Selection of Suitable Landfill Site for Municipal Solid Waste Disposal: A Fuzzy Logic Approach

Abstract

A fuzzy-based model was developed in the proposed work to evaluate the suitability of landfill sites for disposing municipal solid waste considering 32 attributes as proposed by Central Pollution Control Board (CPCB). The work is demonstrated with reference to Asansol Municipal Corporation (AMC) area in West Bengal, India, as a case study. The attributes include both quantitative and qualitative variables. These attributes were assessed qualitatively by the experts related to solid waste management. The weightage of each attributes and sensitivity index of each attributes related to alternative sites were defined in the form of triangular fuzzy membership functions based on the intuitions and data related to available literatures. The fuzzy decisions provided by the experts were then defuzzified by Yager’s unit interval method to get the crisp weightage corresponding to each attribute and sensitivity index for attributes corresponding to each proposed sites. At a time the experts were to evaluate one attribute for both weightage and sensitivity index calculation. The performance of the model was checked by determining relative percentage error considering the model obtained weightage and CPCB mentioned weightage. From the results, it was clear that for most of the attributes the weightage obtained by the two methods was more or less same except for the attributes distance from collection area, distance to nearest drinking water source, public acceptability, distance to nearest surface water, depth of ground water and job opportunities. This implies that during decision-making, experts provide more emphasize to economic viability and social acceptability of the sites along with environmental sustainability. The attributes are mainly qualitative attributes describing socio-economic conditions; hence, fuzzy interpretation of those attributes can capture the ambiguity in the meaning of the attributes and overcome the imprecision and uncertainty of the related data more effectively.