An Integrated Method for the Selection of Software Requirements Using Fuzzy TOPSIS and Genetic Algorithm

Software requirements selection is one of the key activities of the software development process. In this activity the requirements are selected based on their ranking order. Various methods have been developed for selecting the requirements using fuzzy logic, rough set theory, and Metaheuristic algorithms, etc. One of the limitations of the fuzzy based methods is that the membership functions of fuzzy numbers are manually decided by the decision makers. In these methods less attention is given to the automated generation of fuzzy membership function. To address this issue, this paper presents a method for the selection of software requirements in which genetic algorithm has been used for automatically generating the fuzzy numbers. In the proposed method, a random population of size 15 is initialized then the process of reproduction is started by using the selection, crossovers, and mutation operators of genetic algorithm for 23 generations. The best chromosome with the fitness value of 0.883333333 is selected from the last generation as an input in fuzzy TOPSIS method. The applicability of the proposed method is discussed by the requirements of an institute examination system