Hybrid Graph-Based Neighborhood Search and GA for ICS Crowd-Sourced Test Scheduling

As Industrial Control Systems (ICS) play an increasingly central role in critical infrastructure, ensuring the reliability and security of their software faces severe challenges. Cross-domain crowd-sourced testing, as an emerging testing paradigm, can leverage the diverse backgrounds and skills of testers from different domains to effectively enhance test coverage and defect detection capabilities. This paper constructs a scheduling model that can accurately characterize task-time heterogeneity and large-scale concurrency features, and proposes a resource scheduling algorithm that balances efficiency and precision, combines the graph-based neighborhood search algorithm with the elite genetic algorithm, and utilizes the advantages of the graph structure to identify a set of high-quality initial solutions, achieving the reasonable allocation of test resources and the efficient execution of tasks. Experiments show that the method proposed in this paper is superior to the traditional ones.