, having completed the implementation process, the
responsibility for monitoring the system is typically shifted to an operating team.

6.5 Chapter Summary

This chapter provides a brief overview of operational research. Also, the historical
development of operational research and its basic concepts have been explained.
Chapter 7

Application of Finite Geometry to

Operational Research

7.1 Introduction

This chapter aims to discuss applications to operational research (OR). The results of
finite geometry can be applied to game theory, probability, and scheduling problems,
which are the some of the important aspects of operational research.

7.2 Application to Probability

An affine plane of order n has n2 points, n2 + n lines and n + 1 parallel classes. Then
some problems of probability, scheduling, and game theory can be solved using affine
planes.

Problem 1: There are nine teams to match up in a schedule of four days Di , where
each team plays once in group of 3.

Solution To design this schedule use AG(2, q). Since in PG(2, 3) there are 13 points,
13 lines, each line has 4 points and 4 lines pass through each point, then Table 7.1
shows the points of PG(2, 3) and Table 7.2 illustrates lines of PG(2, 3).

105
Chapter 7. Application of Finite Geometry to Operational Research 106

Table 7.1: Points of PG(2, 3)

P1 (1, 0, 0) P2 (0, 1, 0) P3 (0, 0, 1) P4 (1, 2, 1) P5 (1, 1, 2)

P6 (1, 0, 2) P7 (1, 0, 1) P8 (1, 1, 1) P9 (1, 1, 0) P10 (0, 1, 1)

P11 (1, 2, 0) P12 (0, 1, 2) P13 (1, 2, 2)

Table 7.2: Lines of PG(2, 3)

L1 P1 P2 P9 P11 L2 P2 P3 P10 P12 L3 P3 P4 P11 P13

L4 P4 P5 P12 P1 L5 P5 P6 P13 P2 L6 P6 P7 P1 P3

L7 P7 P8 P2 P4 L8 P8 P9 P3 P5 L9 P9 P10 P4 P6

L10 P10 P11 P5 P7 L11 P11 P12 P6 P8 L12 P12 P13 P7 P9

L13 P13 P1 P8 P10

Since, AG(2, 3) = PG(2, 3) \ L2 , then by deleting line 2 and its points from each line of
PG(2, 3), so there are nine 3-secants. Table 7.3 and Figure 7.2 illustrate the 3-secants
in AG(2, 3). Figure 7.1 shows the coordinates of the points in AG(2, 3).

Table 7.3: 3-secants in AG(2, 3)

S1 P1 P9 P11 S2 P4 P11 P13 S3 P1 P4 P5

S4 P5 P6 P13 S5 P1 P6 P7 S6 P4 P7 P8

S7 P5 P 8 P9 S8 P4 P6 P9 S9 P5 P7 P11

S10 P6 P8 P11 S11 P7 P9 P13 S12 P1 P8 P13

Chapter 7. Application of Finite Geometry to Operational Research 107

line 1 line 5 line 8 line 11

x y x y x y x y
0 0 1 2 1 1 2 0
1 0 0 2 1 0 0 2
2 0 2 2 1 2 1 1

line 3 line 6 line 9 line 12

x y x y x y x y
2 1 0 2 1 0 2 2
2 0 0 1 2 1 0 1
2 2 0 0 0 2 1 0

line 4 line 7 line 10 line 13

x y x y x y x y
2 1 0 1 2 0 2 2
1 2 1 1 1 2 0 0
0 0 2 1 0 1 1 1

Figure 7.1: Coordinates of the points in AG(2, 3)

P6 P5 P13

line1
line 3
line 4
line 5
line 6

P7 P8 P4 line 7
line 8
line 9
line 10
line 11
line 12
line 13

P1 P9 P11

Figure 7.2: Lines in AG(2, 3)

Chapter 7. Application of Finite Geometry to Operational Research 108

Here, there are four parallel classes in AG(2, 3) as shown in Figures 7.3, 7.4, 7.5, 7.6.

P6 P5 P13

line1
P7 P8 P4
line 5

line 7

P1 P9 P11

Figure 7.3: Class 1 of AG(2, 3)

P6 P5 P13

line 3

P7 P8 P4
line 6

line 8

P1 P9 P11

Figure 7.4: Class 2 of AG(2, 3)

Chapter 7. Application of Finite Geometry to Operational Research 109

P6 P5 P13

line 4
P7 P8 P4
line 11
line 12

P1 P9 P11

Figure 7.5: Class 3 of AG(2, 3)

P6 P5 P13

line 9
P7 P8 P4
line 10
line 13

P1 P9 P11

Figure 7.6: Class 4 of AG(2, 3)

Chapter 7. Application of Finite Geometry to Operational Research 110

So, each class represents a group of 3 players. Therefore,

D1 : S1 ∪ S4 ∪ S6 = (P1 , P9 , P11 ) (P5 , P6 , P13 ) (P4 , P7 , P8 )

D2 : S2 ∪ S5 ∪ S7 = (P4 , P11 , P13 ) (P1 , P6 , P7 ) (P5 , P8 , P9 )

D3 : S3 ∪ S10 ∪ S11 = (P1 , P4 , P5 ) (P6 , P8 , P11 ) (P7 , P9 , P13 )

D4 : S8 ∪ S9 ∪ S12 = (P4 , P6 , P9 ) (P5 , P7 , P11 ) (P1 , P8 , P13 )

7.3 Application of Projective Planes to Scheduling

Problems

7.3.1 A simple Schedule Obtained via a Projective Plane

The projective plane PG(2, q) over the field Fq of q elements can be used as a model for
scheduling problems given a certain number of teams and a certain number of weeks in
the desired schedule. This representation can be used to assign the teams to multi-team
meetings so that each team plays each other team exactly once. The model is designed
as follows:

(a) each point represents a team;

(b) each line represents a meeting;

(c) the number of the meetings (competition) is q + 1.

In Problem 1, there are nine players to match up in a schedule of four days Di . Here,
using K = PG(2, 3) \ L2 . The nine points are P1 , P2 , P3 , P4 , P5 , P6 , P8 , P10 , P11 , which
correspond to nine players as shown in Table 7.4.

Table 7.4

p = Player p1 p2 p3 p4 p5 p6 p7 p8 p9
Point in PG(2,3) P1 P2 P3 P4 P5 P6 P8 P10 P11
Chapter 7. Application of Finite Geometry to Operational Research 111

Also, let S = {Si | i ≤ 12} represent the twelve 3-secants as illustrated in Table 7.5.

Table 7.5: 3-secants in PG(2, 3)

S1 1 2 11 S4 1 4 5 S7 2 4 8 S10 5 10 11
S2 2 3 10 S5 2 5 6 S8 3 5 8 S11 6 8 11
S3 3 4 11 S6 1 3 6 S9 4 6 10 S12 1 8 10

S1 ∩ S4 ∩ S6 ∩ S12 = P1
S1 ∩ S2 ∩ S5 ∩ S7 = P2
S2 ∩ S3 ∩ S6 ∩ S8 = P3
S3 ∩ S4 ∩ S7 ∩ S9 = P4
S4 ∩ S5 ∩ S8 ∩ S10 = P5
S5 ∩ S6 ∩ S9 ∩ S11 = P6
S7 ∩ S8 ∩ S11 ∩ S12 = P8
S2 ∩ S9 ∩ S10 ∩ S12 = P10
S1 ∩ S3 ∩ S10 ∩ S11 = P11

Then, looking for sets of three lines which do not meet in K.

S1 ∩ S8 ∩ S9 = ∅
S2 ∩ S4 ∩ S11 = ∅
S3 ∩ S5 ∩ S12 = ∅
S6 ∩ S7 ∩ S10 = ∅

So, the solution of the problem is the following:

D1 : S1 ∪ S8 ∪ S9 = (P1 , P2 , P11 ) (P3 , P5 , P8 ) (P4 , P6 , P10 )

D2 : S2 ∪ S4 ∪ S11 = (P1 , P4 , P5 ) (P2 , P3 , P10 ) (P6 , P8 , P11 )
D3 : S3 ∪ S5 ∪ S12 = (P1 , P8 , P10 ) (P2 , P5 , P6 ) (P3 , P4 , P11 )
D4 : S6 ∪ S7 ∪ S10 = (P1 , P3 , P6 ) (P2 , P4 , P8 ) (P5 , P10 , P11 )

7.3.2 Constraint Satisfaction Problem (CSP )

Assume that

(1) there are m players,

(2) players can be divided into r groups,

(3) any two players play only one time in the same group,

(4) the number of weeks this can take place is w.

Chapter 7. Application of Finite Geometry to Operational Research 112

To solve this problem, first it is necessary to explain the following constraints:

(a) P = {P1 , . . . , Pm } is the set of all players;

(b) the size of each group of players is n;

(c) the number of groups of n players is t = r × w;

(d) S = {S1 , . . . , St } is the set of all groups of players;

(e) Sij ∩ Sik = ∅, where 1 ≤ i ≤ w and 1 ≤ j < k ≤ r;

(f) | Sij ∩ Sl k |≤ 1, where 1 ≤ i < l ≤ w and 1 ≤ j < k ≤ r.

Problem 2: In a local golf club, there are 28 social golfers, each of whom plays golf
once every 9 days, and always in groups of 4. How can this be arranged?

Solution The target is to come up with a schedule for these golfers to last as many
weeks as possible, such that no golfer plays in the same group as any other golfer on
more than one occasion. In other words, according to Section 7.3.2 this problem can
be described more explicitly by enumerating four constraints which must be satisfied.
Then, m = 28, n = 4, r = 7, w = 9, t = 63. Also,

(a) P = {P1 , . . . , P28 } is the set of all players,

(b) the size of each group of players is 4,

(c) S = {S1 , . . . , S63 } is the set of all groups of players,

(d) Sij ∩ Sik = ∅, where 1 ≤ i ≤ 9 and 1 ≤ j < k ≤ 7,

(e) | Sij ∩ Slk |≤ 1, where 1 ≤ i < l ≤ 9 and 1 ≤ j < k ≤ 7.

Here, the problem is to divide 28 players into groups of 4 players. Therefore from the
classification of (k, 4)-arcs in PG(2, 8), the largest (k, 4)-arc K has k = 28 with i-secant
distribution of type (63, 0, 0, 0, 10) as mention in Table 3.26. Now, the sixty-three 4-
secants represent the groups of 4 players, and the 28 players are represented by the 28
points. Table 7.6 shows the points of K which correspond to 28 players, and Table 7.7
illustrates the sixty-three 4-secants which correspond to the groups of 4 players.
Chapter 7. Application of Finite Geometry to Operational Research 113

Table 7.6

p = Player p1 p2 p3 p4 p5 p6 p7 p8 p9 p10 p11 p12 p13 p14

Point in PG(2,8) P1 P2 P3 P4 P10 P11 P12 P15 P20 P22 P28 P30 P34 P35

p = Player p15 p16 p17 p18 p19 p20 p21 p22 p23 p24 p25 p26 p27 p28

Point in PG(2,8) P38 P40 P44 P46 P48 P50 P53 P59 P61 P63 P64 P65 P68 P71

Table 7.7: 4-secants in PG(2, 8)

S1 p1 p2 p4 p25 S2 p11 p12 p13 p24 S3 p16 p20 p21 p26

S4 p2 p3 p15 p26 S5 p11 p14 p22 p25 S6 p5 p8 p21 p22
S7 p3 p4 p5 p13 S8 p5 p12 p17 p26 S9 p6 p13 p21 p27
S10 p4 p6 p14 p16 S11 p6 p9 p12 p23 S12 p7 p14 p17 p23
S13 p7 p9 p22 p27 S14 p7 p13 p15 p18 S15 p18 p22 p24 p28
S16 p5 p10 p15 p23 S17 p10 p14 p24 p27 S18 p8 p9 p15 p25
S19 p6 p8 p17 p28 S20 p13 p16 p18 p25 S21 p19 p22 p23 p26
S22 p5 p7 p16 p24 S23 p8 p13 p14 p26 S24 p1 p10 p16 p22
S25 p5 p6 p18 p25 S26 p14 p15 p20 p28 S27 p2 p20 p23 p24
S28 p1 p6 p7 p26 S29 p15 p16 p17 p27 S30 p3 p23 p25 p27
S31 p2 p7 p8 p19 S32 p1 p11 p15 p21 S33 p4 p9 p24 p26
S34 p3 p9 p11 p17 S35 p2 p9 p16 p18 S36 p17 p21 p24 p25
S37 p4 p8 p20 p27 S38 p3 p12 p16 p28 S39 p10 p25 p26 p28
S40 p8 p10 p12 p18 S41 p4 p10 p17 p19 S42 p11 p18 p26 p27
S43 p9 p19 p21 p28 S44 p1 p17 p18 p20 S45 p1 p12 p19 p27
S46 p9 p10 p13 p20 S47 p2 p13 p17 p22 S48 p2 p5 p27 p28
S49 p1 p5 p9 p14 S50 p3 p14 p18 p19 S51 p3 p6 p20 p22
S52 p2 p6 p10 p11 S53 p4 p18 p21 p23 S54 p4 p7 p11 p28
S55 p3 p7 p10 p21 S56 p5 p11 p19 p20 S57 p1 p13 p23 p28
S58 p4 p12 p15 p22 S59 p6 p15 p19 p24 S60 p2 p12 p14 p21
S61 p8 p11 p16 p23 S62 p7 p12 p20 p25 S63 p1 p3 p8 p24

Here each group meets thirty-two other groups in four different players, for instance,
S1 meets 8 other groups in each of p1 , p2 , p4 , p25 . Therefore, there remain 32 groups
meeting S1 at points other than those of S1 . This gives Table 7.8.
Chapter 7. Application of Finite Geometry to Operational Research 114

