Title: Improving software modularity using many-objective NSGA-III algorithm by utilising many quality parameters

Authors: Naveen; Randeep Singh; Amit Rathee

Addresses: Department of Computer Science and Engineering, IEC University, Baddi, Solan, HP 174103, India ' Department of Computer Science and Engineering, IEC University, Baddi, Solan, HP 174103, India ' Department of Computer Science, Government College, Barota, Sonipat, HR 131001, India

Abstract: Software modularity aims to ensure high cohesion within modules and low coupling between them. However, prolonged maintenance often leads to modular degradation and increased system complexity. To address this, software re-modularisation is employed, though it remains a challenging task due to conflicting design objectives. This paper presents a novel re-modularisation approach that relocates classes to more appropriate modules using eight carefully formulated design quality criteria, ensuring both structural improvement and semantic preservation. A many-objective metaheuristic, NSGA-III, is applied to optimise these criteria via move class refactoring. The approach is evaluated on seven open-source software systems, demonstrating significant improvements in modular quality with minimal structural changes. Furthermore, the re-modularisation problem is formulated and assessed under single-objective, multi-objective, and many-objective scenarios using Bunch-GA, NSGA-II, and MOEA/D, respectively. The empirical results of the proposed technique proves effective and practical, particularly in maintenance contexts where full re-modularisation is cost- and/or time-prohibitive.

Keywords: software re-modularisation; SBSE; search-based software engineering; many-objective optimisation; semantic coherence; maintenance effort; NSGA-III; cohesion; coupling.

DOI: 10.1504/IJISCM.2025.154290

International Journal of Information Systems and Change Management, 2025 Vol.15 No.3, pp.231 - 264

Received: 26 Sep 2024
Accepted: 25 May 2025
Published online: 19 Jun 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article