Archives of Business Research – Vol.4, No.6 Publication Date: December. 25, 2016 DOI: 10.14738/abr.46.2508. Rohani, A., Keramati, A. & Razmi, J. (2016). Presenting an Innovative Approach Combining FA and ANP Methods to Identify and Prioritize Water Pollutants in Petroleum and Petrochemical Industries. Archives of Business Research, 4(6), 405-418. Presenting an Innovative Approach Combining FA and ANP Methods to Identify and Prioritize Water Pollutants in Petroleum and Petrochemical Industries Abolfazl Rohani Department of Industrial Engineering Payam Nour University (PNU), Iran and School of Industrial Engineering, College of Engineering University of Tehran, Tehran, Iran Abbas Keramati School of Industrial Engineering, College of Engineering Alborz Campus University of Tehran, Tehran, Iran Jafar Razmi School of Industrial Engineering, College of Engineering Alborz Campus University of Tehran, Tehran, Iran Abstract Water contamination is one of the important challenges and problems of the world and Iran. One of the main causes of disease and mortality in the world is water contamination. Surface and groundwaters are both subject to different contaminants. Considering the nature, origin of entrance of the pollutants, being natural or man-made, various classifications have been considered for contaminants. The major organic contaminants of water include agricultural, chemical, oil, and food contaminants. Each of these four groups has subcriteria. In this research, accurate investigation, identification, and ranking of water contaminants it all oil and petrochemical industries (urban area of Tehran province) have been dealt with using factor analysis and analytical network process methods. The results of factor analysis method suggest that the contaminants of nitrate, sodium, and TDH claimed the highest score by experts of environment in the field of water contamination. Furthermore, the results of analytical network process showed that agricultural contaminants with the weight of 0.321 stood in the first rank, while oil contaminants with the weight of 0.152 ranked the fourth. Keywords: water contaminants, analytical network process, principal component analysis, factor analysis, oil and petrochemical industry INTRODUCTION Today as the demand market has become something central, it has resulted in development of the significance of supply chain such that it has changed into one of the vital instruments in commercial competitions. On the other hand, environmentalism has changed into an important commercial- social issue. Considering the significance of these two issues, green chain concept was developed and has attracted the attention of pioneering companies across the global competition. Today, the managers of leading companies try to benefit from improving their Copyright © Society for Science and Education, United Kingdom 405 Rohani, A., Keramati, A. & Razmi, J. (2016). Presenting an Innovative Approach Combining FA and ANP Methods to Identify and Prioritize Water Pollutants in Petroleum and Petrochemical Industries. Archives of Business Research, 4(6), 405-418. environmental performance throughout the entire supply chain as a strategic weapon for gaining competitive advantage through developing utility and satisfaction throughout the entire supply chain (Razmi and Nasrollahi, 1392). Furthermore, the environment that has surrounded the human and allowed him to leave has been threatened by the human itself. Today, we live in a world in which the plight of environmental contamination due to the rapid growth of population, industry, and constraint of natural resources has become of interest to scientists and experts more than ever it has also been paid attention to by the public as a tangible issue. In today's societies, the significance of environmental protection seems to be something essential and evident. Undoubtedly, conducting and implementing any program requires sufficient knowledge and essential understanding of the environment and its contaminants. Currently, the environmental crises caused by the contaminations is threatening many countries seriously. Therefore, countries can control these environmental crises through serious and logical protection of their environment and scientific programs. In this regard, recognizing water contamination and the industries that develop these contaminations has facilitated the way to combating such contaminations (Ramudhin et al, 2010). Tom Buchert et al (2015) presented development of sustainable product- advantages and obstacles in a research called multicriteria decision-making as an instrument for. In that paper, they merged discrete decision-making trees with lifecycle sustainability assessment (LCSA) for development of multi-criteria quantitative analysis. Further, to improve decision-making