Table 7.8

S1 ∩ S24 ∩ S28 ∩ S32 ∩ S44 ∩ S45 ∩ S49 ∩ S57 ∩ S63 = p1

S1 ∩ S4 ∩ S27 ∩ S31 ∩ S35 ∩ S47 ∩ S48 ∩ S52 ∩ S60 = p2
S4 ∩ S7 ∩ S30 ∩ S34 ∩ S38 ∩ S50 ∩ S51 ∩ S55 ∩ S63 = p3
S1 ∩ S7 ∩ S10 ∩ S33 ∩ S37 ∩ S41 ∩ S53 ∩ S54 ∩ S58 = p4
S6 ∩ S7 ∩ S8 ∩ S16 ∩ S22 ∩ S25 ∩ S48 ∩ S49 ∩ S56 = p5
S9 ∩ S10 ∩ S11 ∩ S19 ∩ S25 ∩ S28 ∩ S51 ∩ S52 ∩ S59 = p6
S12 ∩ S13 ∩ S14 ∩ S22 ∩ S28 ∩ S31 ∩ S54 ∩ S55 ∩ S62 = p7
S6 ∩ S18 ∩ S19 ∩ S23 ∩ S31 ∩ S37 ∩ S40 ∩ S61 ∩ S63 = p8
S11 ∩ S13 ∩ S18 ∩ S33 ∩ S34 ∩ S35 ∩ S43 ∩ S46 ∩ S49 = p9
S16 ∩ S17 ∩ S24 ∩ S39 ∩ S40 ∩ S41 ∩ S46 ∩ S52 ∩ S55 = p10
S2 ∩ S5 ∩ S32 ∩ S34 ∩ S42 ∩ S52 ∩ S54 ∩ S56 ∩ S61 = p11
S2 ∩ S8 ∩ S11 ∩ S38 ∩ S40 ∩ S45 ∩ S58 ∩ S60 ∩ S62 = p12
S2 ∩ S7 ∩ S9 ∩ S14 ∩ S20 ∩ S23 ∩ S46 ∩ S47 ∩ S57 = p13
S5 ∩ S10 ∩ S12 ∩ S17 ∩ S23 ∩ S26 ∩ S49 ∩ S50 ∩ S60 = p14
S4 ∩ S14 ∩ S16 ∩ S18 ∩ S26 ∩ S29 ∩ S32 ∩ S58 ∩ S59 = p15
S3 ∩ S10 ∩ S20 ∩ S22 ∩ S24 ∩ S29 ∩ S35 ∩ S38 ∩ S61 = p16
S8 ∩ S12 ∩ S19 ∩ S29 ∩ S34 ∩ S36 ∩ S41 ∩ S44 ∩ S47 = p17
S14 ∩ S15 ∩ S25 ∩ S35 ∩ S40 ∩ S42 ∩ S44 ∩ S50 ∩ S53 = p18
S20 ∩ S21 ∩ S31 ∩ S41 ∩ S43 ∩ S45 ∩ S50 ∩ S56 ∩ S59 = p19
S3 ∩ S26 ∩ S27 ∩ S37 ∩ S44 ∩ S46 ∩ S51 ∩ S56 ∩ S62 = p20
S3 ∩ S6 ∩ S9 ∩ S32 ∩ S36 ∩ S43 ∩ S53 ∩ S55 ∩ S60 = p21
S5 ∩ S6 ∩ S13 ∩ S15 ∩ S21 ∩ S24 ∩ S47 ∩ S51 ∩ S58 = p22
S11 ∩ S12 ∩ S16 ∩ S21 ∩ S27 ∩ S30 ∩ S53 ∩ S58 ∩ S61 = p23
S2 ∩ S15 ∩ S17 ∩ S22 ∩ S27 ∩ S33 ∩ S36 ∩ S59 ∩ S63 = p24
S1 ∩ S5 ∩ S18 ∩ S20 ∩ S25 ∩ S30 ∩ S36 ∩ S39 ∩ S62 = p25
S3 ∩ S4 ∩ S8 ∩ S21 ∩ S23 ∩ S28 ∩ S33 ∩ S39 ∩ S42 = p26
S9 ∩ S13 ∩ S17 ∩ S29 ∩ S30 ∩ S37 ∩ S42 ∩ S45 ∩ S48 = p27
S15 ∩ S19 ∩ S26 ∩ S38 ∩ S39 ∩ S43 ∩ S48 ∩ S54 ∩ S57 = p28

Also,

S1 ∩ S8 ∩ S14 ∩ S17 ∩ S43 ∩ S51 ∩ S61 =∅

S4 ∩ S10 ∩ S13 ∩ S30 ∩ S36 ∩ S40 ∩ S56 =∅
S3 ∩ S7 ∩ S12 ∩ S15 ∩ S18 ∩ S45 ∩ S52 =∅
S5 ∩ S9 ∩ S16 ∩ S31 ∩ S33 ∩ S38 ∩ S44 =∅
S19 ∩ S20 ∩ S27 ∩ S42 ∩ S49 ∩ S55 ∩ S58 =∅
S11 ∩ S22 ∩ S32 ∩ S37 ∩ S39 ∩ S47 ∩ S50 =∅
S21 ∩ S25 ∩ S29 ∩ S46 ∩ S54 ∩ S60 ∩ S63 =∅
S2 ∩ S6 ∩ S26 ∩ S28 ∩ S35 ∩ S41 ∩ S57 =∅
S23 ∩ S24 ∩ S34 ∩ S48 ∩ S53 ∩ S59 ∩ S62 =∅
Chapter 7. Application of Finite Geometry to Operational Research 115

Therefore, all the sixty-three groups of 4 players can be divided into nine days Di ,
where there are seven 4 players in each day.

D1 : S1 ∪ S8 ∪ S14 ∪ S17 ∪ S43 ∪ S51 ∪ S61

D2 : S4 ∪ S10 ∪ S13 ∪ S30 ∪ S36 ∪ S40 ∪ S56
D3 : S3 ∪ S7 ∪ S12 ∪ S15 ∪ S18 ∪ S45 ∪ S52
D4 : S5 ∪ S9 ∪ S16 ∪ S31 ∪ S33 ∪ S38 ∪ S44
D5 : S19 ∪ S20 ∪ S27 ∪ S42 ∪ S49 ∪ S55 ∪ S58
D6 : S11 ∪ S22 ∪ S32 ∪ S37 ∪ S39 ∪ S47 ∪ S50
D7 : S21 ∪ S25 ∪ S29 ∪ S46 ∪ S54 ∪ S60 ∪ S63
D8 : S2 ∪ S6 ∪ S26 ∪ S28 ∪ S35 ∪ S41 ∪ S57
D9 : S23 ∪ S24 ∪ S34 ∪ S48 ∪ S53 ∪ S59 ∪ S62

Problem 3: Sixteen players need to be arranged in groups of four and three players
to play for six days, where each player plays in the same number of groups.

Solution The target of this problem is to divide 16 players into groups of 3 and 4
players. From the classification of (16, 4)-arcs K in PG(2, 5), with i-secant distribution
of type (12, 16, 0, 0, 3) as mention in Table 3.27. Then, the twelve 4-secants represent
the groups of 4 players and the sixteen 3-secants represent the group of 3 players and
the 16 players are represented by the 16 points. Table 7.9 shows the points of K which
correspond to 16 players, Table 7.10 shows the groups of 4 players and Table 7.11 shows
the groups of 3 players.

Table 7.9

p = Player p1 p2 p3 p4 p5 p6 p7 p8
Point in PG(2,5) P1 P2 P3 P5 P6 P7 P8 P12

p = Player p9 p10 p11 p12 p13 p14 p15 p16

Point in PG(2,5) P13 P16 P18 P20 P23 P25 P30 P31

Table 7.10

(p1 , p2 , p25 , p30 ) (p2 , p3 , p16 , p13 ) (p3 , p5 , p6 , p25 ) (p6 , p7 , p20 , p30 )
(p5 , p7 , p8 , p31 ) (p1 , p5 , p12 , p13 ) (p1 , p7 , p16 , p18 ) (p2 , p8 , p12 , p20 )
(p3 , p13 , p18 , p20 ) (p6 , p12 , p16 , p23 ) (p8 , p18 , p23 , p25 ) (p13 , p23 , p30 , p31 )
Chapter 7. Application of Finite Geometry to Operational Research 116

Table 7.11

(p1 , p3 , p23 ) (p2 , p5 , p18 ) (p1 , p6 , p8 ) (p2 , p7 , p23 )

(p3 , p8 , p30 ) (p12 , p25 , p31 ) (p2 , p6 , p13 ) (p3 , p7 , p12 )

(p8 , p13 , p16 ) (p5 , p16 , p30 ) (p6 , p18 , p31 ) (p5 , p20 , p23 )

(p7 , p13 , p25 ) (p16 , p20 , p25 ) (p12 , p18 , p30 ) (p1 , p20 , p31 )

So, the solution of this problem is the following:

D1 : (p3 , p5 , p6 , p25 ) (p2 , p8 , p12 , p20 ) (p1 , p7 , p16 , p18 ) (p13 , p23 , p30 , p31 )
D2 : (p1 , p2 , p25 , p30 ) (p5 , p7 , p8 , p31 ) (p3 , p13 , p18 , p20 ) (p6 , p12 , p16 , p23 )
D3 : (p6 , p7 , p20 , p30 ) (p1 , p3 , p23 ) (p2 , p5 , p18 )(p12 , p25 , p31 ) (p8 , p13 , p16 )
D4 : (p1 , p5 , p12 , p13 ) (p2 , p7 , p23 ) (p3 , p8 , p30 )(p6 , p18 , p31 ) (p16 , p20 , p25 )
D5 : (p2 , p3 , p16 , p31 ) (p1 , p6 , p8 ) (p5 , p20 , p23 )(p7 , p13 , p25 ) (p12 , p18 , p30 )
D6 : (p8 , p18 , p23 , p25 ) (p1 , p20 , p31 ) (p2 , p6 , p13 )(p3 , p7 , p12 ) (p5 , p16 , p30 )

Here, each players participated in 3 groups of 4 players and 3 groups of 3 players.

7.4 Generalisation of Kirkman’s Problem

Definition 7.1 (Kirkman’ Problem). A Kirkman triple system is a method of choosing

3-sets called blocks from a set of v objects, and of partitioning the set of blocks into
subsets called rounds, so that each object occurs exactly once per round and each
object-pair occurs in exactly one triple in the system.

A generalisation of Kirkman’s problem as follow. If m = q(n − 1) + n schoolgirls go

walking each day in q + 1 − q/n groups of n, they can walk for q + 1 days so that each
girl has walked in the same group as has every other girl and with no girl twice.

Problem 4: A school has 64 girls, who need to be arrange in 8 groups of 8 girls. How
can they be arranged walk for nine days, such that no two girls walk together twice?
Chapter 7. Application of Finite Geometry to Operational Research 117

Solution Here, m = 64, r = 8, w = 8, t = 64. According to Section 7.3.2 this prob-

lem can be described more explicitly by enumerating four constraints which must be
satisfied. Then,

(a) P = {P1 , . . . , P64 } is the set of all girls,

(b) the size of each group of girls is 8,

(c) S = {S1 , . . . , S64 } is the set of all groups of girls,

(d) Sij ∩ Sik = ∅, where 1 ≤ i ≤ 9 and 1 ≤ j < k ≤ 8,

(e) | Sij ∩ Slk |≤ 1, where 1 ≤ i < l ≤ 9 and 1 ≤ j < k ≤ 8.

The classification in Table 7.12 and Table 7.13 shows the solution of this problem.

Table 7.12: 8-secants in PG(2, 8)

1 2 4 8 16 32 37 55 63 66 70 5 21 26 44 53
D1 61 62 68 3 19 24 42 51 57 58 60 72 15 20 38 47
49 50 52 56 7 12 30 39 33 34 36 40 48 69 14 23
28 29 31 35 43 59 9 18 10 11 13 17 25 41 46 73
2 3 5 9 17 33 38 56 66 68 72 7 23 28 46 55
D2 62 63 69 4 20 25 43 52 58 59 61 73 16 21 39 39
50 51 53 57 8 13 31 40 34 35 37 41 49 70 15 24
29 30 32 36 44 60 10 19 11 12 14 18 26 42 47 1
3 4 10 18 34 39 57 66 72 73 2 14 30 35 53 62
D3 48 49 51 55 63 11 29 38 43 44 46 50 58 1 24 33
25 26 28 32 40 56 61 15 16 17 19 23 31 47 52 70
7 9 13 21 37 42 60 69 5 8 12 20 36 41 59 68
4 5 7 11 19 35 40 58 68 70 1 9 25 30 48 57
D4 66 69 73 8 24 29 47 56 60 61 63 2 18 23 41 50
52 53 55 59 10 15 33 42 36 37 39 43 51 72 17 26
31 32 34 38 46 62 12 21 13 14 16 20 28 44 49 3
7 8 10 14 38 43 61 70 59 60 62 66 1 17 40 49
D5 41 42 44 48 56 72 4 31 32 33 35 39 47 63 68 13
23 25 29 37 53 58 3 12 21 24 28 36 52 57 2 11
19 20 26 34 50 55 73 9 15 16 18 30 46 51 69 5
8 9 11 15 23 39 44 62 72 1 5 13 29 34 52 61
D6 70 73 4 12 28 33 51 60 68 69 2 10 26 31 49 58
56 57 59 63 14 19 37 46 40 41 43 47 55 3 21 30
35 36 38 42 50 66 16 25 17 18 20 24 32 48 53 7
Chapter 7. Application of Finite Geometry to Operational Research 118

Table 7.13: 8-secants in PG(2, 8)

9 10 12 16 24 40 63 72 55 56 58 62 70 13 18 36
D7 46 48 52 60 3 8 26 35 44 47 51 59 2 7 25 34
42 43 49 57 73 5 23 32 38 39 41 53 69 1 19 28
30 31 33 37 61 66 11 20 14 15 17 21 29 50 68 4
12 13 15 19 43 48 66 2 69 70 72 3 11 32 50 59
D8 46 47 49 53 61 4 9 36 37 38 40 44 52 68 73 18
28 30 34 42 58 63 8 17 26 29 33 41 57 62 7 16
24 25 31 39 55 60 5 14 20 21 23 35 51 56 1 10
18 19 21 25 33 49 72 8 73 1 3 7 15 31 36 63
D9 55 57 61 69 12 17 35 44 53 56 60 68 11 16 34 43
51 52 58 66 9 14 32 41 47 48 50 62 5 10 28 37
39 40 42 46 70 2 20 29 23 24 26 30 38 59 4 13

Problem 5: A school has ten girls. How can they arranged to walk for six days, such
that no two girls walk together twice?

Solution The target of this problem is to divide 10 girls into groups of 2 and 3 girls.
From the classification of (10, 3)-arcs K in PG(2, 5), with i-secant distribution of type
(10, 15, 0, 0, 6). Then, the ten 3-secants represent the groups of 3 girls and the fifteen 2-
secants represent the group of 2 girls, and the 10 girls are represented by the 10 points.
Table 7.14 shows the points of K which correspond to 10 girls, Table 7.15 shows the
groups of 3 girls and Table 7.16 shows the groups of 2 girls.

