Environmental Impact Assessment using Fuzzy logic inference model (Case study: Kamal Saleh Dam

Document Type : Research Paper


1 MSc, Natural Resources-Environment Engineering, faculty of Natural Resources, Isfahan University of Technology

2 Associate Professor, Faculty of Natural Resources, Isfahan University of Technology


Determining the importance of environmental impacts is one of the main issues and concerns in the process of environmental impact assessment (EIA) of projects. Ecological impacts assessment is very complicated and requires estimates and anticipates of all environmental impacts and thus always associated with uncertainty. A major problem in assessment of ecological impacts is that they cannot be formulated in one algorithm, because the vital elements and their interactions cannot be identified completely unambiguous (Crisp). On the other hand the spatial heterogeneity of ecological systems and the complexity of decision making have made the impact assessment more difficult. Therefore, the concept of environmental impacts is often ambiguous. The main problem is that EIA models are incapable of managing qualitative data. Fuzzy inference will avoid these difficulties. Fuzzy logic brings a method for a broad range of objective data, quantitative data, opinions and subjective judgments to a natural language to describe the environmental impacts. Fuzzy logic is an especial, powerful technique for classifying and describing the environmental conditions, with both natural and human origin. Fuzzy logic has the ability to quantify and classify the environmental impacts with subjective nature. The main objective of Fuzzy inference method for EIA is to calculate the significance of effects based on fuzzy logic. In this study, fuzzy logic used to determine and rank the significance of impacts, as a method to assess qualitative data. Thus the performance of fuzzy logic inference method in comparison with mathematical matrix method is discussed.
1. vfaramarzy@gmail.com
2. soffianian@cc.iut.ac.ir
Materials and Methods
Study area
Kamal Saleh dam basin with 49˚ 4ˊ 2˝ to 49˚ 27ˊ 11˝ east longitude and 33˚ 33ˊ 13˝ to 33˚ 55ˊ 55˝ North latitude, with an area of ​​655 square kilometers, located in south west of Markazi province and North East of Lorestan province in western Iran. This dam has been constructed at a distance of 74 km from the Arak city and 46 km of Shazand city on the Tireh River.
Study Methods
Mathematical Matrix Method
In this research, a case study (Kamal Saleh DAM) used to assess the environmental impacts using fuzzy logic as a novel method for the environmental impact assessment. Therefore an evaluation was done by two methods: mathematical matrix and fuzzy logic. The mathematical matrix method was used after indigenization, so that the matrix is ​​composed of two parts: a complementary index and a basic index, each index includes three criteria which include six criteria overall (magnitude (Mij), duration (Dij) & occurrence Time (Tij) as basic index parameters, and synergy effects (Sij), cumulative effects (Aij) & Probability of occurrence (Pij)  as complementary index parameters) along with the nature of impact (Nij)  showed by + and – symbols which indicate the desirable and undesirable effects of the impact (Iij), respectively. Finally, using the mentioned mathematical relation, we achieved the importance of each action’s impact on the environment.







Then, the importance of effects classified as very low (VL: 0.2≥X>0.36), low (L: 0. 36≥X>0.52), medium (M: 0.52≥X>0.68), high (H: 0.68≥X>0.84) and very high (VH: 0.84≥X>1).
Fuzzy logic study method
In the fuzzy logic method, above mentioned mathematical matrix indices were considered as fuzzy inference system's input. Criteria got fuzzificationed and after determination of membership functions similar to the groups of mathematical matrix classification, and forming rule base center the importance of impact calculated by using the center of gravity method as Defuzzzification approach. The output of the fuzzy logic inference actually is the effect of each activity on the environment and ultimately, the efficiency of two mentioned methods was compared for assessment of effect importance. These two methods have quite similar inputs and finally classified outputs which actually is the importance of the impact, were compared. To do this, in mathematical matrix method and fuzzy logic, 6 criteria for 2 indices (complementary index & basic index) were used (magnitude (Mij), duration (Dij) & occurrence Time (Tij) as basic index parameters, and synergy effects (Sij), cumulative effects (Aij) & probability of occurrence (Pij) as complementary index parameters).
In the method of fuzzy inference system using Matlab Ver R2012a software and applying Mamadani implication method and use the same mathematical matrix indices as system input was implemented.
Discussion of Results and Conclusions
According to the below chart review (Fig.1), the difference in the number of linguistic variables in mathematical matrix and fuzzy methods is obvious. These differences arise from the decision making method in Aristotelian logic and fuzzy logic. In mathematical matrix if the number is placed in border area (high or low range), still belongs to the same range. The importance of the impact calculated based on a mathematical matrix class can create Uncertainty, which is more important in borders of classification (where X is increasing along with the value of impact from very low to very high). i.e. as we move towards increasing the variable X, the value of linguistic variable have increased. This can be seen as several classes in output matrix. For example, if the variable is X=0.53, belongs to medium-class and if variable is X=0.67, still belongs to the same class, even though there has been a major numerical increase; on the other hand, with the increase of 0.01 at 0.67 point, the importance of impact will change from medium to high.

