An Adaptive Requirements Prioritization Technique for Large Scale Software Projects

Abstract

In the software development, varied decisions need to be made for ensuring the fulfillment newlineCustomers frequently seek a wide range of functions in large newlinesoftware projects, resulting in a vast set of requirements. Due to project timeframes and resource newlineconstraints, implementing all of the requirements is usually not possible. Setting priorities for a newlinelarge number of requirements takes time and is challenging. As a result, an organised method of newlineprioritizing and subsequently choosing the primary set of needs based on several factors is newlinerequired. Diverse techniques were obtainable to prioritize the requirements effectively. But, the newlineaccuracy and time consumption for Requirements Prioritization was not optimized. Also, during newlinethe large-scale Requirements Prioritization, multiple aspects such as time, cost are not considered. newlineTherefore, three novel methods are proposed for enhancing the performance of large-scale newlineRequirements Prioritization with better accuracy and lesser time. newlineFor large-scale software Requirements Prioritization, the Interdependency-aware Qubit newlineand BrownRoost Rank (IQ-BR) technique is created in the first phase. The Interdependency-aware newlineQubit Requirement Selection algorithm and BrownBoost Rank Requirements Prioritization newlineLearning model are used to create the IQ-BR approach. The Interdependency-aware Qubit newlineRequirement Selection algorithm, initially, solves the volatile and interdependencies issues among newlinerequirements in the prioritzation process. The requirements are then ranked using the BrownBoost newlineRank Requirements Prioritization Learning method based on the BrownBoost Rank function. newlineThe Pugh Trapezoidal Fuzzy and Gradient Reinforce Learning (PTF-GRL) method is newlineintroduced in the second phase for large-scale software requirement prioritizing in the shortest newlineperiod possible. PTF-GRL method is developed with the contribution of Pugh Decision-based newlineTrapezoidal Fuzzy Requirement Selection model and Gradient Orientation-based Reinforce newlineRequirements Prioritization model.