Table 7.14

g = girl g1 g2 g3 g4 g5 g6 g7 g8 g9 g10

Point in PG(2,5) P1 P2 P3 P5 P6 P7 P9 P11 P17 P24

Table 7.15

(g1 , g3 , g9 ) (g2 , g4 , g10 ) (g3 , g4 , g5 ) (g1 , g5 , g7 ) (g2 , g6 , g7 )

(g3 , g8 , g10 ) (g2 , g5 , g8 ) (g4 , g7 , g9 ) (g1 , g6 , g8 ) (g6 , g9 , g10 )

Chapter 7. Application of Finite Geometry to Operational Research 119

Table 7.16

(g1 , g2 ) (g2 , g3 ) (g5 , g6 ) (g4 , g6 ) (g7 , g8 )

(g1 , g4 ) (g3 , g6 ) (g5 , g9 ) (g2 , g9 ) (g3 , g7 )

(g4 , g8 ) (g5 , g10 ) (g7 , g10 ) (g8 , g9 ) (g1 , g10 )

So, the solution of this problem is the following:

D1 : (g1 , g3 , g9 ) (g2 , g5 , g8 ) (g4 , g6 ) (g7 , g10 )

D2 : (g2 , g4 , g10 ) (g1 , g6 , g8 ) (g3 , g7 ) (g5 , g9 )
D3 : (g3 , g4 , g5 ) (g2 , g6 , g7 ) (g1 , g10 ) (g8 , g9 )
D4 : (g1 , g5 , g7 ) (g6 , g9 , g10 ) (g2 , g3 ) (g4 , g8 )
D5 : (g3 , g8 , g10 ) (g4 , g7 , g9 ) (g1 , g2 ) (g5 , g6 )
D6 : (g1 , g4 ) (g2 , g9 ) (g3 , g6 ) (g7 , g8 ) (g5 , g10 )

Here, each girl participated in 3 groups of 3 girls and 3 groups of 2 girls.

7.5 Chapter Summary

In this chapter some application of finite geometry to operational research, probability

and scheduling problems are explained. Constraint satisfaction problems is considered
an important application to (k, n)-arcs to AG(2, q), PG(2, q). Also, a generalisation of
Kirkman’s problem is showed. Finally, the chapter presents, in details, solutions of five
problems to show the application of projective plane to operational research.
Appendix A

Points and Lines

A.1 Projective Plane of Order Five

Since PG(2, 5) contains 31 points and 31 lines, every line contains 6 points and every
point passes through it 6 lines. To find the points of PG(2, 5). Let

f (x) = x3 − x2 − 1

be an irreducible polynomial over F5 , then the matrix

 
0 1 0
 
C(f ) =  0 0 1
 

 
1 0 1

gives a cyclic projectivity by right multiplication on the points of PG(2, 5). Let P0
be represented by vector (1, 0, 0). Then P0 T i = Pi , i = 0, 1, ..., 30, are 31 points of
PG(2, 5). To find the lines of PG(2, 5) Let L1 be the line which contains the points
1, 2, 15, 21, 25, 30, then L1 T i = Li+1 , i = 1, 2, ..., 30 are the lines of PG(2, 5). Table
A.1 and Table A.2 show the points and lines of PG(2, 5).

120
Appendix A. Points and Lines 121

Table A.1: Points of PG(2, 5)

P1 (1, 0, 0) P7 (1, 3, 4) P13 (1, 2, 2) P19 (1, 1, 3) P25 (1, 1, 0) P31 (0, 1, 4)

P2 (0, 1, 0) P8 (1, 4, 3) P14 (1, 3, 2) P20 (1, 2, 3) P26 (0, 1, 1)

P3 (0, 0, 1) P9 (1, 2, 4) P15 (1, 3, 0) P21 (1, 2, 0) P27 (1, 0, 2)

P4 (1, 0, 1) P10 (1, 4, 4) P16 (0, 1, 3) P22 (0, 1, 2) P28 (1, 3, 1)

P5 (1, 1, 1) P11 (1, 4, 2) P17 (1, 0, 3) P23 (1, 0, 4) P29 (1, 1, 4)

P6 (1, 1, 2) P12 (1, 3, 3) P18 (1, 2, 1) P24 (1, 4, 1) P30 (1, 4, 0)

Table A.2: Lines of PG(2, 5)

L1 P1 P2 P15 P21 P25 P30 L17 P17 P18 P31 P6 P10 P15

L2 P2 P3 P16 P22 P26 P31 L18 P18 P19 P1 P7 P11 P16

L3 P3 P4 P17 P23 P27 P1 L19 P19 P20 P2 P8 P12 P17

L4 P4 P5 P18 P24 P28 P2 L20 P20 P21 P3 P9 P13 P18

L5 P5 P6 P19 P25 P29 P3 L21 P21 P22 P4 P10 P14 P19

L6 P6 P7 P20 P26 P30 P4 L22 P22 P23 P5 P11 P15 P20

L7 P7 P8 P21 P27 P31 P5 L23 P23 P24 P6 P12 P16 P21

L8 P8 P9 P22 P28 P1 P6 L24 P24 P25 P7 P13 P17 P22

L9 P9 P10 P23 P29 P2 P7 L25 P25 P26 P8 P14 P18 P23

L10 P10 P11 P24 P30 P3 P8 L26 P26 P27 P9 P15 P19 P24

L11 P11 P12 P25 P31 P4 P9 L27 P27 P28 P10 P16 P20 P25

L12 P12 P13 P26 P1 P5 P10 L28 P28 P29 P11 P17 P21 P26

L13 P13 P14 P27 P2 P6 P11 L29 P29 P30 P12 P18 P22 P27

L14 P14 P15 P28 P3 P7 P12 L30 P30 P31 P13 P19 P23 P28

L15 P15 P16 P29 P4 P8 P13 L31 P31 P1 P14 P20 P24 P29

L16 P16 P17 P30 P5 P9 P14

Appendix A. Points and Lines 122

A.2 Projective Plane of Order Seven

Since PG(2, 7) contains 57 points and 57 lines, every line contains 8 points and every
point passes through it 8 lines. To find the points of PG(2, 7). Let

f (x) = x3 − x2 − 2x − 4

be an irreducible polynomial over F7 , then the matrix

 
0 1 0
 
C(f ) =  0 0 1
 

 
4 2 1

gives a cyclic projectivity by right multiplication on the points of PG(2, 7). Let P0
be represented by vector (1, 0, 0). Then P0 T i = Pi , i = 0, 1, ..., 56, are 57 points of
PG(2, 7). To find the lines of PG(2, 7) Let L1 be the line which contains the points
1, 2, 8, 25, 37, 39, 50, 55, then L1 T i = Li+1 , i = 1, 2, ..., 56 are the lines of PG(2, 7).
Table A.3 and Table A.4 show the points and lines of PG(2, 7).

Table A.3: Points of PG(2, 7)

P1 (1, 0, 0) P11 (1, 3, 5) P21 (1, 5, 3) P31 (1, 5, 4) P41 (1, 4, 6) P51 (0, 1, 1)

P2 (0, 1, 0) P12 (1, 3, 6) P22 (1, 0, 3) P32 (1, 1, 1) P42 (1, 2, 1) P52 (1, 4, 4)

P3 (0, 0, 1) P13 (1, 2, 3) P23 (1, 0, 2) P33 (1, 6, 4) P43 (1, 6, 6) P53 (1, 1, 4)

P4 (1, 4, 2) P14 (1, 0, 1) P24 (1, 5, 2) P34 (1, 1, 5) P44 (1, 2, 4) P54 (1, 1, 6)

P5 (1, 5, 6) P15 (1, 6, 2) P25 (1, 5, 0) P35 (1, 3, 1) P45 (1, 1, 3) P55 (1, 2, 0)

P6 (1, 2, 6) P16 (1, 5, 1) P26 (0, 1, 5) P36 (1, 6, 1) P46 (1, 0, 5) P56 (0, 1, 2)

P7 (1, 2, 5) P17 (1, 6, 5) P27 (1, 4, 1) P37 (1, 6, 0) P47 (1, 3, 2) P57 (1, 4, 3)

P8 (1, 3, 0) P18 (1, 3, 3) P28 (1, 6, 3) P38 (0, 1, 6) P48 (1, 5, 5)

P9 (0, 1, 3) P19 (1, 0, 4) P29 (1, 0, 6) P39 (1, 4, 0) P49 (1, 3, 4)

P10 (1, 4, 5) P20 (1, 1, 2) P30 (1, 2, 2) P40 (0, 1, 4) P50 (1, 1, 0)
Appendix A. Points and Lines 123

Table A.4: Lines of PG(2, 7)

L1 P1 P2 P8 P25 P37 P39 P50 P55 L30 P30 P31 P37 P54 P9 P11 P22 P27
L2 P2 P3 P9 P26 P38 P40 P51 P56 L31 P31 P32 P38 P55 P10 P12 P23 P28
L3 P3 P4 P10 P27 P39 P41 P52 P57 L32 P32 P33 P39 P56 P11 P13 P24 P29
L4 P4 P5 P11 P28 P40 P42 P53 P1 L33 P33 P34 P40 P57 P12 P14 P25 P30
L5 P5 P6 P12 P29 P41 P43 P54 P2 L34 P34 P35 P41 P1 P13 P15 P26 P31
L6 P6 P7 P13 P30 P42 P44 P55 P3 L35 P35 P36 P42 P2 P14 P16 P27 P32
L7 P7 P8 P14 P31 P43 P45 P56 P4 L36 P36 P37 P43 P3 P15 P17 P28 P33
L8 P8 P9 P15 P32 P44 P46 P57 P5 L37 P37 P38 P44 P4 P16 P18 P29 P34
L9 P9 P10 P16 P33 P45 P47 P1 P6 L38 P38 P39 P45 P5 P17 P19 P30 P35
L10 P10 P11 P17 P34 P46 P48 P2 P7 L39 P39 P40 P46 P6 P18 P20 P31 P36
L11 P11 P12 P18 P35 P47 P49 P3 P8 L40 P40 P41 P47 P7 P19 P21 P32 P37
L12 P12 P13 P19 P36 P48 P50 P4 P9 L41 P41 P42 P48 P8 P20 P22 P33 P38
L13 P13 P14 P20 P37 P49 P51 P5 P10 L42 P42 P43 P49 P9 P21 P23 P34 P39
L14 P14 P15 P21 P38 P50 P52 P6 P11 L43 P43 P44 P50 P10 P22 P24 P35 P40
L15 P15 P16 P22 P39 P51 P53 P7 P12 L44 P44 P45 P51 P11 P23 P25 P36 P41
L16 P16 P17 P23 P40 P52 P54 P8 P13 L45 P45 P46 P52 P12 P24 P26 P37 P42
L17 P17 P18 P24 P41 P53 P55 P9 P14 L46 P46 P47 P53 P13 P25 P27 P38 P43
L18 P18 P19 P25 P42 P54 P56 P10 P15 L47 P47 P48 P54 P14 P26 P28 P39 P44
L19 P19 P20 P26 P43 P55 P57 P11 P16 L48 P48 P49 P55 P15 P27 P29 P40 P45
L20 P20 P21 P27 P44 P56 P1 P12 P17 L49 P49 P50 P56 P16 P28 P30 P41 P46
L21 P21 P22 P28 P45 P57 P2 P13 P18 L50 P50 P51 P57 P17 P29 P31 P42 P47
L22 P22 P23 P29 P46 P1 P3 P14 P19 L51 P51 P52 P1 P18 P30 P32 P43 P48
L23 P23 P24 P30 P47 P2 P4 P15 P20 L52 P52 P53 P2 P19 P31 P33 P44 P49
L24 P24 P25 P31 P48 P3 P5 P16 P21 L53 P53 P54 P3 P20 P32 P34 P45 P50
L25 P25 P26 P32 P49 P4 P6 P17 P22 L54 P54 P55 P4 P21 P33 P35 P46 P51
L26 P26 P27 P33 P50 P5 P7 P18 P23 L55 P55 P56 P5 P22 P34 P36 P47 P52
L27 P27 P28 P34 P51 P6 P8 P19 P24 L56 P56 P57 P6 P23 P35 P37 P48 P53
L28 P28 P29 P35 P52 P7 P9 P20 P25 L57 P57 P1 P7 P24 P36 P38 P49 P54
L29 P29 P30 P36 P53 P8 P10 P21 P26
Appendix A. Points and Lines 124

A.3 Projective Plane of Order Eight

F8 = {0, 1, α, α2 , α3 , α4 , α5 , α6 : α3 + α2 + 1 = 2 = 0}
 
0 1 0
 
C(f ) =  0 0 1  .
 
 
4
α 1 0

Table A.5 and Table A.6 show the points and lines of PG(2, 8).

Table A.5: Points of PG(2, 8)

P1 (1, 0, 0) P2 (0, 1, 0) P3 (0, 0, 1) P4 (1, α3 , 0) P5 (0, 1, α3 ) P6 (1, α3 , 1)

P7 (1, 0, α6 ) P8 (1, α6 , 0) P9 (0, 1, α6 ) P10 (1, α3 , α4 ) P11 (1, α4 , α2 ) P12 (1, 1, α5 )
P13 (1, α2 , α5 ) P14 (1, α2 , 1) P15 (1, 0, α5 ) P16 (1, α2 , 0) P17 (0, 1, α2 ) P18 (1, α3 , α)
P19 (1, α5 , α5 ) P20 (1, α2 , α3 ) P21 (1, α, α2 ) P22 (1, 1, α2 ) P23 (1, 1, α) P24 (1, α5 , α2 )
P25 (1, 1, α6 ) P26 (1, α6 , α4 ) P27 (1, α4 , α5 ) P28 (1, α2 , α2 ) P29 (1, 1, α3 ) P30 (1, α, 1)
P31 (1, 0, α4 ) P32 (1, α4 , 0) P33 (0, 1, α4 ) P34 (1, α3 , α6 ) P35 (1, α6 , 1) P36 (1, 0, α2 )
P37 (1, 1, 0) P38 (0, 1, 1) P39 (1, α3 , α3 ) P40 (1, α, α3 ) P41 (1, α, α) P42 (1, α5 , α3 )
P43 (1, α, α5 ) P44 (1, α2 , α6 ) P45 (1, α6 , α6 ) P46 (1, α6 , α3 ) P47 (1, α, α6 ) P48 (1, α6 , α5 )
P49 (1, α2 , α4 ) P50 (1, α4 , α) P51 (1, α5 , α6 ) P52 (1, α6 , α2 ) P53 (1, 1, 1) P54 (1, 0, α3 )
P55 (1, α, 0) P56 (0, 1, α) P57 (1, α3 , α2 ) P58 (1, 1, α4 ) P59 (1, α4 , α6 ) P60 (1, α6 , α)
P61 (1, α5 , α) P62 (1, α5 , 1) P63 (1, 0, α) P64 (1, α5 , 0) P65 (0, 1, α5 ) P66 (1, α3 , α5 )
P67 (1, α2 , α) P68 (1, α5 , α4 ) P69 (1, α4 , α4 ) P70 (1, α4 , α3 ) P71 (1, α, α4 ) P72 (1, α4 , 1)
P73 (1, 0, 1)

Table A.6: Lines of PG(2, 8)

L1 P1 P2 P4 P8 P16 P32 P37 P55 P64

L2 P2 P3 P5 P9 P17 P33 P38 P56 P65
..
.
L73 P73 P1 P3 P7 P15 P31 P36 P54 P63
Appendix A. Points and Lines 125

A.4 Projective Plane of Order Nine

F9 = {0, 1, 2, β, β 2 , β 3 , β 4 , β 5 , β 6 : β 2 − β − 1 = 3 = 0}
 
0 1 0
 
C(f ) =  0 0 1  .
 
