Recent Research in Science and Technology 2014, 6(1): 122-125

ISSN: 2076-5061
Available Online: http://recent-science.com/

Fuzzy expert system for drinking water quality index

Nidhi Mishra1 and P. Jha2

1Raipur Institute of Technology & Engineering College, Chhattisgarh, India.

2J. Y. Chhattisgarh College, Chhattisgarh, India.

Abstract
The role of water in human life is significant as it plays a very essential role in the procedure of human body. Fuzzy logic
provides an efficient and useful device for classifying drinking water quality based on limited observations. In this study, a
Fuzzy Drinking Water Quality Index (FDWQI) is proposed for evaluation of water quality for drinking purpose. Fuzzy expert
system makes it possible to combine the certainty levels for the acceptability of water based on an approved parameter.

Keywords: Water Quality Index, Fuzzy Logic, Fuzzy Expert System, Water Quality Classification. Drinking water Quality
Index.

INTRODUCTION

Now a day’s environmental protection and water quality sufficient pure and clean water, they can damage every aspect of
management has become a significant issue in public policies all their physiology.
over the world. Availability of fresh and clean water for human Water quality of any specific area or specific source can be
utilization is one of the most important issues. Classification of water assessed using physical, chemical and biological parameters. The
resources to satisfy water quality standards is an important issue in values of these parameters are unsafe for human health if they
this respect. Because Human body is a water mechanism designed occurred more than defined limits [1, 2, 3, and 4]. Therefore, the
to run primarily on water and minerals. The movement of water suitability of water sources for human utilization has been described
within our cellular system also transports very important blood in terms of Water quality index (WQI), which is one of the most
plasma, 92% of which is made up of water. It confirms the quality of efficient and useful ways to explain the quality of water. It has been
water we drink will have severe impact on our overall condition of realized that the use of human being water quality parameter in order
health. Our brain contains more than 80% water and controls each to explain the water quality for general public is not logical and
and every progression that happens within our body. Considering the simply understandable [5,6]. So for that WQI has the ability express
very important role that water plays in our brain and nervous system the data in a simplified and logical form it takes information from a
is key to long life. The purity of water we drink causes impact on our number of sources and combines them to develop an overall status
strength and energy level. The quality of drinking-water is a powerful of a water system. They increase the understanding ability of
environmental determinant of health. Assurance of drinking-water highlighted water quality issues by the policy makers as well as for
safely is a foundation for the prevention and control of waterborne the general public as users of the water resources. The present
disease. The biggest problem with water today is that the public does study developed a fuzzy expert system for water quality index, which
not really believe there is anything wrong with it. People think water is easy to understand, logical and useful for common people.
problem begin when the water starts to develop bad smell, taste or Initially, WQI (water quality index) was developed by Horton
look dirty, however, harmful pollutants and contaminants usually (1965) [7] in United States by selecting 10 most commonly used
don't make water smell or taste bad. water quality parameter. Like dissolved oxygen (DO), pH, coli forms,
Today an estimated 50 to 70% of our populations, in rural turbidity specific conductance, alkalinity and chloride etc. and has
areas are drawing water form "Poisoned lakes, streams and water been usually applied and accepted in European, African and Asian
tap ". Chemicals and Viruses are actually passed into households for countries. Furthermore, a new WQI similar to Horton’s index has also
everyday use and can cause cancer, Birth defects and genetic been developed by the group of Brown in 1970 [8], which were
damage. based on weights to individual parameter. Recently, many
According to the World Health Organization- "80% of all modifications have been considered for WQI concept through
diseases are water borne". We can go without food for almost two various scientists and experts [9, 10].
months, but without water only a few days. So if people don’t drink A fuzzy logic approach to analyzing water quality could help
reduce the number of people in the developing world forced to drink
polluted and diseased water for survival. In this paper, we explain a
new approach to water quality evaluation uses fuzzy logic to
*Corresponding Author combine different problems and provide a more precise indicator of
overall quality [6].
Nidhi Mishra Fuzzy logic provides a structure to model ambiguity. Fuzzy
Raipur Institute of Technology & Engineering College, Chhattisgarh, India. logic was first introduced by Zadeh (1965) [11]. Kumar et al. [12] has
used the concept of degree of match in fuzzy environment to study
the water pollution of Sangam Zone, Allahabad in 2004 and
Recent Research in Science and Technology 2014, 6(1): 122-125 123

