Abstract
In today’s time and budget intensive software development market, quick delivery is the basicmotive of teams. Software development teams strive to gain customer satisfaction by allpossible means. Requirements prioritization is the most challenging customer input dependenttask in the software development life cycle that decides the fate of a project. Selection of awell-suited requirements prioritization technique may result in customer satisfaction and ontime delivery time. Literature reports on many requirements prioritization techniques inpractice. However, each has its own features that can outperform the rest for a certain case.Therefore, this research is conducted to empirically evaluate the existing techniques in termsof certain quality measures (i.e., accuracy, efficiency, and scalability). The selected techniquesare evaluated for the small, medium and large scale of requirements sets. For that, we selectedfive existing techniques that are multi-criteria-decision-making techniques and have userinvolvement (i.e., Analytical Hieratical Process (AHP), Analytical Network Process (ANP),FuzzyAHP, FuzzyANP and Interactive Genetic Algorithm (IGA)). The experimental resultsshowed that among the five selected techniques, FuzzyAHP is the most efficient and accuratetechnique for the large dataset of requirements.
Highlights
There are a lot of requirements prioritization techniques for both manual and automated prioritization
As the data is not normal and for checking the homogeneity of variance, we applied the “Ansari” and “Mood” test which was true for Analytical hieratical Process (AHP), Analytical Network Process (ANP), and Interactive Genetic Algorithm (IGA) while rejecting for FAHP and FANP
We identified that the fuzzy RP techniques have a significant difference as compared to AHP, ANP, and IGA in terms of accuracy
Summary
There are a lot of requirements prioritization techniques for both manual and automated prioritization. AI-based prioritization techniques are proposed using different algorithms that are fuzzy logic and evolutionary algorithms (Agrawal, Singh, & Sharma, 2016; Jawale, Patnaik, & Bhole, 2017; Tonella et al, 2010). A very limited focus on the empirical validation of these techniques. The empirically evaluated techniques still have some limitation that is a lack of focus on scalability, easy to use and learnability (Pergher et al, 2013). There are still some studies that do not fully address the scalability, easy to use and learnability (Pergher et al, 2013), especially the recent AI proposed Algorithm for requirements prioritization (Ahuja & Batra, 2018; Jawale et al, 2017; Tonella et al, 2010)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