 
1 0 β

Table A.7 and Table A.8 show the points and lines of PG(2, 9).

Table A.7: Points of PG(2, 9)

P1 (1, 0, 0) P2 (0, 1, 0) P3 (0, 0, 1) P4 (1, 0, β) P5 (1, β 5 , β) P6 (1, β 5 , β 4 )

P7 (1, β 2 , 1) P8 (1, 1, 2) P9 (1, β 6 , β 3 ) P10 (1, β 3 , 2) P11 (1, β 6 , 2) P12 (1, β 6 , 0)
P13 (0, 1, β 6 ) P14 (1, 0, 1) P15 (1, 1, β) P16 (1, β 5 , 0) P17 (0, 1, β 5 ) P18 (1, 0, β 5 )
P19 (1, β, β) P20 (1, β 5 , β 2 ) P21 (1, β 4 , 2) P22 (1, β 6 , β 4 ) P23 (1, β 2 , β 5 ) P24 (1, β, β 6 )
P25 (1, 2, 2) P26 (1, β 6 , β 2 ) P27 (1, β 4 , β 4 ) P28 (1, β 2 , β 2 ) P29 (1, β 4 , β 2 ) P30 (1, β 4 , β 5 )
P31 (1, β, β 3 ) P32 (1, β 3 , 0) P33 (0, 1, β 3 ) P34 (1, 0, β 4 ) P35 (1, β 2 , β) P36 (1, β 5 , 1)
P37 (1, 1, 0) P38 (0, 1, 1) P39 (1, 0, 4) P40 (1, β 4 , β) P41 (1, β 5 , 2) P42 (1, β 6 , β 6 )
P43 (1, 2, β 2 ) P44 (1, β 4 , 0) P45 (0, 1, β 4 ) P46 (1, 0, 2) P47 (1, β 6 , β) P48 (1, β 5 , β 6 )
P49 (1, 2, β 3 ) P50 (1, β 3 , β 6 ) P51 (1, 2, β 6 ) P52 (1, 2, β 5 ) P53 (1, β, 2) P54 (1, β 6 , 1)
P55 (1, 1, β 3 ) P56 (1, β 3 , β 4 ) P57 (1, β 2 , β 3 ) P58 (1, β 3 , β 3 ) P59 (1, β 3 , β 2 ) P60 (1, β 4 , 1)
P61 (1, 1, β 6 ) P62 (1, 2, 1) P63 (1, 1, 1) P64 (1, 1, β 2 ) P65 (1, β 4 , β 6 ) P66 (1, 2, 0)
P67 (0, 1, 2) P68 (1, 0, β 3 ) P69 (1, β 3 , β) P70 (1, β 5 , β 5 ) P71 (1, β, β 2 ) P72 (1, β 4 , β 3 )
P73 (1, β 3 , 1) P74 (1, 1, β 4 ) P75 (1, β 2 , 2) P76 (1, β 6 , β 5 ) P77 (1, β, 1) P78 (1, 1, β 5 )
P79 (1, β, β 5 ) P80 (1, β, β 4 ) P81 (1, β 2 , β 6 ) P82 (1, 2, β 4 ) P83 (1, β 2 , β 4 ) P84 (1, β 2 , 0)
P85 (0, 1, β 2 ) P86 (1, 0, β 6 ) P87 (1, 2, β) P88 (1, β 5 , β 3 ) P89 (1, β 3 , β 5 7) P90 (1, β, 0)
P91 (0, 1, β)

Table A.8: Lines of PG(2, 9)

L1 P1 P2 P12 P16 P32 P37 P44 P66 P84 P90

L2 P2 P3 P13 P17 P33 P38 P45 P67 P85 P91
..
.
L91 P91 P1 P11 P15 P31 P36 P43 P65 P83 P89
Appendix A. Points and Lines 126

A.5 Projective Plane of Order Eleven

0 1 0
 
C(f ) =  0 0 1  .
 
 
3 1 1

Table A.9 and Table A.10 show the points and lines of PG(2, 11).
Table A.9: Points of PG(2, 11)

P1 (1, 0, 0) P2 (0, 1, 0) P3 (0, 0, 1) P4 (1, 4, 4) P5 (1, 5, 8) P6 (1, 10, 1)

P7 (1, 8, 0) P8 (0, 1, 8) P9 (1, 4, 10) P10 (1, 0, 10) P11 (1, 0, 4) P12 (1, 5, 4)
P13 (1, 5, 9) P14 (1, 2, 5) P15 (1, 7, 10) P16 (1, 0, 9) P17 (1, 2, 4) P18 (1, 5, 6)
P19 (1, 1, 0) P20 (0, 1, 1) P21 (1, 4, 8) P22 (1, 10, 6) P23 (1, 1, 7) P24 (1, 3, 3)
P25 (1, 9, 8) P26 (1, 10, 3) P27 (1, 9, 10) P28 (1, 0, 1) P29 (1, 8, 4) P30 (1, 5, 1)
P31 (1, 8, 2) P32 (1, 6, 9) P33 (1, 2, 3) P34 (1, 9, 3) P35 (1, 9, 5) P36 (1, 7, 9)
P37 (1, 2, 1) P38 (1, 8, 1) P39 (1, 8, 3) P40 (1, 9, 0) P41 (0, 1, 9) P42 (1, 4, 2)
P43 (1, 6, 1) P44 (1, 8, 6) P45 (1, 1, 2) P46 (1, 6, 6) P47 (1, 1, 8) P48 (1, 10, 10)
P49 (1, 0, 8) P50 (1, 10, 4) P51 (1, 5, 3) P52 (1, 9, 7) P53 (1, 3, 6) P54 (1, 1, 6)
P55 (1, 1, 1) P56 (1, 8, 8) P57 (1, 10, 8) P58 (1, 10, 9) P59 (1, 2, 6) P60 (1, 1, 9)
P61 (1, 2, 2) P62 (1, 6, 8) P63 (1, 10, 7) P64 (1, 3, 5) P65 (1, 7, 2) P66 (1, 6, 7)
P67 (1, 3, 9) P68 (1, 2, 9) P69 (1, 2, 0) P70 (0, 1, 2) P71 (1, 4, 6) P72 (1, 1, 3)
P73 (1, 9, 9) P74 (1, 2, 8) P75 (1, 10, 5) P76 (1, 7, 1) P77 (1, 8, 10) P78 (1, 0, 5)
P79 (1, 7, 4) P80 (1, 5, 0) P81 (0, 1, 5) P82 (1, 4, 7) P83 (1, 3, 0) P84 (0, 1, 3)
P85 (1, 4, 9) P86 (1, 2, 7) P87 (1, 3, 2) P88 (1, 6, 10) P89 (1, 0, 2) P90 (1, 6, 4)
P91 (1, 5, 10) P92 (1, 0, 6) P93 (1, 1, 4) P94 (1, 5, 5) P95 (1, 7, 8) P96 (1, 10, 2)
P97 (1, 6, 2) P98 (1, 6, 5) P99 (1, 7, 0) P100 (0, 1, 7) P101 (1, 4, 3) P102 (1, 9, 2)
P103 (1, 6, 0) P104 (0, 1, 6) P105 (1, 4, 1) P106 (1, 8, 9) P107 (1, 2, 10) P108 (1, 0, 7)
P109 (1, 3, 4) P110 (1, 5, 7) P111 (1, 3, 10) P112 (1, 0, 3) P113 (1, 9, 4) P114 (1, 5, 2)
P115 (1, 6, 3) P116 (1, 9, 1) P117 (1, 8, 7) P118 (1, 3, 7) P119 (1, 3, 1) P120 (1, 8, 5)
P121 (1, 7, 6) P122 (1, 1, 5) P123 (1, 7, 7) P124 (1, 3, 8) P125 (1, 10, 0) P126 (0, 1, 10)
P127 (1, 4, 0) P128 (0, 1, 4) P129 (1, 4, 5) P130 (1, 7, 5) P131 (1, 7, 3) P132 (1, 9, 6)
P133 (1, 1, 10)
Appendix A. Points and Lines 127

Table A.10: Lines of PG(2, 11)

L1 P1 P 2 P9 P22 P40 P44 P49 P55 P74 P106 P118 P132

L2 P2 P3 P10 P23 P41 P45 P50 P56 P75 P107 P119 P133
..
.
L133 P133 P1 P8 P21 P39 P43 P48 P54 P73 P105 P117 P131

A.6 Projective Plane of Order Thirteen

0 1 0
 
C(f ) =  0 0 1  .
 
 
2 0 1

Table A.11, Table A.12 and Table A.13 show the points and lines of PG(2, 13).

Table A.11: P1 , . . . P84 of PG(2, 13)

P1 (1, 0, 0) P2 (0, 1, 0) P3 (0, 0, 1) P4 (1, 0, 7) P5 (1, 1, 7) P6 (1, 1, 8)

P7 (1, 9, 3) P8 (1, 11, 2) P9 (1, 10, 0) P10 (0, 1, 10) P11 (1, 0, 9) P12 (1, 8, 7)
P13 (1, 1, 2) P14 (1, 10, 4) P15 (1, 5, 5) P16 (1, 4, 1) P17 (1, 7, 9) P18 (1, 8, 11)
P19 (1, 3, 5) P20 (1, 4, 6) P21 (1, 12, 3) P22 (1, 11, 9) P23 (1, 8, 4) P24 (1, 5, 8)
P25 (1, 9, 0) P26 (0, 1, 9) P27 (1, 0, 2) P28 (1, 10, 7) P29 (1, 1, 4) P30 (1, 5, 12)
P31 (1, 6, 11) P32 (1, 3, 12) P33 (1, 6, 12) P34 (1, 6, 4) P35 (1, 5, 11) P36 (1, 3, 9)
P37 (1, 8, 5) P38 (1, 4, 0) P39 (0, 1, 4) P40 (1, 0, 12) P41 (1, 6, 7) P42 (1, 1, 0)
P43 (0, 1, 1) P44 (1, 0, 1) P45 (1, 7, 7) P46 (1, 1, 1) P47 (1, 7, 1) P48 (1, 7, 4)
P49 (1, 5, 3) P50 (1, 11, 10) P51 (1, 2, 3) P52 (1, 11, 3) P53 (1, 11, 11) P54 (1, 3, 1)
P55 (1, 7, 2) P56 (1, 10, 12) P57 (1, 6, 2) P58 (1, 10, 2) P59 (1, 10, 3) P60 (1, 11, 0)
P61 (0, 1, 11) P62 (1, 0, 10) P63 (1, 2, 7) P64 (1, 1, 9) P65 (1, 8, 2) P66 (1, 10, 9)
P67 (1, 8, 9) P68 (1, 8, 6) P69 (1, 12, 12) P70 (1, 6, 1) P71 (1, 7, 10) P72 (1, 2, 8)
P73 (1, 9, 12) P74 (1, 6, 9) P75 (1, 8, 3) P76 (1, 11, 4) P77 (1, 5, 10) P78 (1, 2, 4)
P79 (1, 5, 4) P80 (1, 5, 6) P81 (1, 12, 2) P82 (1, 10, 10) P83 (1, 2, 1) P84 (1, 7, 8)
Appendix A. Points and Lines 128

Table A.12: P85 , . . . P183 of PG(2, 13)

P85 (1, 9, 5) P86 (1, 4, 4) P87 (1, 5, 1) P88 (1, 7, 3) P89 (1, 11, 6) P90 (1, 12, 9)
P91 (1, 8, 12) P92 (1, 6, 3) P93 (1, 11, 8) P94 (1, 9, 2) P95 (1, 10, 6) P96 (1, 12, 10)
P97 (1, 2, 5) P98 (1, 4, 2) P99 (1, 10, 8) P100 (1, 9, 6) P101 (1, 12, 11) P102 (1, 3, 4)
P103 (1, 5, 9) P104 (1, 8, 8) P105 (1, 9, 1) P106 (1, 7, 5) P107 (1, 4, 9) P108 (1, 8, 0)
P109 (0, 1, 8) P110 (1, 0, 3) P111 (1, 11, 7) P112 (1, 1, 5) P113 (1, 4, 11) P114 (1, 3, 6)
P115 (1, 12, 4) P116 (1, 5, 2) P117 (1, 10, 5) P118 (1, 4, 8) P119 (1, 9, 4) P120 (1, 5, 0)
P121 (0, 1, 5) P122 (1, 0, 11) P123 (1, 3, 7) P124 (1, 1, 10) P125 (1, 2, 9) P126 (1, 8, 10)
P127 (1, 2, 10) P128 (1, 2, 11) P129 (1, 3, 0) P130 (0, 1, 3) P131 (1, 0, 5) P132 (1, 4, 7)
P133 (1, 1, 11) P134 (1, 3, 10) P135 (1, 2, 0) P136 (0, 1, 2) P137 (1, 0, 4) P138 (1, 5, 7)
P139 (1, 1, 12) P140 (1, 6, 0) P141 (0, 1, 6) P142 (1, 0, 6) P143 (1, 12, 7) P144 (1, 1, 6)
P145 (1, 12, 6) P146 (1, 12, 8) P147 (1, 9, 11) P148 (1, 3, 8) P149 (1, 9, 8) P150 (1, 9, 10)
P151 (1, 2, 12) P152 (1, 6, 6) P153 (1, 12, 1) P154 (1, 7, 0) P155 (0, 1, 7) P156 (1, 0, 8)
P157 (1, 9, 7) P158 (1, 1, 3) P159 (1, 11, 5) P160 (1, 4, 12) P161 (1, 6, 5) P162 (1, 4, 5)
P163 (1, 4, 10) P164 (1, 2, 2) P165 (1, 10, 1) P166 (1, 7, 12) P167 (1, 6, 10) P168 (1, 2, 6)
P169 (1, 12, 5) P170 (1, 4, 3) P171 (1, 11, 12) P172 (1, 6, 8) P173 (1, 9, 9) P174 (1, 8, 1)
P175 (1, 7, 11) P176 (1, 3, 2) P177 (1, 10, 11) P178 (1, 3, 11) P179 (1, 3, 3) P180 (1, 11, 1)
P181 (1, 7, 6) P182 (1, 12, 0) P183 (0, 1, 12)

Table A.13: Lines of PG(2, 13)

L1 P1 P2 P9 P25 P38 P42 P60 P108 P120 P129 P135 P140 P154 P182
L2 P2 P3 P10 P26 P39 P43 P61 P109 P121 P130 P136 P141 P155 P183
..
.
L183 P183 P1 P8 P24 P37 P41 P59 P107 P119 P128 P134 P139 P153 P181
Appendix A. Points and Lines 129

A.7 Projective Plane of Order Sixteen

F16 = {0, 1, γ, γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 , γ 9 , γ 10 , γ 11 , γ 12 , γ 13 , γ 14 : γ 4 + γ + 1 = 2 = 0}
 
0 1 0
 
C(f ) =  0 0 1 .
 
