A Hybrid Approach for Requirements Prioritization Using Logarithmic Fuzzy Trapezoidal Approach (LFTA) and Artificial Neural Network (ANN)

Abstract

Requirements prioritization (RP) is a crucial phase of requirements engineering (RE) to rank the requirements as per their priority weight in software development process. The existing technique like FAHP (Fuzzy analytical hierarchy process) is a very suitable methodology for requirements prioritization used in the fuzzy background suffers from a number of limitations like FAHP technique does not grant the accurate priority as per the client hope, create many conflicts in priority vectors and may outcome in dissimilar conclusion which are unacceptable for a fuzzy pair-wise comparison matrix. Fuzzy preference approach (FPA) and extent analysis (EA) based nonlinear techniques are efficient but create many issues like ambiguity, time complexity, scalability, provides negative degree of membership function, inconsistency and generates many non uniqueness optimal solution in fuzzy environment. In this research a hybrid approach for requirements prioritization using ‘LFTA with ANN’ proposed to overcome these issues providing the most client fulfillment with all technical characteristics. The case study performed on MATLAB software and result observed for a real life example ‘college selection’ with three selection criteria’s that illustrates the decision making result for requirements prioritization is better than previous techniques with higher priority. The proposed hybrid approach is oriented to resolve the classical gaps and meet up the client fulfillment of decision making in real applications. A hybrid approach is examined on real-life assignment for students (‘selection of best college’ based on three criteria’s), with existent colleges and college selection criteria’s are discussed in the fuzzy AHP.