they have presented an integrated method for designing sustainable product. The main objective of the selected approach is covering the gap between prospective and retrospective decision- making through quantitative analysis in the preliminary stages of design. The final outcome of this research suggests that for developing sustainable products, design engineers can benefit from multi-criteria quantitative sustainable information which is currently available in the preliminary stages of product design. In another research, Cozzi et al dealt with spatial assessment for agriculture with the aim of increasing the productivity in terms of econometrics in Italy. In this research, which has been done in Basilicata region in Italy, for filtering the refuse resulting from chemical wastes, weighted average method has been used along with verbal quantifier. Accordingly, a suitable map for decreasing the products of a country which was far from chemical wastes and especially water wastes was recognized across the studied region. Based on geological and agricultural scales used in this research, the researchers found that out of the entire studied region, only 25 hectares of the 163 available hectares are cultivable. Furthermore, the results of this research confirmed the fact that if entrance of chemical wastes and especially the water contaminants are not prevented from being introduced into groundwaters of agricultural regions, these regions will be on the verge of devastation. However, Lam and Dai (2015) used a combinational method of developing the quality performance and analytical network process for optimizing the providers of logistics to improve sustainability of green supply chain in order to lower water contaminants. In that research, they concluded that the most important determining factor of sustainability is economic and environmental issues (water contaminants). Martijn Bouwknegt et al (2015) assessed the hazards threatening production of fresh crops despite its green leaves throughout its supply chain in Europe. In that research, based on the studies on a wide range of environmental contaminants and especially water contaminants, they found that there threatening hazards are not merely a general health risk by viruses in foods which are important in supply-chain. Further, they cannot be coped with only through identification and evaluation of viruses in retail products. Accordingly, presentation of a solution in this regard is crucial. In this research, using a conceptual model earned through investigating the indices URL: http://dx.doi.org/10.14738/abr.46.2508. 406 A rchives of Business Research (ABR) Vol.4, Issue 6, December-2016 affecting the quality of agricultural crops including the level of water contaminants such as phenol, nitride, ammonium, and the nitrate present in some foods cultivated close to chemical industries in several points of the member countries of the European Union, the negative effect of these contaminants on the quantity of products has been observed abundantly. The results of research on several specific products such as lettuce and soft fruits such as strawberry, in contaminated regions, when compared with cleaned regions, the quantity of these crops had a drop in quality by 97.5% and this reduction in the quality is due to liberating the contaminants of the current chemical industries in the region to agricultural farms. The structure of this paper is as follows: Sections 2 and 3 deal with the significance and necessity of water resources and the present research. Section 3 deals with the methodology. In Section 4, the stages of the research are dealt with. Statement of the introduction of the effective criteria and subcriteria in the assessment process has been thoroughly discussed in Section 5. Eventually, Section 6 states the conclusion of the research. THE SIGNIFICANCE AND NECESSITY OF PROTECTING WATER RESOURCES Water is one of the most abundant compounds which is found in the nature and occupies almost 75% of the earth surface. However, several factors have contributed to water shortage across the world. The first factor is heterogeneous geographical distribution; over 97% of the entire water resources are found in oceans and seas, where due to the high level of soluble salts, they are not suitable for direct consumption (Alaimo, Peter & Amanda-Lynn Marshall). Over 2% of the entire water resources are present in glaciers and Polar Regions or are found as humidity in the soil and the earth's atmosphere. Due to unavailability, planning for using them is not justifiable in terms of economic feasibility. Thus, for survival and meeting industrial, agricultural, and drinking purposes, human beings can have access to only 62% of the water remain in the lakes, rivers, and groundwaters. This not only has