A study of the bi-objective next release problem

Abstract

One important issue addressed by software companies is to determine which features should be included in the next release of their products, in such a way that the highest possible number of customers get satisfied while entailing the minimum cost for the company. This problem is known as the Next Release Problem (NRP). Since minimizing the total cost of including new features into a software package and maximizing the total satisfaction of customers are contradictory objectives, the problem has a multi-objective nature. In this work, we apply three state-of-the-art multi-objective metaheuristics (two genetic algorithms, NSGA-II and MOCell, and one evolutionary strategy, PAES) for solving NRP. Our goal is twofold: on the one hand, we are interested in analyzing the results obtained by these metaheuristics over a benchmark composed of six academic problems plus a real world data set provided by Motorola; on the other hand, we want to provide insight about the solution to the problem. The obtained results show three different kinds of conclusions: NSGA-II is the technique computing the highest number of optimal 30 Empir Software Eng (2011) 16:29-60 solutions, MOCell provides the product manager with the widest range of different solutions, and PAES is the fastest technique (but with the least accurate results). Furthermore, we have observed that the best solutions found so far are composed of a high percentage of low-cost requirements and of those requirements that produce the largest satisfaction on the customers as well.

Search Based Software Engineering: Techniques, Taxonomy, Tutorial

“…This approach is described in more detail elsewhere [48] and was first proposed as the 'best' way to handle multiple objectives for all SBSE problems by Harman in 2007 [36]. Since then, there has been a rapid uptake of Pareto optimal SBSE to requirements [27,31,84,90,113], planning [5,98], design [17,88,95], coding [9,99], testing [33,35,47,76,90,96,107], and refactoring [52].…”

A Survey on Software Release Planning Models

“…We received 5 proposals of papers which were considered relevant in the context of this project. From these 5 references, we discarded 3 of them: Ruhe's book because it was not providing new models but a compilation of knowledge [10], a paper by Durillo et al [4] (see justification below) and a tool demo report since it was only 4 LNCS pages long [1]. The two remaining papers are listed in Table 3.…”

“…• Papers that present some experimentation (through benchmarks) on advanced algorithms that are applied to SRP (e.g., Durillo et al [4] performing a sensitivity analysis of three genetic algorithms or Luna et al [9] which surveys different metaheuristics for solving a multi-objective based formulation of the problem).…”

Multi-objective ant colony optimization for requirements selection

“…The decision maker will select a solution from this set according to personal criteria. The works of Saliu and Ruhe (2007), Zhang et al (2007), Finkelstein et al (2008Finkelstein et al ( , 2009 and Durillo et al (2009), study the NRP problem from the multi-objective point of view, either as an interplay between requirements and implementation constraints (Saliu and Ruhe, 2007) or considering multiple objectives as cost-value (Zhang et al, 2007) or different measures of fairness (Finkelstein et al, 2008(Finkelstein et al, , 2009, or applying several algorithms based on genetic inspiration as NSGA-II, MOCell and PAES (Durillo et al, 2009(Durillo et al, , 2011. However, to date, no ACO approach has been applied to multi-objective NRP.…”

“…In the case of NRP, the crossover and mutation methods are more specific and difficult than in other problems because we need to take into account the resources bound in order to obtain new valid individuals (see del Sagrado et al (2010a)). The works of Durillo et al (2011) andHarman (2010) show that NSGA-II can solve NRP offering a set of comparable solutions with those obtained by other metaheuristics.…”