The Issues of Solving Staffing and Scheduling Problems in Software Development Projects

Abstract

Search-Based Software Engineering (SBSE) applies search-based optimization techniques in order to solve complex Software Engineering problems. In the recent years there has been a dramatic increase in the number of SBSE applications in areas such as Software Test, Requirements Engineering, and Project Planning. Our focus is on the analysis of the literature in Project Planning, specifically the researches conducted in software project scheduling and resource allocation. SBSE project scheduling and resource allocation solutions basically use optimization algorithms. Considering the results of a previous Systematic Literature Review, in this work, we analyze the issues of adopting these optimization algorithms in what is considered typical settings found in software development organizations. We found few evidence signaling that the expectations of software development organizations are being attended.

Benchmarking and Comparison of Software Project Human Resource Allocation Optimization Approaches

“…The comparison shows that few approaches among them all are satisfying the criteria adopted and capable for the illustrated aspects by this study as the one in [7]. Skills and productivity of resources found are the least aspects considered by the approaches used by Pixoto et al [4]. In addition, time and cost of software projects are the goals adopted by overwhelming majority of SSSP approaches.…”

“…The first study by Pixoto et al [4] evaluates the solution provided by SSSP approaches regarding their applicability in real-world software development projects. Criteria used by Pixoto et al to evaluate the description of solutions are usefulness, work compatibility, and ease of use attributes.…”

“…However, approaches can have different performance characteristics such as the accuracy of the allocation results or computational time required, characteristics that are critical for successful SSSP but very hard to determine without a systematic manner. Limited number of studies in this context [3,4] were published that compare SSSP approaches but neither of these studies performs an empirical evaluation of SSSP approaches using a unified basis and data set.…”

“…These studies have compared the approaches by a comprehensive survey [3] or systematic literature review [4] by extracting the text that describing the problem and solution of the approaches. Thus, these studies are more formally systematic literature review with comprehensive survey of wide software project management approaches.…”

“…The assignment of those resources in a software development department to projects and tasks within those projects is one of the most critical tasks for a project manager, with limited resources, dependent tasks, and available skillsets needing to be considered to achieve an optimal project delivery time. This problem of staffing and scheduling a software project (SSSP) in order to minimize the project completion time has been attracting researchers since the end of last century [1][2][3][4] and different optimization techniques have been used to address it in various incarnations [3,5,6]. These approaches typically consider specific attributes when optimizing the resource allocation such as task length, resource availability or skills, and the traversal of the optimization space is typically performed by using exact, heuristic, and meta-heuristic techniques in order to deal the NP-Complete nature of the allocation problem [3].…”