reduced the quality of available water resources and brought about further limitations, but the resources have also faced decreased and altered quality of waters due to development of urbanization and industrial as well as agricultural activities. It is many years that it is investing in proper planning of management of water resources (Azadfallah, M., &Azizi). THE ENVIRONMENT AND ENVIRONMENTAL CONTAMINATIONS AND THE SIGNIFICANCE OF THE RESEARCH The environment refers to all environments in which life exists. A set of external physical factors and living creatures which are interacting with each other constitute the environment and influence the growth and behavior of creatures. Environmental protection in the 21st century is known as one of the eight objectives of the millennial development and one of the three pillars of sustainable development (Azadi et al, 2015). Natural environment is a combinational term from different sciences, which involves a set of biological and environmental factors in the form of environment and non-biological factors (physical, chemical) which influence the life of a person or species and are also influenced by them. Today, this definition is mainly related to human beings and their activities, where the environment can be summarized as a set of the natural agents of the earth such as air, water, atmosphere, rocks, plants, etc. which are surrounding the human. The difference between the environment and nature lies in the fact that definition of nature involves a set of natural, biological, and non-biological factors which are considered exclusively, while the term "environment" has been described based on the interactions between human and nature and according to their viewpoint (Lam and Dai, 2015). The characteristic of the environment of modern business is ever increasing competition and economic globalization. Under such conditions, producers benefit from novel the strategies and Copyright © Society for Science and Education, United Kingdom 407 Rohani, A., Keramati, A. & Razmi, J. (2016). Presenting an Innovative Approach Combining FA and ANP Methods to Identify and Prioritize Water Pollutants in Petroleum and Petrochemical Industries. Archives of Business Research, 4(6), 405-418. technologies to achieve the environment. Therefore, throughout the recent years, supply-chain management 3 has attracted the attention of both academics and industries as an effective business philosophy (Pishvaee & Razmi, 2012). As green supply-chain methods 5 have developed a green revolution across the supply chain, companies should use their internal and external resources to implement the green supply-chain methods. Contamination of water and underground resources with different contaminants brings about various environmental negative effects. The contamination caused by industrial activities has always brought concern for industry stakeholders and the Institutes supervising the environment. Selecting adaptive policies and logical solutions for cleaning the environment in a way congruent with environmental considerations is inevitable. Considering the effect of environment in the life and in turn human survival, protecting the environment is considered something essential. Therefore, to achieve this aim, recognition of the issues that contribute to environmental contamination seems to be evident, as the country has many industrial towns (production industries), thus understanding and being aware of the factors influencing environmental contaminations are considered to be important. Through this, these factors can be mitigated and for eliminating them some important measures can be taken to protect the environment. As water is not a consumer and worthless product. Today, water is a strategic and economic good which is very valuable and those who own it have bargaining power and are able to disturb political and regional equations. Furthermore, water shortage in dry and semidry regions, on the one hand, and wasteful and wrong consumption of water resources and their contamination on the other are serious threats for sustainable development and environmental protection. Based on the points mentioned above and according to the previous studies, we achieved a study gap and absence of an efficient method for identification and prioritization of a variety of water contaminations in green supply-chain. Based on this fact, in this research attempts have been made to present a method with expert system design for identifying and prioritizing a variety of water contaminations in green supply-chain as an instrument for responding to the needs of the target population. Briefly, the novelties of this research include: - in this research, it is predicted that major contaminants and their level in the Iranian oil and petrochemical industry