ReproTizer: a fully implemented software requirements prioritization tool

• Main Authors: Achimugu, P., Selamat, A., Ibrahim, R.
• Format: Conference or Workshop Item
• Published: Springer Verlag 2016
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://eprints.utm.my/id/eprint/73481/; https://www.scopus.com/inward/record.uri?eid=2-s2.0-84960854553&doi=10.1007%2f978-3-662-49619-0_5&partnerID=40&md5=dada60fbd5689231bc5ebbcc0ebc053f

Before software is developed, requirements are elicited. These requirements could be over-blown or under-estimated in a way that meeting the expectations of stakeholders becomes a challenge. To develop a software that precisely meets the expectations of stakeholders, elicited requirements need to be prioritized. When requirements are prioritized, contract breaches such as budget over-shoot, exceeding delivery time and missing out important requirements during implementation can be totally avoided. A number of techniques have been developed but these techniques do not addresses some of the crucial issues associated with real-time prioritization of software requirements such as computational complexities and high time consumption rate, inaccurate rank results, inability of dealing with uncertainties or missing weights of requirements, scalability problems and rank update issues. To address these problems, a tool known as ReproTizer (Requirements Prioritizer) is proposed to engender real-time prioritization of software requirements. ReproTizer consist of aWS (Weight Scale) which avails project stakeholders the ability to perceive the influence, different requirements weights may have on the final results. The WS combines a single relative weight decision matrices to determine the weight vectors of requirements with an aggregation operator (AO) which computes the global weights of requirements. The tool was tested for scalability, computational complexity, accuracy, time consumption and rank updates. Results of the performance evaluation showed that the tool is highly reliable (98.89% accuracy), scalable (prioritized over 1000 requirements), less time consumption and complexity ranging from 500–29,804 milliseconds (ms) of total prioritization time and able to automatically update ranks whenever changes occurs. Requirements prioritization, a multi-criteria decision making task is therefore an integral aspect of the requirements engineering phase of the development life cycle phases. It is used for software release planning and leads to the development of software systems based on the preferential requirements of stakeholders.