Optimal allocation of time redundancy and abort policies for multiple sequential tasks

Mission abort policies play a pivotal role in safeguarding system safety during mission execution, while time redundancy allocation becomes a critical prerequisite for balancing task safety and efficiency when multiple sequential tasks compete for shared time resources in practical engineering scenarios. Existing research on mission abort policies primarily focuses on single-task scenarios or assumes identical time windows for multi-task systems, failing to address the joint optimization of time redundancy allocation and abort policies for sequential multi-tasks. To fill this research gap, this paper proposes a novel integrated framework that combines time redundancy allocation and abort policies for systems executing multiple sequential tasks. Under the framework, each task is assigned a dedicated time window and a limited number of attempts, with subsequent task initiation independent of the success or abort outcomes of preceding tasks. A recursive method based on event transitions is employed to derive analytical expressions for the success probability of each task and the overall system failure probability, explicitly accounting for the interdependence of remaining execution time across tasks and distinguishing task failure causes. Furthermore, a joint optimization problem is formulated to determine the optimal time window and shock-based abort threshold for each task, aiming to minimize the system’s expected total loss. A genetic algorithm (GA) is utilized to solve the optimization problem, ensuring computational efficiency and accuracy. Finally, a detailed case study involving an unmanned aerial vehicle (UAV) performing sequential power grid inspection tasks validates the effectiveness of the proposed time redundancy allocation strategy and the practical applicability of the associated abort policies.