and especially Tehran province are extracted and after presenting a mathematical model with the aim of decreasing environmental contaminants and presenting it to the relevant organizations, an effective step is taken towards preventing devastation of the environment of the development of contaminants this industry. - As Iran is an oil-rich country and has many refineries and petrochemical industries along with relevant industries, the results of the above-mentioned plan can be used in these industries and other similar areas. STAGES OF THE PROCEDURE OF THE RESEARCH In this research, the methodology is of applied type, in terms of objective, while in terms of type, it is an assessment method based on questionnaire analysis in the real world; it can be put forward as a case study. Determining the input and output indices, obtaining and extracting the variables in this type of research itself are important parts of implementation of this research. The following steps were taken: - Determining the criteria through the research background and PCA method - Development of questionnaire - Examining the reliability of the questionnaire - Data collection URL: http://dx.doi.org/10.14738/abr.46.2508. 408 A rchives of Business Research (ABR) Vol.4, Issue 6, December-2016 - Determining the validity of the items of the collective questionnaires - Giving weights to the variables - Analysis and ranking using ANP method Principal component analysis (PCA) Principal component analysis is one of the classic multivariate methods and perhaps their oldest and most famous method. This method was first developed to analyze the structure of variance-covariance matrices and correlation coefficient. As with many multivariate methods, prior to invention of computers, this method was not widely used due to complexity of calculations. Thereafter, in terms of theory and application, it was widely developed and employed. This type of analysis can be focused on from several perspectives. - Conversion of dependent variables to uncorrelated variables - Finding linear combinations with large or small relative variability - Reduction of the volume of data - Data interpretation This type of analysis is not usually considered a final analysis, rather it is mainly used as an intermediate tool for further studies and investigations. The mathematical aspects used in this research involved eigenvalues and the Eigen vectors of symmetrical always positive matrices. Reduction of the volume of data is the main objective of this analysis, where these data consist of a large number of variables with internal correlations, such that the maximum available information possible in the data is kept. This takes place through converting the data (variables) to new variables which are called principal component and are uncorrelated. They are prioritized in an order that a small number of them often bring the changes in the initial variables with them. In the analysis of principal component, although seemingly the main focus is on the variance of variables, considering the relationships between the variances and covariance’s, this method also takes covariances or correlation coefficients into consideration implicitly. Analytical network process (ANP) Saati (1996) presented a method for multi-criteria decision-making. This method is called analytical network process (ANP) which has the aim of developing a model through which complex multi-criteria decision-making problems are analyzed into smaller components and through logical initialization they are converted to simpler components and finally combination of these values help in final decision-making. ANP method is the developed form of AHP method which is able to model the correlations and feedbacks between the elements affecting a decision-making and can take all internal effects of the effective components in decision-making into consideration and introduce them into the calculations. Therefore, with the help of this characteristic, this technique is superior then the previous relevant models. Indeed, it can be stated that AHP is a special form of ANP method. ANP method has two main parts, which combines these two parts in a process. The first part involves groups consisting of controlling criteria and subcriteria as well as the group which takes the options of the volunteer. The second part is a network of vectors and arcs which represent the dependencies and correlations and the feedbacks present in the decision-making system. Eventually, this method is based on conducting paired comparisons which is similar to the paired comparisons performed in AHP method. Analytical network process can be called the most complete multi- criteria decision-making method which has been presented so far. However, the single important problem in this model