assessment of quality of water rivers Ganga and Yamuna during

Ardh Kumbh 2007 was studied by Yadav [13] under the supervision Here FWQI based on five input parameter. System output is
of Pankaj Srivastava. Raman et al. (2009) believed that fuzzy logic actually giving the drinking water quality which is divided into four
concepts, if used logically, could be a capable tool for some of the types. The knowledge base describing the system’s behavior is
environmental strategy matters [14]. Jinturkar et al. (2010) used the represented by the membership functions designing the linguistic
fuzzy logic for evaluating the water quality index on the basis of variables. Hence, for the proposed system’s behavior, six linguistic
which, water quality rankings are given to determine the quality of variables are defined. Out of these six, five is input variables
water. The fuzzy index has been shown to be efficient in avoiding the namely- turbidity (T), dissolved oxygen (DO), biochemical oxygen
loss or non-detection of information critical for classification of water demand (BOC), pH value (pH), and Fecal Coli form (FC) and one is
quality (Roveda et al., 2010) [15]. Babaei Semiromi, F,,Hassani, output variable drinking water quality. fuzzyfier unit computes the
A.H.,Torabian A., Karbassi, A.R and Hosseinzadeh Lotfi was membership values of each input variable in accordance with the
developed a water quality index for karnoor river of Iran during 2011 fuzzy values defined in the database. Inference engine interprets the
[16]. A water quality index was developed by Pankaj Sriivastava, rules combined in the rule base. The Inference engine is performed
Anjali Burande and Neeraja Sharmal at sangam jone during in three steps, (1) antecedent establishment (2) implication (3)
mahakubh 2013 [17]. aggregation. Finally, defuzzification convert the fuzzy output into
In this study, the fuzzy logic has been used to assessment of cripes value.
drinking water quality by developing a Fuzzy Expert System for
drinking water quality index based on fuzzy logic. Membership Function of Input and Output:

MATERIALS AND METHOD Water quality is determined by its physical, chemical, and
Fuzzy expert System bacteriological characteristics and in order to evaluate quality of
water. The parameters measured for analysis are turbidity (T),
Fuzzy Expert System (figure no.1) is a rule based expert dissolved oxygen (DO), biochemical oxygen demand (BOC), pH
system where fuzzy logic is used as a tool for representing different value (pH), and Fecal Coli form (FC) Figures 3, 4, 5, 6 and 7. Figure
form of knowledge about the problem. no. 8 shows the membership value of output result. In order to
design a user friendly environment model for evaluating drinking
water quality.

Fig 1. Fuzzy Expert System

The proposed system consists of 5 input, 1 output and 10

rules and actual flow diagram for fuzzy expert system is shown in Fig. Fig 3. Membership function for Turbidity
2. The present system is used for developing a drinking water quality
index.

Fig 4. Membership function for dissolved oxygen.

Fig 2. Flow diagram of fuzzy Expert system

Fig 5. Membership function for Biochemical oxygen dissolved
124 Mishra and Jha

Fecal Coli form (FC) - FC is categorized into four fuzzy sets and
their membership functions are given below:
(1) Good (G) : <5000,
(2) Not Good (NG) : >5000.

Output variable is quality of drinking water. There are three classes

of drinking water quality and the range for membership function is
from 0 to 10.
(1)Good for drinking water: 7 -10
(2) Good for drinking water after treatment: 5-7
Fig 6. Membership function for pH factor (3) Not good for drinking water: 0-5

In this study we use triangle and trapezoidal membership

function that defines how each point in the input space is mapped to
a membership value between 0 and 1. In order to suggest a user
friendly environment model for evaluating Drinking Water Quality
Index. Steps of this expert system are - fuzzification, assessment of
inference rules, and defuzzification of fuzzy output results.
Conception of inference rule is an if–then rule has the form: “If x is A
then z is C”, the if–part is called the antecedent, while the then–part
. is called the consequent. The antecedent and the consequent of a
Fig 7. Membership function for Fecal Coli form rule can have multiple parts. Fuzzification is process to define inputs
and outputs as well as their individual membership function that
convert the numerical value of a variable into a membership grade to
a fuzzy set. After the inputs are fuzzified, the degree to which each
part of the antecedent is satisfied for each rule. If the antecedent of a
given rule has more than one part, the fuzzy operator is applied to
obtain one number that represents the result of the antecedent for
that rule. This number is then applied to the output function. The
input to the fuzzy operator is two or more membership values from
fuzzified input variables. The output is a only truth value. The input
for the inference process is a single number given by the antecedent,
Fig 8. Membership function for Output Result and the output is a fuzzy set. Implication is implemented for each
rule. Because decisions are based on the testing of all of the rules in
Input Parameters an FES, the rules must be combined in some way in order to make a
decision. Aggregation is the procedure by which the fuzzy sets that
Turbidity (T)- Turbidity is categorized into three fuzzy sets with correspond to the outputs of each rule are combined into a particular
membership functions as follows: fuzzy set. Finally, the input for the defuzzification procedure is a
(1) Good (G): 0–5NTU, fuzzy set (the aggregate output fuzzy set) and the output is a single
(2) Little bit good (LG): 3–8NTU, number. In this research, each of the six input parameter has been
(3) Poor (P): 7–10NTU. divided into different categories and defined by triangular and
trapezoidal membership function. The membership functions were
Dissolved Oxygen (DO) - DO is categorized into four fuzzy sets and assigned as shown in Figure 3,4,5,6, 7 and 8. Water quality index is
their membership functions are given below: shown in Table 1and testing of that water quality index (WQI) is
(1) Low (L): 0–4 mg/L, shown in table no.2.Five fuzzy sets have been considered to be
(2) Moderate (M): 2–6 mg/L, suitable for this study. Ranges for fuzzy sets were based on CPCB-
(3) High (H): 5–9 mg/L, Indian standard specifications for drinking water is-10500.
(4) Very high (VH): 7–10mg/L. Five quality parameters have been selected to estimate water
quality by means of an aggregated index called fuzzy drinking water
Biochemical Oxygen Dissolved (BOD) - BOD is categorized into quality index (FDWQI). For the selected set of five parameters, the
four fuzzy sets and their membership functions are given below: most prominent 10 rules have been used. For example, fuzzy rule is
(1)Low (L): 0–3 mg/L, chosen. Rule 1: If TDS is good, and DO is high .BOD is low, PH is
(2) High (H): 2–6 mg/L, desirable and fecal coli form is good then WQI is good. Same way,
(3) V. High (VH): 5–10 mg/L, other rules can be enunciated. Strength of the system depends on
the number and quality of the rules. Every rule in fuzzy rule base
PH Factor (pH)- The pH factor is categorized into three fuzzy sets system consists of rule antecedents (inputs) and rules consequents
and their membership functions are given below: (output). Table 1 shows the rule antecedents and consequents for
(1) Acidic (A): 0-6 this system.
(2) Desirable (D): 6 – 9
(3) Basic (B): 8-10 Table 1. Rule Antecedent and Consequence
Recent Research in Science and Technology 2014, 6(1): 122-125 125