Figure 1. Comparing the numeric summation of whole impacts importance (positive and negative) in mathematical matrix and fuzzy logic methods
But fuzzy logic approach solved this problem and its output defined based on membership grade. For example, if the output of fuzzy logic is 0.67, then the fuzzy logic determines a degree of membership for two membership functions, and thus the uncertainty in the mathematical matrix classification, which is acting as a binary logic, would improve. Impact importance of Ȳ=0.67 in fuzzy logic belongs to two membership functions with different membership levels, moderate linguistic variable with 0.06 degree of membership and high linguistic variable with 0.94 membership degree. The concept of environmental impact Assessment is unambiguous and ecological effects cannot be explicitly defined, for this reason the fuzzy logic has a very high performance in formulating the importance of each impact in an appropriate manner. Fuzzy logic is capable of using qualitative criteria or linguistic variables for assessment and solves the problem of the variables formulation and simultaneously is capable to use and synthesis both qualitative and quantitative data derived from environmental assessors. As a result, the fuzzy logic method leads to modification of uncertainty which always is a problem in unambiguous and complicated matters such as EIA. Since one of the main issues in environmental impact assessment (regarding project approval and determination of appropriate corrective solutions) is to define the impact significance correctly; the fuzzy logic with its spectacular capabilities is an appropriate method. Determining the importance of environmental impacts is one of the main issues in the process of environmental impact assessment (EIA). Ecological impacts assessment is very complicated and always associated with uncertainty because the assessment data are often qualitative and common EIA methods are incapable of managing these kind of data.