is performing paired comparisons. This problem which also exists in AHP method is considered problematic as a decision-maker is not always facing accurate states of commenting and in many decision-makings of the real world, the decision- Copyright © Society for Science and Education, United Kingdom 409 Rohani, A., Keramati, A. & Razmi, J. (2016). Presenting an Innovative Approach Combining FA and ANP Methods to Identify and Prioritize Water Pollutants in Petroleum and Petrochemical Industries. Archives of Business Research, 4(6), 405-418. makers cannot make decisions about paired comparisons with certainty. The modeling process involves the following stages: - The first step, basing the model and structure of the problem - The second step, paired comparisons matrix and estimation of the relative weight - The third step, development of the preliminary super matrix - The fourth step, development of weighted super matrix - The fifth step, calculation of the general weighted vector - The sixth step, calculation of the final weight of the criteria INTRODUCTION OF THE INFLUENTIAL CRITERIA AND SUBCRITERIA IN THE EVALUATION PROCESS In this research, by presenting a framework, to identify and prioritize the extent of greenness of production industries and presentation of a solution for enhancing the greenness of supply chain. According to the background and interview with experts, the criteria effective in the evaluation process have been developed as the steps presented in the research according to Fig. 1. In this model, the aim is identification and prioritization of industrial contaminants in water (wastewater) which lies at level I. the production industries which include food industries, chemical industries, agriculture industries, and oil industries in two petrochemical and refineries of the oil in Tehran have been shown at Level II, which are in the form of nitrate (NO3), nitride (N3-), and ammonium (NH3) present in food industries, phosphate (PO4), sulfate (SO4), and chloride (Cl-) present in chemical industries, phenol (C6H5OH), sulfide (H2S), oil, mercury (Hg), and cadmium (Cd) present in oil industries, and nitrate (NO3), ammonium (NH3), phosphate (PO4), and chloride (Cl-) present in agricultural industries, which are considered as subcriteria. BOD, COD, TDS, and TSS which are known as comparative evaluation indices of contaminants across all production industries are considered as the subcriteria at Level III. Finally, these contaminants are prioritized in industry in the order of contamination in the environment. The list of the contaminants includes the following (Cozzi et al, 2015): Introduction of the criteria that detect the contaminants - Biochemical oxygen demand (BOD): oxygen consumption rate inside water by organisms. If BOD is low, water is clean and lacks any organism or the organisms inside water are dead and have no need to consume oxygen. BOD is the amount of oxygen required for biological stability in water. The size of facilities of biological treatment especially the wastewater aeration rate in aeration basins can be calculated by BOD level. If water BOD is 1 ppm, then the water is almost pure. Water with a BOD up to 5 ppm is considered to be relatively pure, but when it exceeds 5 ppm, purity of water is questioned. However if its value goes beyond 20 ppm, the public health is jeopardized. BOD experiments provide a realistic estimation of the quality of oxygen that has been introduced into the water (Bouwknegt et al, 2015). - Chemical oxygen demand: it is the amount of oxygen that is required so that the organic compounds present in the sample are chemically stabilized. The utilized oxidizer is usually potassium dichromate in the presence of sulfuric acid. Wastewater contamination caused by the external compounds that enter water as suspended or soluble form, cause its contamination and production of wastewater. Evidently, the higher the level of these compounds in wastewater, the greater its contamination load. Therefore, measurement of the amount of external compounds of wastewater is the major key in determining the extent of contamination of wastewater (Liu et al, 2011). - Total dissolved solids or concentration of minerals: it is the amount of organic compounds or mixed inside a liquid, where these compounds can exist in water as URL: http://dx.doi.org/10.14738/abr.46.2508. 