KNAPSACK PROBLEM

SUBMIT TED BY : DEEPSIKHA(51810090)

S U N N Y KU M A R JA I S WA L
(51810091)
 CONTENTS :
1. Understanding the problem
2. Approaches to Solve.
3. Their Procedures
4. Examples
5. Algorithms

NIT, Kurukshetra 2
 KNAPSACK PROBLEM
C o n s i d e r, y o u a r e g i v e n a c o n t a i n e r w i t h l i m i t e d c a p a c i t y ‘ W ’ .
a n d s o m e i t e m s I 0, I 1, . . . . I n w i t h v a l u e s v 0, v 1, … . v n a n d w i t h
t h e i r w e i g h t s W 0 , W 1 , … . W n r e s p e c t i v e l y.

GOAL :- We have to put each item in knapsack so that the total weight
is less than or equal to ‘W’ and total value of items is as
large as possible.

“ THIS IS AN OPTIMIZATION PROBLEM”

NIT, Kurukshetra 3
 Each item has 2 attributes : -
• Va l u e : i . e . , v i f o r i t e m I i.

• W e i g h t : i . e . , W i f o r i t e m I i.

Ty p e s o f K n a p s a c k p r o b l e m s : -
• Fractional Knapsack problem :- Can take fraction of an item.

(Using Greedy method).

• 0-1 Knapsack problem :- Either can take or leave an item.

(Using Dynamic programming)

No distribution of a single item.

NIT, Kurukshetra 4
 (FRACTIONAL KNAPSACK PROBLEM)
In this case,items can be broken into smaller pieces:
Given problem :- Consider Capacity of knapsack ‘W’ = 60 kgs.

Items Weight(kg.) Value

1 40 280
2 10 100
3 20 120
4 24 120

STEP :- 1. C a l c u l a t e D e n s i t y ( v a l u e / w e i g h t ) o f e a c h i t e m :

D e n s i t y = Va l u e ( v i)
W e i g h t ( W i)

and place them in Descending order According to Density of items;

NIT, Kurukshetra 5
 (FRACTIONAL KNAPSACK PROBLEM)
After Sorting Place the items in different table :

Items Weight(kg.) Value Density

2 10 100 10
REMEMBER
1 40 280 7
‘W’=60
3 20 120 6
4 24 120 5

STEP:-2. Now,We will pick items such that our value is Max. and total Weight is
less than Capacity of knapsack(W=60).

Condition :- i s _ w e i g h t [ I i] + t o t a l w e i g h t < = W (initially total weight=0),

if yes
take whole item
else
take fractional part of item.

NIT, Kurukshetra 6
 (FRACTIONAL KNAPSACK PROBLEM)
STEP :-3. Place items in new table according to condition;

Items Weight(kg. Value Density Total Total

) weight Value
2 10 100 10 10+0<=60 100 REMEMBER
‘W’=60
1 40 280 7 40+10<=60 380
3 20 120 6 50+20*1/2<=60 440
4 24 120 5 - -
i s _ w e i g h t [ I 2] + 0 < = 6 0 i s _ w e i g h t [ I 1] + 1 0 < = 6 0 i s _ w e i g h t [ I 3] + 5 0 < = 6 0 So,as total weight
10 + 0 <= 60 40 + 10 <= 60 20 + 50 <= 60 inside knapsack is
if yes if yes if yes r e a c h e d i t s c a p a c i t y,
take whole item take whole item …………… Hence,
F i l l v a l u e o f I 2. F i l l v a l u e o f I 1. else
20*1/2+50<=60 Max. value = 440.
To t a l w e i g h t = 1 0 To t a l w e i g h t = 5 0
To t a l v a l u e = 1 0 0 . To t a l v a l u e = 3 8 0 . To t a l w e i g h t = 6 0 (optimized solution)
To t a l v a l u e = 4 4 0 .

NIT, Kurukshetra 7
 (FRACTIONAL KNAPSACK ALGORITHM)

W Time
complexity:
O(nlog(n))

NIT, Kurukshetra 8
 0-1 KNAPSACK PROBLEM
In 0-1 Knapsack, items cannot be broken which means the thief should take the
item as a whole or should leave it. This is reason behind calling it as
“0-1 Knapsack”.
Hence, in case of 0-1 Knapsack, the value of xi can be either 0 or 1,
where other constraints remain the same.

xi : amount of object is selected, where 0< xi <1.

Satisfying condition :-
∑ x iW i < C a p a c i t y o f K n a p s a c k .

Objective :-
∑ x iV i s h o u l d b e o p t i m i z e d . ( m a x i m u m )

“ T H I S I S S O LV E D B Y D Y N A M I C P R O G R A M M I N G . ”

NIT, Kurukshetra 9
 (0-1 KNAPSACK PROBLEM)
In this case,items can’t be broken into smaller pieces:
Given problem :- Consider Capacity of knapsack ‘W’ = 5 kgs.

