Towards Automated Requirements Triage

Budgetary restrictions and time-to-market deadlines often require stakeholders to prioritize requirements and decide which ones to include in a given product release. Lack of an effective prioritization and triage process can lead to problems such as missed deadlines, disorganized development efforts, and late discovery of architecturally significant requirements. Existing prioritization techniques do not provide sufficient automation for large projects with hundreds of stakeholders and thousands of potentially conflicting requests and requirements. This paper therefore proposes an approach for automating a significant part of the prioritization process. The proposed method utilizes a probabilistic traceability model combined with a standard hierarchical clustering algorithm to cluster incoming stakeholder requests into hierarchical feature sets. Additional cross-cutting clusters are then generated to represent factors such as architecturally significant requirements or impacted business goals. Prioritization decisions are initially made at the feature level and then more critical requirements are promoted according to their relationships with the identified cross-cutting concerns. The approach is illustrated and evaluated through a case study applied to the requirements of the ice breaker system.