 
γ 0 γ

Table A.14, . . . , Table A.17 show the points and lines of PG(2, 16).

Table A.14: P1 , . . . P85 of PG(2, 16)

P1 (1, 0, 0) P2 (0, 1, 0) P3 (0, 0, 1) P4 (1, 0, 1) P5 (1, γ 14 , 1)

P6 (1, γ 14 , γ 6 ) P7 (1, γ 8 , γ 9 ) P8 (1, γ 5 , γ 6 ) P9 (1, γ 8 , γ 6 ) P10 (1, γ 8 , γ 4 )

P11 (1, γ 10 , γ 14 ) P12 (1, 1, γ 5 ) P13 (1, γ 9 , γ 7 ) P14 (1, γ 7 , γ 4 ) P15 (1, γ 10 , γ 8 )

P16 (1, γ 6 , γ 4 ) P17 (1, γ 10 , γ 4 ) P18 (1, γ 10 , γ 10 ) P19 (1, γ 4 , γ 3 ) P20 (1, γ 11 , 0)

P21 (0, 1, γ 11 ) P22 (1, 0, γ 14 ) P23 (1, 1, 1) P24 (1, γ 14 , γ 3 ) P25 (1, γ 11 , γ 5 )

P26 (1, γ 9 , γ 10 ) P27 (1, γ 4 , γ 6 ) P28 (1, γ 8 , γ 11 ) P29 (1, γ 3 , γ 12 ) P30 (1, γ 2 , γ 10 )

P31 (1, γ 4 , γ 13 ) P32 (1, γ, α10 ) P33 (1, γ 4 , γ 10 ) P34 (1, γ 4 , γ 2 ) P35 (1, γ 12 , γ 4 )

P36 (1, γ 10 , γ 9 ) P37 (1, γ 5 , 0) P38 (0, 1, γ 5 ) P39 (1, 0, γ 7 ) P40 (1, γ 7 , 1)

P41 (1, γ 14 , γ 13 ) P42 (1, γ, 0) P43 (0, 1, γ) P44 (1, 0, γ 6 ) P45 (1, γ 8 , 1)

P46 (1, γ 14 , γ 9 ) P47 (1, γ 5 , γ) P48 (1, γ 13 , γ 14 ) P49 (1, 1, γ 6 ) P50 (1, γ 8 , γ 2 )

P51 (1, γ 12 , γ 10 ) P52 (1, γ 4 , γ 4 ) P53 (1, γ 10 , γ 3 ) P54 (1, γ 11 , γ 13 ) P55 (1, γ, γ 11 )

P56 (1, γ 3 , γ) P57 (1, γ 13 , γ 4 ) P58 (1, γ 10 , γ 2 ) P59 (1, γ 12 , γ 9 ) P60 (1, γ 5 , γ 8 )

P61 (1, γ 6 , γ 12 ) P62 (1, γ 2 , γ 2 ) P63 (1, γ 12 , γ 3 ) P64 (1, γ 11 , γ 2 ) P65 (1, γ 12 , γ 2 )

P66 (1, γ 12 , γ 7 ) P67 (1, γ 7 , γ) P68 (1, γ 13 , γ 10 ) P69 (1, γ 4 , γ 8 ) P70 (1, γ 6 , γ 5 )

P71 (1, γ 9 , 0) P72 (0, 1, γ 9 ) P73 (1, 0, γ 10 ) P74 (1, γ 4 , 1) P75 (1, γ 14 , γ 14 )

P76 (1, 1, γ 3 ) P77 (1, γ 11 , γ 12 ) P78 (1, γ 2 , γ 6 ) P79 (1, γ 8 , γ 5 ) P80 (1, γ 9 , γ 8 )

P81 (1, γ 6 , 0) P82 (0, 1, γ 6 ) P83 (1, 0, γ 2 ) P84 (1, γ 12 , 1) P85 (1, γ 14 , γ 12 )
Appendix A. Points and Lines 130

Table A.15: P86 , . . . P200 of PG(2, 16)

P86 (1, γ 2 , γ 4 ) P87 (1, γ 10 , γ 11 ) P88 (1, γ 3 , γ 6 ) P89 (1, γ 8 , γ 12 ) P90 (1, γ 2 , γ 5 )

P91 (1, γ 9 , γ 12 ) P92 (1, γ 2 , γ 12 ) P93 (1, γ 2 , γ) P94 (1, γ 13 , 0) P95 (0, 1, γ 13 )

P96 (1, 0, γ 4 ) P97 (1, γ 10 , 1) P98 (1, γ 14 , 8) P99 (1, γ 7 , γ 13 ) P100 (1, γ, γ 2 )

P101 (1, γ 12 , γ 6 ) P102 (1, γ 8 , γ 10 ) P103 (1, γ 4 , γ 11 ) P104 (1, γ 3 , γ 9 ) P105 (1, γ 5 , γ 2 )

P106 (1, γ 12 , γ 8 ) P107 (1, γ 6 , γ 14 ) P108 (1, 1, γ 13 ) P109 (1, γ, γ 4 ) P110 (1, γ 10 , γ 12 )

P111 (1, γ 2 , γ 11 ) P112 (1, γ 3 , γ 10 ) P113 (1, γ 4 , γ 9 ) P114 (1, γ 5 , γ 7 ) P115 (1, γ 7 , γ 11 )

P116 (1, γ 3 , γ 5 ) P117 (1, γ 9 , γ 11 ) P118 (1, γ 3 , γ 11 ) P119 (1, γ 3 , γ 13 ) P120 (1, 2, 2)

P121 (1, γ 13 , γ 3 ) P122 (1, γ 11 , γ 7 ) P123 (1, γ 7 , γ 14 ) P124 (1, 1, γ 9 ) P125 (1, γ 5 , γ 10 )

P126 (1, γ 4 , γ 7 ) P127 (1, γ 7 , γ 12 ) P128 (1, γ 2 , γ 7 ) P129 (1, γ 7 , γ 7 ) P130 (1, γ 7 , γ 3 )

P131 (1, γ 11 , γ 14 ) P132 (1, 1, γ 12 ) P133 (1, γ 2 , γ 8 ) P134 (1, γ 6 , γ 2 ) P135 (1, γ 12 , γ 14 )

P136 (1, 1, γ 11 ) P137 (1, γ 3 , γ 14 ) P138 (1, 1, γ 14 ) P139 (1, 1, 0) P140 (0, 1, 1)

P141 (1, 0, γ 3 ) P142 (1, γ 11 , 1) P143 (1, γ 14 , γ 5 ) P144 (1, γ 9 , γ 2 ) P145 (1, γ 12 , γ 13 )

P146 (1, γ, γ 6 ) P147 (1, γ 8 , γ 7 ) P148 (1, γ 7 , 0) P149 (0, 1, γ 7 ) P150 (1, 0, γ 9 )

P151 (1, γ 5 , 1) P152 (1, γ 14 , γ) P153 (1, γ 13 , γ 11 ) P154 (1, γ 3 , γ 4 ) P155 (1, γ 10 , γ 6 )

P156 (1, γ 8 , γ 14 ) P157 (1, 1, γ 2 ) P158 (1, γ 12 , γ 11 ) P159 (1, γ 3 , 0) P160 (0, 1, γ 3 )

P161 (1, 0, γ 12 ) P162 (1, γ 2 , 1) P163 (1, γ 14 , γ 4 ) P164 (1, γ 10 , γ 7 ) P165 (1, γ 7 , γ 8 )

P166 (1, γ 6 , γ 6 ) P167 (1, γ 8 , γ 3 ) P168 (1, γ 11 , γ) P169 (1, γ 13 , γ 7 ) P170 (1, γ 7 , γ 10 )

P171 (1, γ 4 , γ 12 ) P172 (1, γ 2 , γ 13 ) P173 (1, γ, γ 14 ) P174 (1, 1, γ 4 ) P175 (1, γ 10 , γ 5 )

P176 (1, γ 9 , γ) P177 (1, γ 13 , γ 9 ) P178 (1, γ 5 , γ 14 ) P179 (1, 1, γ 10 ) P180 (1, γ 4 , γ)

P181 (1, γ 13 , γ 8 ) P182 (1, γ 6 , γ) P183 (1, γ 13 , γ) P184 (1, γ 13 , γ 12 ) P185 (1, γ 2 , 0)

P186 (0, 1, γ 2 ) P187 (1, 0, γ 11 ) P188 (1, γ 3 , 1) P189 (1, γ 14 , γ 8 ) P190 (1, γ 6 , γ 10 )

P191 (1, γ 4 , γ 5 ) P192 (1, γ 9 , γ 6 ) P193 (1, γ 8 , γ 8 ) P194 (1, γ 6 , γ 3 ) P195 (1, γ 11 , γ 8 )

P196 (1, γ 6 , γ 8 ) P197 (1, γ 6 , γ 11 ) P198 (1, γ 3 , γ 7 ) P199 (1, γ 7 , γ 5 ) P200 (1, γ 9 , γ 4 )
Appendix A. Points and Lines 131

Table A.16: P201 , . . . P273 of PG(2, 16)

P201 (1, γ 10 , γ) P202 (1, γ 13 , γ 2 ) P203 (1, γ 12 , γ 5 ) P204 (1, γ 9 , γ 13 ) P205 (1, γ, γ 5 )
P206 (1, γ 9 , γ 5 ) P207 (1, γ 9 , γ 14 ) P208 (1, 1, γ 7 ) P209 (1, γ 7 , γ 9 ) P210 (1, γ 5 , γ 11 )
P211 (1, γ 3 , γ 2 ) P212 (1, γ 12 , 0) P213 (0, 1, γ 12 ) P214 (1, 0, γ 8 ) P215 (1, γ 6 , 1)
P216 (1, γ 14 , γ 10 ) P217 (1, γ 4 , γ 14 ) P218 (1, 1, γ) P219 (1, γ 13 , γ 6 ) P220 (1, γ 8 , γ 13 )
P221 (1, γ, γ 7 ) P222 (1, γ 7 , γ 2 ) P223 (1, γ 12 , γ) P224 (1, γ 13 , γ 5 ) P225 (1, γ 9 , γ 9 )
P226 (1, γ 5 , γ 3 ) P227 (1, γ 11 , γ 4 ) P228 (1, γ 10 , γ 13 ) P229 (1, γ, γ 12 ) P230 (1, γ, γ 14 )
P231 (1, 1, γ 8 ) P232 (1, γ 6 , γ 13 ) P233 (1, γ, γ 11 ) P234 (1, γ 5 , γ 13 ) P235 (1, γ, γ 13 )
P236 (1, γ, γ 8 ) P237 (1, γ 6 , γ 9 ) P238 (1, γ 5 , γ 12 ) P239 (1, γ 2 , γ 9 ) P240 (1, γ 5 , γ 9 )
P241 (1, γ 5 , γ 5 ) P242 (1, γ 9 , γ 3 ) P243 (1, γ 11 , γ 10 ) P244 (1, γ 4 , 0) P245 (0, 1, γ 4 )
P246 (1, 0, γ 5 ) P247 (1, γ 9 , 1) P248 (1, γ 14 , γ 2 ) P249 (1, γ 12 , γ 12 ) P250 (1, γ, γ 3 )
P251 (1, γ 11 , γ 6 ) P252 (1, γ 8 , γ) P253 (1, γ 13 , γ 13 ) P254 (1, γ, γ 3 ) P255 (1, γ 11 , γ 11 )
P256 (1, γ 3 , γ 3 ) P257 (1, γ 11 , γ 3 ) P258 (1, γ 11 , γ 9 ) P259 (1, γ 5 , γ 4 ) P260 (1, γ 10 , 0)
P261 (0, 1, γ 10 ) P262 (1, 0, γ) P263 (1, γ 13 , 1) P264 (1, γ 14 , γ 11 ) P265 (1, γ 3 , γ 8 )
P266 (1, γ 6 , γ 7 ) P267 (1, γ 7 , γ 6 ) P268 (1, γ 8 , 0) P269 (0, 1, γ 8 ) P270 (1, 0, γ 13 )
P271 (1, γ, 1) P272 (1, γ 14 , 0) P273 (0, 1, γ 14 )

Table A.17: Lines of PG(2, 16)

L1 P1 P2 P20 P37 P42 P71 P81 P94 P139 P148 P159 P185 P212 P244

P260 P268 P272

L2 P2 P3 P21 P38 P43 P72 P82 P95 P140 P149 P160 P186 P213 P245

P261 P269 P273

..
.
L307 P273 P1 P19 P36 P41 P70 P80 P93 P138 P147 P158 P184 P211 P243

P259 P267 P271

Appendix A. Points and Lines 132

A.8 Projective Plane of Order Seventeen

0 1 0
 
C(f ) =  0 0 1  .
 
 
1 0 8

Table A.18, . . . , Table A.21 show the points and lines of PG(2, 17).

