Experience Report on the Development of an Application of the Genetic Algorithm Heuristic for Solving the Next Version Problem

• Thiago D. de C. Q. Gama Faculdade de Tecnologia SENAI
• Celso G. C. Junior UFG
• Gilmar T. Junior UEG
• Ana Clara A. G. da Silva UEG
• Ricardo Manuel G. Martins PUC Goiás

DOI: https://doi.org/10.5753/erigo.2023.237260

Abstract

This study investigates the application of the Genetic Algorithm heuristic to solve the complex Next Version Problem in software engineering. The problem involves selecting and prioritizing features for the next version of software. The study adapts the Genetic Algorithm to address this issue, demonstrating its effectiveness compared to other configurations through experiments on real data sets. The results indicate that this approach generates efficient and balanced solutions for project objectives, offering valuable insights for requirements management in software development projects.

References

Elvassore, V. (2016). "Experimenting with generic algorithms to resolve the next release problem". Dissertação de Mestrado. Universitat Politècnica de Catalunya.

Goldberg, D. E. (1989). Genetic Algorithms in Search, Optimization, and Machine Learning. Addison Wesley.

Gorschek, T., Wohlin, C., & Östberg, O. (2006). "A staged model for systematic review". In Empirical Software Engineering, 11(4), 543-562.

Niu, N., Huang, C., & Jin, H. (2008). "An evolutionary algorithm for feature selection based on mutual information". In Information Sciences, 178(14), 2799-2813.

Sommerville, I. (2011). Software Engineering (9th ed.). Addison Wesley.