410 A rchives of Business Research (ABR) Vol.4, Issue 6, December-2016 molecular or ionized form or very tiny grains at micron level as a suspended form. The primary resources (major applications) for TDS in receiving waters, the waters resulting from agriculture and residential areas, soil contamination washing and the contaminated water resources discharged from industrial units and wastewaters. The most important aspect of TDS, taking water quality into consideration, is its effect on the taste of water. Clarity and transparency of water with a TDS lower than 600 mg/l is generally considered a good state. Drinking water with a degree above 1200 mg/l is considered an unfavorable state by the majority of consumers. The difference between TDS and TSS lies in the fact that in TSS the particles cannot pass through a filter with a 2-micron scale and remain suspended in the solution for an unknown time (Mulberry et al, 2010). Classification of water contaminants The processes for selecting the selection criteria of the extent of contamination of industries are based on the two following areas: - Investigation of the criteria used by previous researchers - Study of the contamination indices and important contaminating industries according to managers and experts of environment Based on the mentioned process, a set of the most important selection indices of industries and contaminants were considered as the main measurement criteria. 1. The criterion of contamination of food industries: Based on the extent of significance and development of these contaminants in groundwaters, the criterion of food industries consists of three subcriteria of "nitrate", "phosphorus", and "protein" (Zhang et al, 2013). 2. The criterion of contamination of chemical industries: The four following criteria are considered the most important subcriteria of the characteristics of this industry in laboratory trusted by the environment, which have been selected at this stage: lead, mercury, copper, alkalinity of suspended solids and heavy metals (Wu et al, 2012). 3. The criterion of contamination of oil industries: the following subcriteria are among the most important ones of this industry in measuring the extent of contamination: phenol, hydrogen sulfide, oil, ammonium, and TPH (Zailani et al, 2015). 4. The criterion of contamination of agricultural industries: furthermore, the following criteria have been considered as the subcriteria associated with agricultural industries: nitrate, chemical fertilizers, sediments, and TOG (Yongming et al, 2006). Fig. 1. Represents the major contaminants by individual types. Water contaminants Agricultural Chemical Oil contamination Food contamination contamination contamination Ammonia Lead Nitrate Phenols Mercury Nitrate Chemical fertilizers Hydrogen sulfide Copper Phosphorus Sediments OIL Alkaline suspension Protein TOG TPH Heavy metals Fig. 1. Classification of the contaminants Copyright © Society for Science and Education, United Kingdom 411 Rohani, A., Keramati, A. & Razmi, J. (2016). Presenting an Innovative Approach Combining FA and ANP Methods to Identify and Prioritize Water Pollutants in Petroleum and Petrochemical Industries. Archives of Business Research, 4(6), 405-418. RESULTS Next comes explanation of the results of the open-ended questionnaire and analytical network process method. Statistical analysis According to the previous studies, 20 water contaminant indices were extracted and then distributed among 145 experts, so that based on their previous background, a score between 1 (the least significant) and 10 (the most significant) would be given. Accordingly, for ranking the contaminants, statistical methods and especially factor analysis and PCA were used. Then, we deal with explanation of the results of the open-ended questionnaire. Tables 1-3 present the preliminary factor analysis of the indices for estimating the contributing performance of each of the contaminants on the problem's objective. Table 1. The ratio of the variance of each of the contaminants Communalities Initial Extraction Nitrate 1.000 .783 phosphor 1.000 .671 PH 1.000 .667 OIL 1.000 .667 TDS 1.000 .712 heavy metals 1.000 .579 TSS 1.000 .559 Ammonia 1.000 .599 Phenols 1.000 .532 Protein 1.000 .491 Chemical fertilizers 1.000 .627 COD 1.000 .591 Sediments 1.000 .712 Hydrogen sulfide 1.000 .684 Lead 1.000 .708 Mercury 1.000 .725 Copper 1.000 .726 DO 1.000 .572 BOD 1.000 .472 Extraction Method: Principal Component Analysis. Table 1 provides the ratio of the variance of the scores taken by each contaminant by the experts. As can be observed in the table, the contaminants nitrate, sodium, and TDS have claimed the highest score by the experts of environment regarding water contamination. Table 2 presents the frequency distribution of the sum of the variance of contaminants. In this table, the results were investigated based on PCA method. In the above table, Initial eigenvalues column represents the variance associated with the complete set of contaminants from the initial results. Furthermore, the column of extraction sums of squared loadings demonstrate the variance of the most important contaminants among all the contaminants. URL: http://dx.doi.org/10.14738/abr.46.2508. 