Items Weight(kg.) Value

1 3 100
2 2 20
3 4 60
4 1 40

Approach :- Compute the solutions to the sub-problems once and store the
solutions in a table, So that they can be reused repeatedly using
bottom-up approach.

“ Our objective is here to fill the knapsack in such a manner that value is
maximum.”

NIT, Kurukshetra 10
 (0-1 KNAPSACK PROBLEM)
STEP :-1. Create a table V(i,w), Where i = item no.
w = Weight of item.

Here, I = 0 to 4;
and, w = 0 to 5; because max. capacity W = 5.

V(i,w) w=0 1 2 3 4 5
i=0 0 0 0 0 0 0
1 0
2 0
3 0
4 0

Fill the row i = 0 with 0. This means when 0 item is considered weight is 0.
Fill the column w = 0 with 0. This means when weight is 0 then items considered is 0.
NIT, Kurukshetra 11
 (0-1 KNAPSACK PROBLEM)
Procedure to fill the V[i,w] table.
Items Weight(kg.) Value
if weight[i] > w then 1 3 100
V[i,w] = V[i-1,w] n=4 & W=5
2 2 20

else if weight[i] <= w then 3 4 60

V[i,w] = max( V[i-1,w], val[i] + V[i-1, w - weight[i]] ) 4 1 40

V(i,w) w = 0 1 2 3 4 5 For i=1 and w=1, For i=1 and w=2,

i=0 0 0 0 0 0 0 If weight[1]>w If weight[1]>w
3 > 1 3 > 2
1 0 0 0
So, V[1,1]= V[0,1] So, V[1,2]= V[0,2]
2 0 V[1,1]= 0 V[1,2]= 0
3 0
4 0

NIT, Kurukshetra 12
 (0-1 KNAPSACK PROBLEM)
if weight[i] > w then Items Weight(kg.) Value
V[i,w] = V[i-1,w]
1 3 100
else if weight[i] <= w then
2 2 20
V[i,w] = max( V[i-1,w], val[i] + V[i-1, w - weight[i]] )
3 4 60
4 1 40

V(i,w w = 0 1 2 3 4 5 For i=1 and w=3,

If weight[1]>w
) 3 > 3 No,
i=0 0 0 0 0 0 0 V[1,3]=max(V[0,3], val[1]+V[i-1,w-weight[i]])
=max(0,100+V[0,3-3])
1 0 0 0 100 100 =max(0,100+V[0,0])
=100.
2 0
For i=1 and w=4,
3 0 If weight[1]>w
3 > 4 No,
4 0 V[1,4]=max(V[0,4], val[1]+V[i-1,w-weight[i]])
=max(0,100+V[0,4-3])
=max(0,100+V[0,1])
=100.
NIT, Kurukshetra 13
 (0-1 KNAPSACK PROBLEM)
A f t e r, f i l l i n g o u t t h e t a b l e w e h a v e t o f i n d t h e Items Weight(kg.) Value
maximum value V[n,w].
1 3 100
max V[n,w]=140. 2 2 20
3 4 60
4 1 40

V(i,w w = 0 1 2 3 4 5
)
i=0 0 0 0 0 0 0
1 0 0 0 100 100 100
2 0 0 20 100 100 120
3 0 0 20 100 100 120
4 0 40 40 100 140 140

NIT, Kurukshetra 14
 (0-1 KNAPSACK PROBLEM)
Rule for back tracking the items that we have put Items Weight(kg.) Value
inside into knapsack.
set i=n and w=W 1 3 100
while I and w > 0 do
if V[i,w] != V[i-1,w] 2 2 20
then mark ith item 3 4 60
w= w-weight[i]
i= i-1 4 1 40
else i= i-1
end if i=3 and w=4 ;
V[3,4] !=V[2,4]
end while
i =3-1=2.
V(i,w w = 0 1 2 3 4 5 i=4 and w=5 ;
) V[4,5] !=V[3,5] i=2 and w=4 ;
m a r k 4 th i t e m V[2,4] !=V[1,4]
i=0 0 0 0 0 0 0 w=5-weight[4] i =2-1=1.
1 0 0 0 100 100 100 w=5-1=4.
i =4-1=3. i=1 and w=4 ;
2 0 0 20 100 100 120 V[1,4] !=V[0,4]
So ,item 1 & 4 were m a r k 1 st i t e m
3 0 0 20 100 100 120 w=4-weight[1]
put inside the
4 0 40 40 100 140 140 knapsack. w=4-3=1.
NIT, Kurukshetra i =1-1=0. 15
 (0-1 KNAPSACK ALGORITHM)

100

// bi is value of ith item

Time
complexity:
O(n.W)

NIT, Kurukshetra 16
 THANK YOU

NIT, Kurukshetra 17
 REFERENCES
https://www.dyclassroom.com/dynamic-programming/0-1-knapsack-problem

https://www.tutorialspoint.com/design_and_analysis_of_algorithms/design_and_analysis_of_algorithm
s_01_knapsack.htm

https://www.sanfoundry.com/dynamic-programming-solutions-0-1-knapsack-problem/

18