Requirement Prioritization Using Analytical Network Process (ANP) (T-0670) (MFN 3603)

Abstract

In software product/project development, there may be few or many of the requirements to be fulfilled. Softwares are developed in increments, so it is difficult for the stakeholders or requirements engineers to implements all the requirements in a single increment. For this purpose we will prioritize requirements in order to deliver right product on right time. It seems from the research that none of the requirements prioritization techniques prioritize interdependent requirements and also performance of existing requirements techniques needs improvement regarding results. This leads to a research question directly, that is it will have to develop a requirements prioritization technique that support prioritization of interdependent requirements and then compare performance of newly proposed technique with existing requirements prioritization techniques to produce better results. In order to find answer to question literature review were carried out. A systematic literature review was carried out to find solution to the question through identifying, analyzing and interpreting all related studies. In literature review corresponding data related to requirements prioritization were critically reviewed, in order to make a strong background in software requirements prioritization. Most of the requirements prioritization techniques were studied, as how they prioritized software requirements. After developing a strong literature review a frame work or model were developed for our newly proposed requirements prioritization technique, that will prioritized both interdependent and not dependent requirements. Interdependent requirements are prioritized with newly proposed requirements prioritization technique (ANP) with example. So the answer to the question is given that, develop a requirements prioritization technique that prioritize both dependent and not dependent requirements. To compare the performance of newly proposed requirement prioritization technique (ANP) with existing requirements prioritization techniques (binary search tree, AHP, hierarchy AHP, spanning tree matrix, priority group and bubble sort) an experiment is conducted. A case study of local project (automation of Mobilink franchise system) is selected for the experiment. Then parameters/ measures selected for the experiments are (consistency indication, scale of measurement, interdependence, required number of decisions, total time consumption, time consumption per decision, ease of use, reliability of results and fault tolerance) on which requirements prioritization techniques are evaluated. The results of experiment show that ANP is the best requirements prioritization technique amongst all the requirements prioritization techniques.