412 A rchives of Business Research (ABR) Vol.4, Issue 6, December-2016 Table 2. The sum of distributional variance of the contaminants Total Variance Explained Initial Eigenvalues Extraction Sums of Squared Loadings Component Total % of Variance Cumulative % Total % of Variance Cumulative % 1 1.759 9.257 9.257 1.759 9.257 9.257 2 1.621 8.533 17.789 1.621 8.533 17.789 3 1.529 8.049 25.839 1.529 8.049 25.839 4 1.392 7.325 33.163 1.392 7.325 33.163 5 1.374 7.234 40.397 1.374 7.234 40.397 6 1.178 6.199 46.596 1.178 6.199 46.596 7 1.142 6.013 52.609 1.142 6.013 52.609 8 1.078 5.672 58.281 1.078 5.672 58.281 9 1.000 5.266 63.546 1.000 5.266 63.546 10 .970 5.103 68.649 11 .898 4.728 73.377 12 .857 4.509 77.886 13 .811 4.267 82.153 14 .739 3.890 86.043 15 .667 3.511 89.554 16 .610 3.209 92.763 17 .523 2.750 95.513 18 .469 2.471 97.984 19 .383 2.016 100.000 Extraction Method: Principal Component Analysis. Table 3. The correlation matrix between each of the contaminants Component Matrix(a) Component 1 2 3 4 5 6 7 8 9 TSS -.583 .098 .315 .224 .149 .011 .181 .057 .046 TDS .572 -.169 .230 .350 .044 .000 - .056 - .416 -.046 Lead .534 -.087 .248 - .086 -.303 - .040 .035 .270 -.423 Chemical fertilizers .070 .566 - .202 .171 .043 - .031 .003 - .432 -.204 heavy metals .213 -.503 .260 - .230 .278 - .149 .183 - .077 .149 Ammonia .365 .488 .227 - .112 .088 .253 - .048 .300 -.009 Sediments .126 .138 - .674 .185 .150 .217 .243 .029 .244 Mercury -.215 .205 .472 .187 -.392 .307 .042 - .343 .103 Phenols .104 .327 .275 .379 .275 .299 .041 .147 -.077 BOD .362 .040 - .095 .375 .246 .001 - .109 .085 .331 Copper -.215 -.314 .110 - .170 .525 .319 - .249 - .061 -.312 DO .158 -.202 .291 .404 .432 - .156 - .156 .065 .137 COD -.124 -.350 - .055 .358 -.383 - .206 - .257 - .123 .227 Nitrate .131 .112 .283 .044 -.030 - .402 .712 - .039 -.012 phosphor .198 .348 .174 - .172 -.178 - .300 - .561 .005 .120 OIL -.179 -.235 .170 .344 -.382 .321 .022 .422 .070 Protein .033 .053 - .250 .312 -.019 - .356 - .021 .414 -.170 Hydrogen sulfide .414 -.098 .016 - .360 -.156 .363 .124 .022 .447 PH -.269 .362 .305 - .257 .217 - .276 - .078 .153 .390 Extraction Method: Principal Component Analysis. a. 9 components extracted. Copyright © Society for Science and Education, United Kingdom 413 Rohani, A., Keramati, A. & Razmi, J. (2016). Presenting an Innovative Approach Combining FA and ANP Methods to Identify and Prioritize Water Pollutants in Petroleum and Petrochemical Industries. Archives of Business Research, 4(6), 405-418. Table 3 indicates the correlation matrix between all of the contaminants with nine selected contaminants which had the greatest contamination variance as commented by experts according to Table 2. Based on the results of this table, all of these contaminants have a low correlation with each other. Table 4. The final ranking of the contaminants according to expert comments Final Ranking Score contaminants 2 0.955 Nitrate 10 0.734 phosphor 4 0.808 PH 13 0.719 OIL 15 0.711 TDS 12 0.726 heavy metals 11 0.729 TSS 12 0.726 Ammonia 3 0.83 Phenols 14 0.717 Protein 16 0.707 Chemical fertilizers 7 0.764 COD 8 0.754 Sediments 14 0.717 Hydrogen sulfide 1 0.97 Lead 17 0.703 TOG 6 0.785 Mercury 5 0.788 Copper 9 0.743 DO 17 0.703 BOD As can be observed in Table 4, Lead contaminants is considered the most important water contaminant by the experts. It is followed by nitrate with the score of 0.955, as the second rank. Furthermore, in the above table it can be deduced that several contaminants had the same score according to the experts, which include hydrogen sulfide and heavy metals, both of which had a score equal to 0.717. Analytical Network Process In order to achieve the objective of the research, paired comparisons questionnaires were designed and distributed among the experts. Considering the verbal judgment approach in this research, the statements and numbers written in Table 5 were used. Table 5. The defined numbers relative comparison of the indices (verbal judgment) Priority number Absolute significance 9 Very strong significance 7 Strong significance 5 Weak significance 3 Equal significance 1 Intermediate values 2,4,6,8 In this section, based on the hierarchical network, the prepared paired comparisons tables and the modified method of Asgharpour et al (1388), the weight of the components was obtained and they were then prioritized. The results were calculated by Super Decision software. In this research, ANP technique was used for determining the priority of objectives. For this purpose, using ANP questionnaire, 15 (Wibisono, D., & Khan) managers and experts of the field of environment related to oil and petrochemical contaminants were requested to give their comments. Thereafter, the data written in the matrices were analyzed by Excel and Super URL: http://dx.doi.org/10.14738/abr.46.2508. 414