AN OPTIMIZATION METHOD FOR TASK RESOURCE RESCHEDULING IN AVIONICS SYSTEMS CONSIDERING PARTITIONS FAILURES

In Integrated Modular Avionics (IMA) systems, the partitioned resource-sharing model leads to multiple tasks competing for limited computational and communication resources, which may result in resource conflicts and task delays, especially under partition failures. Such failures introduce greater uncertainty and burden to task scheduling, leading to degraded system performance and potential safety risks. Therefore, optimizing task resource scheduling in the presence of partition failures, balancing task completion time with system load, is a critical issue that remains inadequately addressed in current research. This paper presents a multi-objective task allocation and scheduling reconstruction optimization method under partition failure conditions. First, a mathematical model is established to optimize task rescheduling, with the objectives of minimizing task completion time and balancing system load, while considering constraints such as task priority and resource thresholds. To solve this model, we adopt a hybrid algorithm combining Multi-Objective Particle Swarm Optimization (MOPSO) and Non-dominated Sorting Genetic Algorithm II (NSGA-II). MOPSO leverages the global search capability of particle swarms to explore the solution space, while NSGA-II ensures the diversity and distribution of the Pareto optimal solution set through non-dominated sorting and crowding distance mechanisms. By integrating the strengths of both algorithms, the proposed hybrid method efficiently solves complex optimization problems with multiple objectives. Finally, the effectiveness of the method is validated through a case study on an IMA system, yielding an optimized scheduling solution.