The objective of this paper is to conduct a controlled experiment with the three requirements prioritization techniques: Numerical Assignment (NA), Analytic Hierarchy Process (AHP) and Extensive Numerical Assignment (ENA), each based on ordinal, ratio and interval scales respectively. NA and AHP are widely used traditional requirements prioritization techniques. ENA is a novel technique introduced by the authors, which acknowledges the uncertain and incomplete nature of human judgment about requirements priorities, which are in turn uncertain guesses about the upcoming product. The aim of the experiment is to examine the three techniques using various objective and subjective measures like number of decisions, time consumption, ease of use, attractiveness, scalability and reprioritizability. The experiment was executed by prioritizing the requirements of a university website system with students as participants in the experiment. The results of the experiment proved that ENA transcends NA and AHP.

No abstract

Abstract.A good requirements prioritization technique is one which involves all the relevant stakeholders, provides them the flexibility of assessing a requirement by means of subjective and uncertain inputs, and aggregates these assessments to produce reliable requirements priorities. This paper addresses this by applying Interval Evidential Reasoning (IER) algorithm. Analytic Hierarchy Process (AHP) is employed to determine the varying contribution of stakeholders. The degree of satisfaction with the requirements priorities will be obtained by following the same procedure followed for requirements assessment and aggregation.

No abstract

No abstract