A Semi-Markov Reliability and Digital Twin Framework for Mission Readiness Evaluation of Repairable Tethered HAPS

High-Altitude Platform Stations (HAPS) are emerging as a key component of future aerial communication networks, where sustained platform availability is essential for continuous mission execution and network service continuity. This paper presents a semi-Markov reliability framework for mission readiness evaluation of a repairable tethered HAPS equipped with hot standby support. The propulsion subsystem is modeled as a repairable k-out-of-n redundant system, while the semi-Markov formulation relaxes the memoryless assumption and provides a more realistic representation of system failure and repair dynamics than conventional Markov models. Exact analytical expressions are derived for steady-state availability, reliability, mean time to failure, and failure rate, followed by an investigation of repair-time variability and the reliability benefits of hot standby redundancy. To support operational decision-making, availability-based mission readiness criteria are developed by determining admissible repair regions, minimum engine lifetime requirements, and maximum allowable repair durations for prescribed readiness levels. Furthermore, a Bayesian digital twin framework is introduced to update engine failure intensity using operational observations and dynamically re-evaluate availability, mission readiness, repair thresholds, and service continuity throughout the mission. Numerical results demonstrate that the proposed framework effectively integrates analytical reliability modeling with adaptive online updating, providing a practical approach for mission readiness assessment and resilient HAPS-assisted communication services.