Table A.18: P1 , . . . P114 of PG(2, 17)

P1 (1, 0, 0) P2 (0, 1, 0) P3 (0, 0, 1) P4 (1, 0, 8) P5 (1, 15, 8) P6 (1, 15, 12)

P7 (1, 10, 5) P8 (1, 7, 10) P9 (1, 12, 7) P10 (1, 5, 0) P11 (0, 1, 5) P12 (1, 0, 15)

P13 (1, 8, 8) P14 (1, 15, 9) P15 (1, 2, 4) P16 (1, 13, 0) P17 (0, 1, 13) P18 (1, 0, 12)

P19 (1, 10, 8) P20 (1, 15, 5) P21 (1, 7, 11) P22 (1, 14, 4) P23 (1, 13, 3) P24 (1, 6, 1)

P25 (1, 1, 14) P26 (1, 11, 2) P27 (1, 9, 5) P28 (1, 7, 3) P29 (1, 6, 16) P30 (1, 16, 2)

P31 (1, 9, 16) P32 (1, 16, 16) P33 (1, 16, 9) P34 (1, 2, 6) P35 (1, 3, 14) P36 (1, 11, 7)

P37 (1, 5, 12) P38 (1, 10, 7) P39 (1, 5, 7) P40 (1, 5, 16) P41 (1, 16, 3) P42 (1, 6, 2)

P43 (1, 9, 11) P44 (1, 14, 15) P45 (1, 8, 1) P46 (1, 1, 16) P47 (1, 16, 7) P48 (1, 5, 3)

P49 (1, 6, 4) P50 (1, 13, 1) P51 (1, 1, 4) P52 (1, 13, 4) P53 (1, 13, 7) P54 (1, 5, 5)

P55 (1, 7, 9) P56 (1, 2, 5) P57 (1, 7, 5) P58 (1, 7, 6) P59 (1, 3, 12) P60 (1, 10, 4)

P61 (1, 13, 2) P62 (1, 9, 6) P63 (1, 3, 1) P64 (1, 1, 11) P65 (1, 14, 5) P66 (1, 7, 4)

P67 (1, 13, 14) P68 (1, 11, 15) P69 (1, 8, 11) P70 (1, 14, 1) P71 (1, 1, 5) P72 (1, 7, 15)

P73 (1, 8, 13) P74 (1, 4, 6) P75 (1, 3, 3) P76 (1, 6, 9) P77 (1, 2, 3) P78 (1, 6, 3)

P79 (1, 6, 10) P80 (1, 12, 12) P81 (1, 10, 9) P82 (1, 2, 11) P83 (1, 14, 2) P84 (1, 9, 15)

P85 (1, 8, 12) P86 (1, 10, 3) P87 (1, 6, 0) P88 (0, 1, 6) P89 (1, 0, 11) P90 (1, 14, 8)

P91 (1, 15, 14) P92 (1, 11, 3) P93 (1, 6, 6) P94 (1, 3, 9) P95 (1, 2, 14) P96 (1, 11, 13)

P97 (1, 4, 1) P98 (1, 1, 12) P99 (1, 10, 1) P100 (1, 1, 1) P101 (1, 1, 9) P102 (1, 2, 10)

P103 (1, 12, 15) P104 (1, 8, 2) P105 (1, 9, 12) P106 (1, 10, 13) P107 (1, 4, 14) P108 (1, 11, 1)

P109 (1, 1, 2) P110 (1, 9, 0) P111 (0, 1, 9) P112 (1, 0, 10) P113 (1, 12, 8) P114 (1, 15, 1)
Appendix A. Points and Lines 133

Table A.19: P115 , . . . P252 of PG(2, 17)

P115 (1, 1, 6) P116 (1, 3, 11) P117 (1, 14, 16) P118 (1, 16, 11) P119 (1, 14, 11) P120 (1, 14, 0)

P121 (0, 1, 14) P122 (1, 0, 2) P123 (1, 9, 8) P124 (1, 15, 7) P125 (1, 5, 15) P126 (1, 8, 14)

P127 (1, 11, 11) P128 (1, 14, 9) P129 (1, 2, 2) P130 (1, 9, 9) P131 (1, 2, 9) P132 (1, 2, 12)

P133 (1, 10, 11) P134 (1, 14, 12) P135 (1, 10, 12) P136 (1, 10, 6) P137 (1, 3, 4) P138 (1, 13, 13)

P139 (1, 4, 9) P140 (1, 2, 16) P141 (1, 16, 6) P142 (1, 3, 5) P143 (1, 7, 12) P144 (1, 10, 10)

P145 (1, 12, 9) P146 (1, 2, 15) P147 (1, 8, 7) P148 (1, 5, 14) P149 (1, 11, 12) P150 (1, 10, 16)

P151 (1, 16, 15) P152 (1, 8, 0) P153 (0, 1, 8) P154 (1, 0, 6) P155 (1, 3, 8) P156 (1, 15, 2)

P157 (1, 9, 7) P158 (1, 5, 2) P159 (1, 9, 2) P160 (1, 9, 4) P161 (1, 13, 6) P162 (1, 3, 13)

P163 (1, 4, 3) P164 (1, 6, 15) P165 (1, 8, 5) P166 (1, 7, 13) P167 (1, 4, 2) P168 (1, 9, 10)

P169 (1, 12, 14) P170 (1, 11, 4) P171 (1, 13, 15) P172 (1, 8, 10) P173 (1, 12, 2) P174 (1, 9, 14)

P175 (1, 11, 5) P176 (1, 7, 0) P177 (0, 1, 7) P178 (1, 0, 13) P179 (1, 4, 8) P180 (1, 15, 0)

P181 (0, 1, 15) P182 (1, 0, 16) P183 (1, 16, 8) P184 (1, 15, 10) P185 (1, 12, 1) P186 (1, 1, 3)

P187 (1, 6, 14) P188 (1, 11, 6) P189 (1, 3, 7) P190 (1, 5, 6) P191 (1, 3, 6) P192 (1, 3, 0)

P193 (0, 1, 3) P194 (1, 0, 14) P195 (1, 11, 8) P196 (1, 15, 3) P197 (1, 6, 13) P198 (1, 4, 15)

P199 (1, 8, 6) P200 (1, 3, 15) P201 (1, 8, 15) P202 (1, 8, 4) P203 (1, 13, 10) P204 (1, 12, 11)

P205 (1, 14, 6) P206 (1, 3, 16) P207 (1, 16, 5) P208 (1, 7, 1) P209 (1, 1, 15) P210 (1, 8, 16)

P211 (1, 16, 0) P212 (0, 1, 16) P213 (1, 0, 7) P214 (1, 5, 8) P215 (1, 15, 15) P216 (1, 8, 9)

P217 (1, 2, 7) P218 (1, 5, 1) P219 (1, 1, 13) P220 (1, 4, 12) P221 (1, 10, 14) P222 (1, 11, 16)

P223 (1, 16, 14) P224 (1, 11, 14) P225 (1, 11, 10) P226 (1, 12, 4) P227 (1, 13, 11) P228 (1, 14, 3)

P229 (1, 6, 7) P230 (1, 5, 4) P231 (1, 13, 5) P232 (1, 7, 14) P233 (1, 11, 0) P234 (0, 1, 11)

P235 (1, 0, 5) P236 (1, 7, 8) P237 (1, 15, 11) P238 (1, 14, 14) P239 (1, 11, 9) P240 (1, 2, 13)

P241 (1, 4, 16) P242 (1, 16, 4) P243 (1, 13, 12) P244 (1, 10, 2) P245 (1, 9, 13) P246 (1, 4, 10)

P247 (1, 12, 5) P248 (1, 7, 7) P249 (1, 5, 9) P250 (1, 2, 1) P251 (1, 1, 10) P252 (1, 12, 3)
Appendix A. Points and Lines 134

Table A.20: P253 , . . . P307 of PG(2, 17)

P253 (1, 6, 12) P254 (1, 10, 0) P255 (0, 1, 10) P256 (1, 0, 3) P257 (1, 6, 8) P258 (1, 15, 13)

P259 (1, 4, 0) P260 (0, 1, 4) P261 (1, 0, 4) P262 (1, 13, 8) P263 (1, 15, 16) P264 (1, 16, 10)

P265 (1, 12, 13) P266 (1, 4, 5) P267 (1, 7, 2) P268 (1, 9, 3) P269 (1, 6, 11) P270 (1, 14, 7)

P271 (1, 5, 10) P272 (1, 12, 0) P273 (0, 1, 12) P274 (1, 0, 1) P275 (1, 1, 8) P276 (1, 15, 6)

P277 (1, 3, 2) P278 (1, 9, 1) P279 (1, 1, 0) P280 (0, 1, 1) P281 (1, 0, 9) P282 (1, 2, 8)

P283 (1, 15, 4) P284 (1, 13, 16) P285 (1, 16, 12) P286 (1, 10, 15) P287 (1, 8, 3) P288 (1, 6, 5)

P289 (1, 7, 16) P290 (1, 16, 1) P291 (1, 1, 7) P292 (1, 5, 13) P293 (1, 4, 11) P294 (1, 14, 13)

P295 (1, 4, 13) P296 (1, 4, 7) P297 (1, 5, 11) P298 (1, 14, 10) P299 (1, 12, 6) P300 (1, 3, 10)

P301 (1, 12, 10) P302 (1, 12, 16) P303 (1, 16, 13) P304 (1, 4, 4) P305 (1, 13, 9) P306 (1, 2, 0)

P307 (0, 1, 2)

Table A.21: Lines of PG(2, 17)

L1 P1 P2 P10 P16 P87 P110 P120 P152 P176 P180 P192 P211 P233 P254
P259 P272 P279 P306
..
.

L307 P307 P1 P9 P15 P86 P109 P119 P151 P175 P179 P191 P210 P232 P253
P258 P271 P278 P305
Appendix B

GAP Program

B.1 GAP Program

——————————————————————————————————-
ps := [];; r := [];; x := [];; `1 := [];; ` := [];; mm := [];;
oo := [] ;; yy := [] ;; nn := [] ;; ac := [];;
——————————————————————————————————-
q := 13;
qq := 183;
n := 3;
u1 := [1, 0, 0];;
cm := [[0, 1, 0], [0, 0, 1], [2, 0, 1]];;
——————————————————————————————————-
Find the Points Of PG(2,13)
——————————————————————————————————-
tic := 0; tt := 0;
for i in [0..qq − 1] do
p := u1 ∗ cim mod q;
if p[1] <> 0 mod q then z := p ∗ p[1]−1 mod q;
elif p[2] <> 0 mod q then z := p ∗ p[2]−1 mod q;
elif p[3] <> 0 mod q then z := p ∗ p[3]−1 mod q;

135
Appendix B. GAP Program 136

fi;
ps [i + 1] := z;
Add (ps , z);
if p[3] = 0 then zz := z;
tt := tt + 1;
`1 [tt] := i;
fi;
od;
——————————————————————————————————-
c := []; nu := []; ad := []; ar := [];
ac := [[1, 2, 3, 46, 7], [1, 2, 3, 46, 8], [1, 2, 3, 46, 18]];
for k in [1 · · · qq] do;
nu [k + 1] := k;
od;
——————————————————————————————————-
for i in [1 · · · Size(ac )] do
wq := Difference (nu , ac [i]);
c := ac [i];
for j in wq do
od; od;
——————————————————————————————————-
Find type of Secants
——————————————————————————————————-
for i in ad do
ch := [];;
for j in [1 · · · qq] do
Add(ch , [j]);
fi;
od;
if Size(ch ) = 0 then
Add(ar , i);
Appendix B. GAP Program 137

fi;
od;
Print(Size(ar ), ” n”);
——————————————————————————————————-
q1 := 6;
——————————————————————————————————-
s := List ( [1 · · · Size (ar )], i− > List(ar [i], j− > ps [j]));
w := Size(s);;
for h in [1 · · · w − 1] do
for j in [h + 1 · · · w] do
v := [];;
——————————————————————————————————-
Find the Permutations Of Arcs
——————————————————————————————————-
n := [s[j][i1 ], s[j][i2 ], s[j][i3 ], s[j][i4 ]];;
Add(v, n);
fi;
——————————————————————————————————-
Find the Projectivity
——————————————————————————————————-
for i in [1 · · · Length(v)] do
t := v[i];
a1 := t[4][1] ∗ (t[2][2] ∗ t[3][3] − t[3][2] ∗ t[2][3]);
a2 := t[2][1] ∗ (t[4][2] ∗ t[3][3] − t[3][2] ∗ t[4][3]);
a3 := t[3][1] ∗ (t[4][2] ∗ t[2][3] − t[2][2] ∗ t[4][3]);
b1 := t[1][1] ∗ (t[4][2] ∗ t[3][3] − t[3][2] ∗ t[4][3]);
b2 := t[4][1] ∗ (t[1][2] ∗ t[3][3] − t[3][2] ∗ t[1][3]);
b3 := t[3][1] ∗ (t[1][2] ∗ t[4][3] − t[4][2] ∗ t[1][3]);
c1 := t[1][1] ∗ (t[2][2] ∗ t[4][3] − t[4][2] ∗ t[2][3]);
c2 := t[2][1] ∗ (t[1][2] ∗ t[4][3] − t[4][2] ∗ t[1][3]);
c3 := t[4][1] ∗ (t[1][2] ∗ t[2][3] − t[2][2] ∗ t[1][3]);
Appendix B. GAP Program 138

a := (a1 − a2 + a3 )mod q;
b := (b1 − b2 + b3 )mod q;
c := (c1 − c2 + c3 )mod q;
T := [[a ∗ t[1][1]mod q, a ∗ t[1][2]mod q, a ∗ t[1][3]mod q],
[b ∗ t[2][1]mod q, b ∗ t[2][2]mod q, b ∗ t[2][3]mod q],
[c ∗ t[3][1]mod q, c ∗ t[3][2]mod q, c ∗ t[3][3]mod q]];
——————————————————————————————————-
Find the Equivalents Arcs
——————————————————————————————————-
for k in[1 · · · q1 ] do
r[k] := (s[h][k] ∗ T )mod q;
od;
for k in[1 · · · q1 ] do
if r[k][1] <> 0 mod qthen x[k] := r[k] ∗ r[k][1]−1 mod q;
elif r[k][2] <> 0 mod qthen x[k] := r[k] ∗ r[k][2]−1 mod q;
elif r[k][3] <> 0 mod qthen x[k] := r[k] ∗ r[k][3]−1 mod q;
fi;
od;
if Set(x) = Set(s[j]) mod q then
for mi in[1 · · · q1 ] do
od;
if pos = fail then
Add (oo, j);
fi; fi;
od; od; od;
——————————————————————————————————-
Find the Distinct Arcs
——————————————————————————————————-
ry := [] ;;ro := 0;
for i in [1 · · · w] do
yy[i] := i;
Appendix B. GAP Program 139

od;
for i in xx do
ro := ro + 1;
ry [ro ] := ar [i] ;
od;
Print( ”The ”, Size(ry ), ” distinct arcs-are: n”, ry , ” n”);
Appendix C

Fortran Program

C.1 Fortran Program

C.1.1 Algorithm Two

program Main implicit none integer G1, G2, G3, ci, cj, pi, pj, Li, Lj, adi, adj, ti, tj, tk, tr
integer sor, sor1, sor2, ui, uj, uk, xi, xj, O1, O2, i1, i2, i3, i4, uuii, uujj
integer Det, Det1, Det2, Det3, M 1, LO, N 1, di, dj, dk, ni, GH, complete, uukk
integer startl oop, tip, tin, comp, major, minor, dist, ii, jj, mm, oo, HH
integer loop1, loop2, clas, C, GF, B1, xr, i, j, IV, Lk, LL, s, ads, ts, tt
integer, dimension (3) :: P 1, AB
integer, dimension (3, 3) :: CM, A, A1, A2, A3, projective
integer, dimension (4) :: IN V
integer, dimension (6) :: L
integer, dimension (31, 6) :: LIN E
integer, dimension (31, 3) :: P OIN T, poin
integer, dimension (4) :: P ERM U T AT ION
integer, dimension (31) :: T d
integer, dimension (5) :: T RR, T Y P, SEC
integer, allocatable :: AD(:, :)
integer, allocatable :: ADD(:, :)

