Optimal task aborting policy and component activation delay in consecutive multi-attempt missions

Recent reliability literature has extended the investigation of mission abort policies for single-attempt missions to multi-attempt missions, where the mission task may be attempted multiple times and each attempt may be performed by a single component or by multiple components simultaneously. To exploit the low operation cost achieved by the single component case and the high mission success probability (MSP) achieved by the multi-component case, this paper proposes a new model of multiple attempts performed consecutively with a prespecified activation delay by different components. Each attempt engages an operation phase and a return/rescue phase following the completion or abortion of the operation phase. During the two phases, components are exposed to random environments with different shock processes. A numerical algorithm is proposed to evaluate mission metrics including the MSP, expected number of lost components and expected mission losses (EML). An optimization problem is formulated and solved, which determines the activation delay as well as the attempt abort policy (composed of the number of shocks experienced and an operation time threshold triggering the abortion) to minimize the EML. The proposed model and influences of several model parameters are demonstrated by a case study of unmanned aerial vehicles performing a reconnaissance mission.