Value-Based Requirements Prioritization: Usage Experiences

摘要

There are usually more requirements than feasible given budget and schedule constraints. Thus it's important to select the most valuable ones for implementation in order to ensure the delivery of a high value system. Simple prioritization approaches like 1- 10 ranking or MoSCoW lead to numerous ties requiring one to repeat the process for the tied items. In a previous study [1] we analyzed 17 different prioritization frameworks that could be used to perform value-based requirements prioritization (VBRP). TheTechnique of Ordered Preference by Similarity to Ideal Solution(TOPSIS) was selected as the framework of choice, as a result of the analysis. TOPSIS was deployed for use by a premier IT company in India. In this paper we present our experiences in using a decision analysis framework like TOPSIS to perform VBRP. We have seen successful applications of using such a decision analysis framework for prioritizing test-cases, scoping to select the most valuable requirements for a release or product roadmap, value focused resource allocation and value-oriented product customization. We aim to provide evidence showing how VBRP and decision analysis frameworks can help channelize various systems engineering activities by focusing on the most valuable items first. Our experience in deploying the framework for various prioritization activities provides strong evidence making explicit the need for VBRP as a first-class citizen for value-based planning, implementation and delivery of systems and software applications.