امینی‌‌‌فسخودی، ع. 1384. کاربرد استنتاج منطق فازی در مطالعات برنامه‌‌ریزی و توسعه منطقه‌‌ای، مجلۀ دانش و توسعه، شمارۀ 17، ص 39 تا 61.
ب شپارد، ر. 2005. ارزیابی اثرات توسعه با منطق فازی (مترجم: عبدالرسول سلمان‌ماهینی)، انتشارات مهر مهدیس، چاپ اول، 268 ص.
پزشکی، و.، زرافشان، ک. 1387. کاربرد منطق فازی در ارائۀ مدل ارزیابی سطوح توسعۀ کشاورزی دهستان‌های شهرستان کرمانشاه، فصلنامۀ روستا و توسعه، سال 11، شمارۀ 4، ص 53 تا 70.
پورقاسمی، ح. ر، م، حمیدرضت.، محمدی، م.، و مهدوی‌‌فر، م. ر. 1387. تهیۀ نقشه حساسیت به خطر زمین‌لغزش و ارزیابی آن با استفاده از اپراتورهای فازی، علوم و فنون کشاورزی و منابع طبیعی، سال دوازدهم، شمارۀ 46، ص 375 تا 389.
خدابخشی، ب.، جعفری، ح. ر. 1389. بررسی کاربرد مدل دسته‌بندی چندمعیارۀ Electre - TRI در تعیین اهمیت آثار محیط‌زیستی (مطالعۀ موردی: ارزیابی آثار محیط‌زیستی طرح سد و شبکۀ آبیاری- زهکشی اردبیل)، مجلۀ پژوهش‌‌های محیط‌زیست، سال اول، شمارۀ 2، ص 31 تا 42.
شکیبایی، ع. 1387. برآورد کشش عرضۀ خدمات درمانی با استفاده از منطق فازی (Fuzzy logic)، مجلۀ توسعه و سرمایه، سال اول، شمارۀ 2، ص 149 تا 181.
صالحی، ج.، مرادی، ح. 1390. منطق فازی و کاربرد آن در ارزیابی اثرات محیط‌‌‌زیستی، محیط‌‌زیست و توسعه، سال 2، شمارۀ 3، ص 37 تا 44.
طاهری، س. م. 1384. سیمای منطق فازی، مجلۀ فرهنگ و اندیشه ریاضی، شمارۀ 35، ص 73 تا 92.
قاسمی، ع.، محمودزاده، س. 1389. ارزیابی طرح‌‌های اقتصادی در شرایط عدم قطعیت (رویکرد فازی)، مجلۀ تحقیقات اقتصادی، شمارۀ 93، ص 83 تا 108.
کریمی، م.، مسگری، م.س.، و شریفی، م. ع. 1388. مدل‌‌سازی توان اکولوژیکی سرزمین، با استفاده از منطق فازی (منطقۀ مورد مطالعه: شهرستان برخوار و میمه)، مجلۀ سنجش از دور و GIS ایران، سال اول، شمارۀ اول، ص 17 تا 38.
مخدوم، م. 1387. چهار نکته در ارزیابی اثرات توسعه، نشریۀ علمی محیط و توسعه، سال دوم، شمارۀ سوم، ص 9 تا 12.
منعم، م. ج.، خرمی، ج.، و حیدری‌‌یان، س. ا. 1386. ارزیابی عملکرد شبکه‌‌های آبیاری با استفاده از منطق فازی: مطالعۀ موردی شبکۀ مارون، مجلۀ فنی و مهندسی مدرس، ص 33 تا 42.
مهندسان مشاور سامانۀ فرایندهای محیطی. 1383. مطالعات ارزیابی زیست‌‌محیطی سد ماشکید، جلد چهارم، ارزیابی آثار محیط‌‌زیستی.
نجفی‌‌نژاد، ع.، مردیان، م.، وروانی، ج.، و شیخ، واحدبردی. 1390. ارزیابی کارایی ضرایب اصلاحی در بهینه‌‌سازی منحنی سنجه رسوب (مطالعۀ موردی: حوضۀ سد کمال صالح استان مرکزی)، مجلۀ پژوهش‌‌های حفاظت آب و خاک، جلد هجدهم، شمارۀ دوم، ص 105 تا 122.
Andriantiatsaholiniaina, L.A., Kouikoglou, V.S., Yannis, A.P. 2004. Evaluating Strategies for Sustainable development: Fuzzy logic reasoning and sensitivity analysis. Ecological economics, 48 (2): 149-172.
Blanco Moron, A., Delgado Calvo-Floresa, M., Martín Ramo, JM., and Polo Almohano, M.P. 2008. AIEIA: Software for Fuzzy Environmental Impact Assessment, Expert Systems with Applications, 36: 9135-9149.
Bojorquez-Tapia, L.A., Ezcurra, E., Garcia., O. 1998. Appraisal of Environmental Impacts and mitigation Measures through Mathematical Matrices, Journal of Environmental Management, 53: 91-99.
Bojorquez-tapia, L.A., Juarez, L., Cruz-Bello, G. 2002. Integrating Fuzzy Logic, Optimization, and GIS for Ecological Impact Assessments, Environmental Management, 30(3): 418- 433.
Canter, L.W. 1996. Environmental Impact Assessment. Mc Grew Hill Book Co. Baltimore.
Liu, K.F.R., Yu, C.W. 2009. Integrating case-based and Fuzzy Reasoning to Qualitatively Predict Risk in an Environmental Impact Assessment Review. Environmental Modeling and Software 24:1241- 1251.
Maria Valente, T., Joao Ferreira, M., Leal Gomes, C. 2011. Application of Fuzzy logic to Qualify the Environmental Impact in Abandoned Mining Sites, Water Air Soil Pollution, 217: 303-315.  
Pislaru, M., Trandabat, A. 2012. Assessment of c Environmental Impact using Fuzzy Logic, International Proceedings of Chemical, Biological & Environmenta, vol 32: 97.
Plillis, Y.A., Andriantiatsaholiniaina, L.A. 2001. Sustainability: an ill-defined Concept and its Assessment using Fuzzy logic, Ecological Economics, 37(3): 435- 456.
Silvert, W. 2000. Fuzzy Indices of Environmental Conditions, Ecological Modeling, 130(1-3): 11-119.