Joint modeling of loading and mission abort policies for systems operating in dynamic environments

Failures of safety-critical systems may cause huge economic losses and irretrievable disasters. The dynamic operating environment of such systems makes it more difficult to evaluate and control the risk of system failure. To enhance system safety, the existing literature mainly focuses on maintenance modeling and optimization, which can interrupt continuous mission execution. As an alternative, a mission can be aborted for quick response to high failure risk during mission execution prior to maintenance. In addition to mission abort, adjusting load is another effective way to control risk due to the dependence between load and failure risk. Improving load accelerates mission progress but increases system failure risk. Thus, an optimal load can be found to balance the risk of failure and the progress of the mission. This paper investigates the joint modeling of loading and mission abort policies for systems operating in dynamic environments. Information about dynamic environments, system degradation, and mission progress is integrated to guide loading and mission abort policies. The long-term average revenue rate of the system is derived and maximized by determining the optimal loads, system degradation and mission progress thresholds. Furthermore, two heuristic policies are proposed and numerical examples are given to illustrate the obtained results.

Automated summary

Failures of safety-critical systems can lead to significant economic losses and irreparable disasters. Evaluating and controlling the risk of system failure becomes more difficult in dynamic operating environments. This paper investigates the joint modeling of loading and mission abort policies to enhance system safety. Optimal loads, system degradation, and mission progress thresholds are determined to maximize the long-term average revenue rate of the system. Two heuristic policies are proposed and numerical examples are provided to illustrate the findings.