Cybernetics, vol. 6, no. 10, pp. 698–703.

[6] R. E. Bellman and L. A. Zadeh, 1970. “Decision making in a
fuzzy environment, Management Science,” Management
Science, vol. 17, no. 4, pp. B141–B164.
[7] Horton, R.K., 1965. “An index number system for rating water
quality”, J. Water Pollu. Cont. Fed., 37(3). 300-305.
[8] Brown, R.M., McClelland, N.I., Deininger, R.A. and Tozer, R.G.,
1970. “Water quality index-do we dare?”, Water Sewage
Works, 117(10). 339-343.
Table 2. Testing of FDWQI [9] Bhargava, D.S, Saxena, B.S. and Dewakar, R.W.,1998. “A study
of geo-pollutants in the Godavary river basin in India”, Asian
Environ., 12. 36-59.
[10] Dwivedi, S., Tiwari, I.C. and Bhargava, D.S., 1997.“Water quality
of the river Ganga at Varanasi”, Institute of Engineers, Kolkota,
78, 1-4.
[11] L. A. Zadeh, 1965.“Fuzzy sets,” Information and Control, vol. 8,
no. 3, pp. 338–353.
[12] M. Kumar, 2004. Application of Fuzzy Theory Approach to
StudyWater Pollution of Sangam Zone, M.Tech. Thesis Civil
CONCLUSION Engineering, Motilal Nehru National Institute of Technology,
Allahabad, India.
Fuzzy logic is the flexible device to develop categorization [13] S. Yadav, 2007. Water quality assesment of water Ganga and
model with an easy to understand framework and constructed with Yamuna during Ardh Kumbh-2007 by Fuzzy Analysis [M.S.
simple and efficient language. In this study, we developed a fuzzy thesis], Environment Science, Allahabad University.
drinking water quality index, which was obtained to express the
categorization of drinking water quality assessment more [14] Raman BV, Reinier B, Mohan S 2009. Fuzzy logic Water Quality
understandable, especially in public consideration. It is an easy to index and importance of Water Quality Parameters. Air, Soil
use and effective fuzzy expert system from which ordinary people Water Res., 2: 51–59.
can access the quality of their drinking water. [15] Roveda SRMM, Bondança APM, Silva JGS, Roveda JAF, Rosa
AH 2010. Development of a water quality index using a fuzzy
REFERENCE logic: A case study for the sorocaba river. 2010 IEEE World
[1] Guidelines for Drinking-water Quality, 2012. Fourth Edition, Congress on Computational Intelligence, WCCI 2010, art. no.
World Health Organization ISBN 978 92 4 154815 1. 5584172
[2] Bureau of Indian Standards, 2012. Specification for drinking [16] Babaei Semiromi,F.,Hassani,A.H.,Torabian,A.,Karbassi,A.R.and
water. IS: 10500, New Delhi, India. Hosseinzadeh Lotfi A, Water quality index development using
fuzzy logic: A case study of the Karoon River of Iran, African
[3] United State EPA 816-F-09-004, May 2009, Journal of Biotechnology Vol. 10(50), pp. 10125-10133,
http://water.epa.gov/drink/contaminants/upload/mcl2.pdf September, 2011 .
(Accessed 12 July 2013).
[17] Srivastava Pankaj,Burande Anjali, Sharaml Neeraja, 2013. Fuzzy
[4] Guide Manual: Water and Waste Water, Central Pollution Control Environmental Model for Evaluating Water Quality of Sangam
Board, New Delhi. Zone during Maha Kumbh 2013.Department of Mathematics,
http://www.cpcb.nic.in/upload/Latest/Latest_67_guidemanualw Motilal Nehru National Institute of Technology, Allahabad,India.
&wwanalysis.pdf (Accessed 12 July 2013).
[5] R. Jain, 1976. “Decision making in the presence of fuzzy
variables,” IEEE Transactions on Systems, Man and