140
Appendix C. Fortran Program 141

integer, allocatable :: ADD1(:, :)

integer, allocatable :: ADDEDS(:, :)
integer, allocatable :: ADDED(:, :)
integer, allocatable :: N EW (:, :)
integer, allocatable :: F AT EN (:, :)
integer, allocatable :: T Y P ES(:)
integer, allocatable :: DistinctT ypes(:, :)
integer, allocatable :: M U LT (:)
integer, allocatable :: DIST IN CT (:, :)
open (unit=1 , file=’COMPLETE-ARCS.fort’)
open (unit=2 , file=’NEW.fort’)
open (unit=3 , file=’Report.fort’)
IV = 0
do i = 1, q − 1
do j = 1, q − 1
if (mod (i ∗ j, q).eq.1) then
IV = IV + 1
IN V (IV ) = j
end if
end do
end do
P 1 = (/1, 0, 0/)
poin= 0
poin (1, :) = P 1
POINT (1, :) = poin(1, :)
do pi = 2, qq
do pj = 1, 3
poin(pi, :) = mod (poin(pi, :) + P 1(pj) ∗ CM (pj, :), q)
end do
if (poin(pi, 1).ne.0) then
poin = mod (IN V (poin(pi, 1)) ∗ poin, q)
Appendix C. Fortran Program 142

end if
P 1 = poin(pi, :); P OIN T (pi, :) = poin(pi, :)
end
LL = 0
do Lk = 1, qq
if (P OIN T (Lk, 3).eq.0) then
L(LL) = Lk
end if
LINE (1, :) = L
end do
do Li = 2, qq
LINE (Li, :) = L + 1
do Lj = 1, q + 1
if (LIN E(Li, Lj).eq.qq + 1) then
LIN E(Li, Lj) = 1
end if
end do
L = LIN E(Li, :)
end do
G1 = 2; G2 = 6
allocate(AD(G1, G2))
AD(1, 1) = 1; AD(1, 2) = 2; AD(1, 3) = 3; AD(1, 4) = 5; AD(1, 5) = 15; AD(1, 6) = 21
AD(2, 1) = 1; AD(2, 2) = 2; AD(2, 3) = 3; AD(2, 4) = 5; AD(2, 5) = 15; AD(2, 6) = 25
do startl oop = 1, 10
allocate(ADD1(G3, G2 + 6))
allocate(ADD(G3, G2 + 1))
tin = 0; complete = 0
do adi = 1, G1; comp = 0; HH = 0
do adj = 1, 31
if(all(AD(adi, :).ne.adj))then;
HH = HH + 1
Appendix C. Fortran Program 143

ADD(HH, :) = (/AD(adi, :), adj/)

end if
end do
do ads = 1, HH; dott = 0, 5; T RR(tt) = 0; end do
do ti = 1, 31; tr = 0
do tj = 1, G2 + 1
do tk = 1, 6
end do
end do
T d(ti) = tr
end do
do ts = 1, 31
do tt = 0, 5
end do
end do
s=0
do i = 4, 0, −1
T Y P (i) = T RR(i)
s = s + T Y P (i)
end do
if(s.eq.31) then;
do sor1 = 1, G2
do sor2 = sor1 + 1, G2 + 1
if(ADD(ads, sor1).gt.ADD(ads, sor2)) then
sor = ADD(ads, sor2)
ADD(ads, sor2) = ADD(ads, sor1)
ADD(ads, sor1) = sor
end if
end do
end do
comp = comp + 1
Appendix C. Fortran Program 144

tin = tin + 1
ADD1(tin, :) = (/ADD(ads, :), T Y P/)
end if
end do
if(comp.eq.0) then
complete = complete + 1
write(1, ∗)AD(adi, :),0 Is Complete(0 , G2,0 ,0 , 4,0 ) − arc0
end if
end do
write (1, ∗)’—————————————————–’
write (1, ∗)0 Total number of complete(0 , G2,0 ,0 , 4,0 )is :0 , complete
write (1, ∗)’—————————————————–’
write (1, ∗)0 COMPLETE(0 , G2,0 ,0 , 4,0 ) − arcs is0 ,0 [0 , complete,0 ]0
Deallocate (AD)
Deallocate (ADD)
allocate (ADDED(tin, G2 + 1))
dist = 0
do uuii = 1, tin
uukk = 0
do uujj = uuii + 1, tin
uukk = uukk + 1
end if
end do
if (uukk.eq.0) then
dist = dist + 1
ADDED(dist, :) = ADD1(uuii, 1 : G2 + 1)
write (2, ∗)ADDED(dist, :)
end if
end do
write(3, ∗)dist,0 (0 , G2 + 1,0 ,0 , 4,0 ) − arcs0
deallocate (ADD1)
Appendix C. Fortran Program 145

G1 = dist; G2 = G2 + 1
allocate(AD(G1, G2))
do i = 1, dist
AD(i, :) = ADDED(i, :)
end do
Deallocate (ADDED)
end do
close (1)
close (3)
close (4)
close (6)
stop
end
Appendix C. Fortran Program 146

C.1.2 Algorithm Five

program Main
implicit none
integer, parameter :: q = 13, qq = q ∗ q + q + 1, N = 2, N U M = q − 3
integer G1, G2, G3, ci, cj, pi, pj, Li, Lj, adi, adj, ti, tj, tk, tr, ts, tt
integer sor, sor1, sor2, ui, uj, uk, xi, xj, O1, O2, i1, i2, i3, i4, uuii, uujj
integer Det, Det1, Det2, Det3, M 1, LO, N 1, di, dj, dk, ni, GH, complete, uukk
integer start − loop, tip, tin, comp, major, minor, dist, ii, jj, mm, oo, HH
integer loop1, loop2, clas, C, GF, B1, xr, i, j, IV, Lk, LL, s, f f
integer, dimension (3) :: P 1, AB
integer, dimension (3, 3) :: CM, A, A1, A2, A3, projective
integer, dimension (q − 1) :: IN V
integer, dimension (q + 1) :: L
integer, dimension (qq, q + 1) :: LIN E
integer, dimension (qq, 3) :: P OIN T, poin
integer, dimension (4) :: P ERM U T AT ION
integer, dimension (qq) :: T d
integer, dimension (N + 1) :: T RR, T Y P, SEC
integer, allocatable :: AD(:, :)
integer, allocatable :: ADD(:)
integer, allocatable :: ADD1(:, :)
integer, allocatable :: ADDEDS(:, :)
integer, allocatable :: ADDED(:, :)
integer, allocatable :: N EW (:, :)
integer, allocatable :: F AT EN (:, :)
integer, allocatable :: T Y P ES(:)
integer, allocatable :: Distinct − T ypes(:, :)
integer, allocatable :: M U LT (:)
integer, allocatable :: DIST IN CT (:, :)
open (unit=0 , file=’ARCS.fort’)
open (unit=1 , file=’COMPLETE-ARCS.fort’)
Appendix C. Fortran Program 147

open (unit=2 , file=’TYPES.fort’)

open (unit=3 , file=’EQUIVALENT.fort’)
open (unit=4 , file=’RESULTS.fort’)
open (unit=5 , file=’INFORMATION.fort’)
open (unit=8 , file=’MAXIMUM.fort’)
IV = 0
do i = 1, q − 1
do j = 1, q − 1
if (mod (i ∗ j, q).eq.1) then
IV = IV + 1
IN V (IV ) = j
end if
end do
end do
P 1 = (/1, 0, 0/)
poin= 0
poin (1, :) = P 1
POINT (1, :) = poin(1, :)
do pi = 2, qq
do pj = 1, 3
poin(pi, :) = mod (poin(pi, :) + P 1(pj) ∗ CM (pj, :), q)
end do
if (poin(pi, 1).ne.0) then
poin = mod (IN V (poin(pi, 1)) ∗ poin, q)
end if
P 1 = poin(pi, :); P OIN T (pi, :) = poin(pi, :)
end
LL = 0
do Lk = 1, qq
if (P OIN T (Lk, 3).eq.0) then
L(LL) = Lk
Appendix C. Fortran Program 148

end if
LINE (1, :) = L
end do
do Li = 2, qq
LINE (Li, :) = L + 1
do Lj = 1, q + 1
LIN E(Li, Lj) = 1
end if
end do
L = LIN E(Li, :)
end do
G1 = 1
G2 = 4
allocate (AD(G1, G2))
do 1ci = 1, G1
1 read (0, ∗)AD(ci, 1), AD(ci, 2), AD(ci, 3), AD(ci, 4)
DO start − loop = 1, N U M
allocate (ADD(G2 + 1))
allocate (ADD1(G3, G2 + 2 + N ))
tin = 0
complete = 0
HH = 1
do adi = 1, G1
comp= 0
do adj = 1, qq
B1 = 0
if(all(AD(adi, :).ne.adj)) then
ADD(:) = (/AD(adi, :), adj/)
do tt = 0, N + 1
T RR(tt) = 0
end do
Appendix C. Fortran Program 149

do ti = 1, qq
tr = 0
do tj = 1, G2 + 1
do tk = 1, q + 1
if (ADD(tj).eq.LIN E(ti, tk)) then
tr = tr + 1
end if
end do
end do
T d(ti) = tr
end do
do ts = 1, qq
do tt = 0, N + 1
if (T d(ts).eq.tt) then
T RR(tt) = T RR(tt) + 1
end if
end do
end do
ff = 0
s=0
do i = N, 0, −1
ff = ff + 1
T Y P (f f ) = T RR(i)
s = s + T Y P (f f )
end do
if (s.eq.qq) then
do sor1 = 1, G2
do sor2 = sor1 + 1, G2 + 1
if (ADD(sor1).gt.ADD(sor2)) then
sor = ADD(sor2)
ADD (sor2) = ADD(sor1)
Appendix C. Fortran Program 150

ADD (sor1) = sor

end if
end do
end do
comp = comp + 1
do xr = 1, tin
if (all(ADD1(xr, 1 : G2 + 1).eq.ADD(:))) then
B1 = 1
end if
end do
if (B1.eq.0) then
tin = tin + 1
ADD1(tin, :) = (/ADD(:), T Y P/)
end if
end if
end if
end do
if (comp.eq.0) then
complete = complete + 1
write (1, ∗)AD(adi, :),0 Is Complete (0 , G2,0 ,0 , N,0 ) − arc0
end if
end do
write (1, ∗) ’—————————————————–’
write (1, ∗)0 Total number of complete (0 , G2,0 ,0 , N,0 )is :0 , complete
write (1, ∗) ’—————————————————–’
write (1, ∗) ’ COMPLETE (0 , G2,0 ,0 , N,0 )-arcs is 0 ,0 [0 , complete,0 ]0
Deallocate (AD)
Deallocate (ADD)
allocate (Distinct − T ypes(tin, N + 1))
dist = 0
do uuii = 1, tin
Appendix C. Fortran Program 151

uukk = 0
do uujj = uuii + 1, tin
if (all(ADD1(uujj, G2 + 2 : G2 + 2 + N ).eq.ADD1(uuii, G2 + 2 : G2 + 2 + N )))then
uukk = uukk + 1
end if
end do
if (uukk.eq.0) then
dist = dist + 1
Distinct − T ypes(dist, :) = ADD1(uuii, G2 + 2 : G2 + 2 + N )
end if
end do
allocate (T Y P ES(dist))
allocate (ADDED(tin, G2 + 1))
oo = 0
do ii = 1, dist
mm = 0
do jj = 1, tin
if (all(ADD1(jj, G2 + 2 : G2 + 2 + N ).eq.Distinct − T ypes(ii, :))) then
oo = oo + 1
ADDED(oo, :) = ADD1(jj, 1 : G2 + 1)
mm = mm + 1
end if
end do
T Y P ES(ii) = mm
write (2, ∗)T Y P ES(ii),0 (0 , G2 + 1,0 ,0 , N,0 ) − arcof type0 , Distinct − T ypes(ii, :)
end do
if (tin.eq.1) then
N1 = 1
goto 44
end if
deallocate (ADD1)
Appendix C. Fortran Program 152

deallocate (Distinct − T ypes)

allocate (N EW (tin, G2 + 1))
allocate (F AT EN (G2 + 1, 3))
allocate (M U LT (G2 + 1))
loop1 = 1
loop2 = T Y P ES(1)
GH = 0
do clas = 1, dist
GF = 0
C = 1 + loop2 − loop1
Do major = loop1, loop2
O1 = 0
do xi = 1, GH
if (all(N EW (xi, :).eq.ADDED(major, :))) then
O1 = O1 + 1
end if
end do
if (O1.eq.0) then
Do minor = major + 1, loop2
O2 = 0
do xj = 1, GH
if (all(N EW (xj, :).eq.ADDED(minor, :))) then
O2 = O2 + 1
end if
end do
if (O2.eq.0) then
Do i1 = 1, G2 + 1
Do i2 = 1, G2 + 1
Do i3 = 1, G2 + 1
Do i4 = 1, G2 + 1
if (i1.ne.i2.And.i1.ne.i3.And.i1.ne.i4.And.i2.ne.i3.And.i2.ne.i4.And.i3.ne.i4) then
Appendix C. Fortran Program 153

