Article: Automated simulation testing for complex software environments using multi-agent reinforcement learning Journal: International Journal of Simulation and Process Modelling (IJSPM) 2026 Vol.23 No.1 pp.1 - 10 Abstract: The growing complexity of software systems poses major challenges for automated testing in continuous integration and continuous delivery pipelines. This paper proposes multi-agent-based software automated testing, a multi-agent reinforcement learning framework that models testing tasks as a decentralised partially observable Markov decision process. Using the Q-network mixing algorithm, the system enables coordinated decisions across testing agents. Evaluation on a TravisTorrent-based simulation environment shows multi-agent-based software automated testing achieves a 95.2% defect detection rate - showing a 4.7% improvement over the best multi-agent baseline - while reducing test execution time to 70% of conventional rule-based scheduling. Compared with multi-agent deep deterministic policy gradient, it demonstrates a large effect size (Cohen's d = 0.89) in defect detection. These results demonstrate the framework's effectiveness in improving testing quality and efficiency, offering a viable solution for intelligent test automation in complex software environments. Inderscience Publishers - linking academia, business and industry through research

Title: Automated simulation testing for complex software environments using multi-agent reinforcement learning

Authors: Qiuming Zhang; Jing Luo

Addresses: School of Computer Science, China University of Geosciences, Wuhan, 430074, China ' Library and Archives, Wuhan Vocational College of Software and Engineering, Wuhan, 430205, China

Abstract: The growing complexity of software systems poses major challenges for automated testing in continuous integration and continuous delivery pipelines. This paper proposes multi-agent-based software automated testing, a multi-agent reinforcement learning framework that models testing tasks as a decentralised partially observable Markov decision process. Using the Q-network mixing algorithm, the system enables coordinated decisions across testing agents. Evaluation on a TravisTorrent-based simulation environment shows multi-agent-based software automated testing achieves a 95.2% defect detection rate - showing a 4.7% improvement over the best multi-agent baseline - while reducing test execution time to 70% of conventional rule-based scheduling. Compared with multi-agent deep deterministic policy gradient, it demonstrates a large effect size (Cohen's d = 0.89) in defect detection. These results demonstrate the framework's effectiveness in improving testing quality and efficiency, offering a viable solution for intelligent test automation in complex software environments.

Keywords: multi-agent reinforcement learning; MARL; automated testing; continuous integration and continuous delivery; CI/CD; Jenkins.

DOI: 10.1504/IJSPM.2026.152088

International Journal of Simulation and Process Modelling, 2026 Vol.23 No.1, pp.1 - 10

Received: 14 Oct 2025
Accepted: 18 Nov 2025
Published online: 06 Mar 2026 *

Title: Automated simulation testing for complex software environments using multi-agent reinforcement learning

Keep up-to-date