P ERM U T AT ION (:) = (/ADDED(minor, i1), ADDED(minor, i2),

ADDED(minor, i3), ADDED(minor, i4)/)
A(:, 1) = P OIN T (P ERM U T AT ION (1), :)
A(:, 2) = P OIN T (P ERM U T AT ION (2), :)
A(:, 3) = P OIN T (P ERM U T AT ION (3), :)
AB = P OIN T (P ERM U T AT ION (4), :)
A1 = A
A2 = A
A3 = A
A1(1, 1) = AB(1)
A1(2, 1) = AB(2)
A1(3, 1) = AB(3)
A2(1, 2) = AB(1)
A2(2, 2) = AB(2)
A2(3, 2) = AB(3)
A3(1, 3) = AB(1)
A3(2, 3) = AB(2)
A3(3, 3) = AB(3)
Det = mod(mod((A(1, 1) ∗ A(2, 2) ∗ A(3, 3) + A(1, 2) ∗ A(2, 3) ∗ A(3, 1)+
A(1, 3) ∗ A(2, 1) ∗ A(3, 2) − A(1, 3) ∗ A(2, 2) ∗ A(3, 1) − A(1, 1) ∗ A(2, 3) ∗ A(3, 2)−
A(1, 2) ∗ A(2, 1) ∗ A(3, 3)), q) + q, q)
if (Det.ne.0) then
Det1 = mod (mod ((A1(1, 1) ∗ A1(2, 2) ∗ A1(3, 3)+
A1(1, 2) ∗ A1(2, 3) ∗ A1(3, 1) + A1(1, 3) ∗ A1(2, 1) ∗ A1(3, 2)−
A1(1, 3) ∗ A1(2, 2) ∗ A1(3, 1) − A1(1, 1) ∗ A1(2, 3) ∗ A1(3, 2)−
A1(1, 2) ∗ A1(2, 1) ∗ A1(3, 3)), q) + q, q)
if (Det1.ne.0) then
Det2 =mod (mod((A2(1, 1) ∗ A2(2, 2) ∗ A2(3, 3)+
A2(1, 2) ∗ A2(2, 3) ∗ A2(3, 1) + A2(1, 3) ∗ A2(2, 1) ∗ A2(3, 2)−
A2(1, 3) ∗ A2(2, 2) ∗ A2(3, 1) − A2(1, 1) ∗ A2(2, 3) ∗ A2(3, 2)−
A2(1, 2) ∗ A2(2, 1) ∗ A2(3, 3)), q) + q, q)
Appendix C. Fortran Program 154

if (Det2.ne.0) then
Det3 =mod (mod ((A3(1, 1) ∗ A3(2, 2) ∗ A3(3, 3)+
A3(1, 2) ∗ A3(2, 3) ∗ A3(3, 1) + A3(1, 3) ∗ A3(2, 1) ∗ A3(3, 2)−
A3(1, 3) ∗ A3(2, 2) ∗ A3(3, 1) − A3(1, 1) ∗ A3(2, 3) ∗ A3(3, 2)−
A3(1, 2) ∗ A3(2, 1) ∗ A3(3, 3)), q) + q, q)
if (Det3.ne.0) then
P ROJECT IV E(1, :) = mod(Det1 ∗ A(:, 1), q)
P ROJECT IV E(2, :) = mod(Det2 ∗ A(:, 2), q)
P ROJECT IV E(3, :) = mod(Det3 ∗ A(:, 3), q)
do M 1 = 1, G2 + 1
F AT EN (M 1, 1) =mod (P OIN T (ADDED(major, M 1), 1) ∗ P ROJECT IV E(1, 1)+
P OIN T (ADDED(major, M 1), 2) ∗ P ROJECT IV E(2, 1)+
P OIN T (ADDED(major, M 1), 3) ∗ P ROJECT IV E(3, 1), q)
F AT EN (M 1, 2) = mod(P OIN T (ADDED(major, M 1), 1) ∗ P ROJECT IV E(1, 2)+
P OIN T (ADDED(major, M 1), 2) ∗ P ROJECT IV E(2, 2)+
P OIN T (ADDED(major, M 1), 3) ∗ P ROJECT IV E(3, 2), q)
F AT EN (M 1, 3) = mod(P OIN T (ADDED(major, M 1), 1) ∗ P ROJECT IV E(1, 3)+
P OIN T (ADDED(major, M 1), 2) ∗ P ROJECT IV E(2, 3)+
P OIN T (ADDED(major, M 1), 3) ∗ P ROJECT IV E(3, 3), q)
if (F AT EN (M 1, 1).ne.0) then
F AT EN =mod (IN V (F AT EN (M 1, 1)) ∗ F AT EN, q)
elseif (F AT EN (M 1, 2).ne.0) then
F AT EN =mod (IN V (F AT EN (M 1, 2)) ∗ F AT EN, q)
elseif (F AT EN (M 1, 3).ne.0) then
F AT EN =mod (IN V (F AT EN (M 1, 3)) ∗ F AT EN, q)
end if
doLO = 1, qq
M U LT (M 1) = LO
exit
end if
end do
Appendix C. Fortran Program 155

end do
do sor1 = 1, G2
do sor2 = sor1 + 1, G2 + 1
if (M U LT (sor1).gt.M U LT (sor2))then
sor = M U LT (sor2)
M U LT (sor2) = M U LT (sor1)
M U LT (sor1) = sor
end if
end do
end do
if (all(M U LT (:).eq.ADDED(minor, :)))then
GH = GH + 1
N EW (GH, :) = M U LT (:)
end if
end if
end if
end if
end if
end if
33 end do
end do
end do
end do
end if
99 end do
end if
GF = GF + 1
print*, GF,0 of 0 , C
100 end do
end do
allocate (DIST IN CT (tin, G2 + 1))
Appendix C. Fortran Program 156

N1 = 0
do di = 1, tin
dk = 0
do dj = 1, GH
if (all(N EW (dj, :).eq.ADDED(di, :))) then
dk = dk + 1
end if
end do
if (dk.eq.0) then
N1 = N1 + 1
DISTINCT (N 1, :) = ADDED(di, :)
write (4, ∗)DIST IN CT (N 1, :)
end if
end do
write (4, ∗)N 1
write (1, ∗) ’Total number of complete (0 , G2,0 ,0 , N,0 ) is:0 , complete
write (3, ∗) ’number of eq.(0 , G2 + 1,0 ,0 , N,0 )-arcs is 0 , GH,0 of total0 , tin
write (5, ∗) ’Total (0 , G2 + 1,0 ,0 , N,0 )-arcs is0 ,0 [0 , tin,0 ]0
write (5, ∗) ’Equivalent(0 , G2 + 1,0 ,0 , N,0 )-arcs is0 ,0 [0 , GH,0 ]0
write (5, ∗) ’Distinc (0 , G2 + 1,0 ,0 , N,0 )-arcs is0 ,0 [0 , N 1,0 ]0
G1 = N 1
G2 = G2 + 1
allocate (AD(G1, G2))
doni = 1, G1
AD(ni, :) = DIST IN CT (ni, :)
end do
Deallocate (N EW )
Deallocate (DIST IN CT )
Deallocate (ADDED)
Deallocate (F AT EN )
Deallocate (M U LT )
Appendix C. Fortran Program 157

Deallocate (T Y P ES)
END Do
goto 55
44 write (8, ∗) ’MAXIMUM (0 , G2 + 1,0 ,0 , N,0 )-arcs is 0 , ADDED(1, :)
write (5, ∗) ’MAXIMUM (0 , G2 + 1,0 ,0 , N,0 )-arcs is 0 ,0 , 1
write (1, ∗) ’Total number of complete (0 , G2 + 1,0 ,0 , N,0 ) is:0 , 1
write (1, ∗)ADDED(1, :),0 Is Complete (0 , G2 + 1,0 ,0 , N,0 )-arc’
write (4, ∗)ADDED(1, :)
write(4, ∗)1
55 stop
end
Bibliography

[1] T. L. Alderson and A. A. Bruen, Maximal AMDS code, Appl. Algebra Engrg.
Comm. Comput. 19 (2008), 87–98.

[2] A. Ali, Classification of arcs in the Galois plane of order thirteen. PhD thesis,
University of Sussex, 1993.

[3] S. Ball, On sets of points in finite planes. PhD thesis, University of Sussex, 1994.

[4] S. Ball, Multiple blocking sets and arcs in finite planes, J. London Math. Soc. 54
(1996), 581–593.

[5] S. Ball, On the size of a triple blocking set in PG(2, q), European J. Combin. 17
(1996), 427–435.

[6] S. Ball, A. Blokhuis, On the size of a double blocking set in PG(2, q), Finite Fields
Appl. 2 (1996), 125–137.

[7] S. Ball, Three-dimensional linear codes, Online table, (2014), https://mat-

web.upc.edu/people/simeon.michael.ball/codebounds.html.

[8] D. Bartoli, S. Marcugini and F. Pambianco, The non-existence of some NMDS

codes and the extremal size of complete (n, 3)-arcs in PG(2, 16), Des. Codes Cryp-
togr. 72 (2014), 129–134.

[9] J. Bierbrauer, (k, n)-arcs of maximal size in the plane of order eight. PhD thesis,
Unpublished manuscript, 1988.

[10] A. Blokhuis, On multiple nuclei and a conjecture of Lurelli and Sce, Bull. Belg.
Math. Soc. Simon Steven 3 (1994), 349–353.
158
Bibliography

[11] A. Blokhuis, L. Storme and T. Szőnyi, Multiple blocking sets and Baer-subplanes,
unpublished manuscipt (1995), 351–356.

[12] A. E. Brouwer and H. A. Wilbrink, Blocking set in translation planes, J. Geom.

19 (1982), 200-200.

[13] A. A. Bruen, Baer subplanes and blocking sets, Bull. Amer. Math. Soc. 76 (1970),
342–344.

[14] A. A. Bruen, Blocking sets in finite projective planes, SIAM J. Appl.Math. 21

(1971), 380–392.

[15] A. A. Bruen and J. A. Thas, Blocking sets, Geom. Dedicata 6 (1977), 193–203.

[16] A. A. Bruen and R. Silverman, Arcs and blocking sets II, European J.Combin. 8
(1987), 351–356.

[17] A. A. Bruen, Polynomial multiplicities over finite and intersection sets, J. Combin.
Theory Ser. A 60 (1992), 19–33.

[18] P. Cameron, Introduction to algebra, Oxford University Press, New York, 1998,
second edition.

[19] R. Casse, Projective geometries over finite fields, Oxford University Press, New
York, 2006, second edition.

[20] R. Chandra, L. Dagum, D. Kohr, J. McDonald and R. Menon, Parallel Program-

ming in OpenMP, Morgan Kaufmann, 2001.

[21] J. M. Chao and H. Kaneta, Classical arcs in PG(r, q) for 11 ≤ q ≤ 19, Discrete
Math. 174 (1997), 87–94.

[22] J. M. Chao and H. Kaneta, A complete 24-arc in PG(2, 29) with automorphism
group PSL(2, 7), Rend. Math. Appl. 16 (1996), 537–544.

[23] J. M. Chao and H. Kaneta, Classical arcs in PG(r, q) for 23 ≤ q ≤ 29, Discrete
Math. 226 (2001), 377–385.
Bibliography

[24] C. Churchman, R. Ackoff and E. Arnoff, Introduction to operations research, New

York, John Wiley and Sons, Inc., 1957, first edition.

[25] J. Ciechanowicz, The non-existence of (37, 6)-arcs in PG(2, 7) and some other
(k, n)-arcs in Galois planes. PhD thesis, University of London, 1980.

[26] K. Coolsaet and H. Sticker, A full classification of the complete k-arcs in PG(2, 23)
and PG(2, 25), J. Combin. Des. 17 (2009), 459–477.

[27] K. Coolsaet and H. Sticker, The complete k-arcs of PG(2, 27) and PG(2, 29), J.
Combin. Des. 19 (2011), 111–130.

[28] K. Coolsaet and H. Sticker, The complete (k, 3)-arcs of PG(2, q), q ≤ 13, J.
Combin. Des. 20 (2012), 89–111.

[29] K. Coolsaet, The complete arcs of PG(2, 31), J. Combin. Des. 23 (2015), 522–533.

[30] R. Daskalov, On the existence and the non-existence of some (k, r)-arcs in
PG(2, 17), Ninth International Workshop on Algebraic Combinatorial Coding The-
ory (19-25 June, 2004), 95–100.

[31] S. Dodurekov and I. Landger, Near-MDS code, J. Geom. 54 (1995), 30–43.

[32] B. Fleischmann, R. Lasch, U. Derigs, W. Domschke and U. Rieder, Operations re-

search proceedings 2000, Springer-Verlag Berlin Heidelberg, Germany, 2000, eight
edition.

[33] W. Gropp, E. Lusk and A. Skjellum, Portable parallel programming with the
message-passing interface, The MIT press, 1994.

[34] R. Hill, A first course in coding theory, Clarendon Press, Oxford, 1986, first edition.

[35] R. Hill and J. R. M. Mason, On (k, n)-arcs and the falsity of the Lunelli-Sce conjec-
ture, Finite Geometries and Designs (Proc. Conf., Chelwood Galo), 49, Cambridge
University (1980), 153–168.

[36] J. W. P. Hirschfeld, Projective geometries over finite fields, Clarendon Press, Ox-
ford, 1979, first edition.
Bibliography

[37] J. W. P. Hirschfeld, Projective geometries over finite fields, Oxford University

Press, 1998, second edition.

[38] J. W. P. Hirschfeld and L. Storme, The packing problem in statistics, coding theory
and finite projective spaces, J. Statist. Plann. Inference 72 (1998), 355–380.

[39] J. W. P. Hirschfeld and E. V. D. Pichanick, Bounds for arcs of arbitrary degree in

finite Desargusian planes, J. Combin. Des. 24 (2016), 184–196.

[40] J. Humphreys, A Course in Group Theory, Oxford University Press, New York,
1996, second edition.

[41] G. Kéri, Types of superregular matrices and the number of n-arcs and complete
n-arcs in PG(r, q), J. Combin. Des. 14 (2006), 363–390.

[42] L. Lunelli and M. Sce, Considerazione arithmetiche e risultati sperimentali sui

{K; n}q -archi, Ist. Lombardo Accad. Sci. Lett. Rend. A 98 (1964), 3–52.

[43] S. Marcugini, A. Milani and F. Pambianco, A computer based classification of the

(n, 3)-arcs in PG(2, 8), Rapporto Tecnico (1998).

[44] S. Marcugini, A. Milani and F. Pambianco, Maximal (n, 3)-arcs in PG(2, 11),
Discrete Math. 208 (1999), 421–426.

[45] S. Marcugini, A. Milani and F. Pambianco, Classification of the [n, 3, n−3]q NMDS
codes over GF(7), GF(8) and GF(9), Ars Combin. 61 (2001), 263–269.

[46] S. Marcugini, A. Milani and F. Pambianco, Classification of the (n, 3)-arcs in

PG(2, 7), J. Geom. 80 (2004), 179–184.

[47] S. Marcugini, A. Milani and F. Pambianco, Maximal (n, 3)-arcs in PG(2, 13),
Discrete Math. 294 (2005), 139–145.

[48] S. Marcugini, A. Milani and F. Pambianco, Minimal complete arcs in PG(2, q),
q ≤ 32, J. Combin. Math. Combin. Comput. (2010), 217–222.

[49] J. R. M. Mason, On (k, n)-arcs and (k, n)-caps in finite projective geometries. PhD
thesis, University of Salford, 1981.
Bibliography

[50] J. R. M. Monios and R. Bergqvist, Inter modal freight transport and logistics, CRC
Press, 2017.

[51] J. W. Di Paola, On minimum blocking coalitions in small projective plane games,

SIAM J. Appl.Math. 17 (1969), 378–392.

[52] J. Rajgopal, Principles and applications of operations research, Maynard’s Indus-

trial Engineering Handbook 1 (2004), 11–27.

[53] M. Richardson, On finite projective games, Proc. Amer. Math. Soc. 17 (1956),
458–465.

[54] A. Sadeh, The classification of k-arcs and cubic surfaces with Twenty Seven Lines
over the Field of Eleven Elements. PhD thesis, University of Sussex, 1984.

[55] J. E. Savage, Models of Computation, Addison Wesley, 1998.

[56] B. Segre, Le geometrie di Galois, Ann. Math. Pura Appl. 48 (1959), 1–96.

[57] J. Singer, A theorem in finite projective geometry and some applications to number
theory, Trans. Amer. Math. Soc. 43 (1938), 377–385.

[58] H. Taha, Operations Research, an Introduction, Pearson International, 2007,

eighth edition.

[59] www.rcs.manchester.ac.uk/services/computational/parallel.

[60] A. Yasin, Cubic Arcs in the Projective Plane of Order Eight. PhD thesis, University
of Sussex, 1986.