Identifying challenges in maintenance planning for on-demand UAM fleets using agent-based simulations

Abstract

The novel aircraft architectures for Urban Air Mobility (UAM), combined with pure on-demand operations, mean a significant change in aircraft operation and maintenance compared to traditional airliners. Future flight missions and related variables such as the aircraft position or utilisation are unknown for on-demand operation. Consequently, existing methods to optimise aircraft assignment and maintenance planning cannot be transferred. This study examines the behaviour of an aircraft fleet in an on-demand UAM transport system regarding the interlinking between operation and maintenance. Initially, a potential maintenance schedule for UAM vehicles is deduced. A transport and maintenance simulation is introduced where aircraft are modelled as agents servicing a simple network. As aircraft reach their maintenance intervals, they transfer to one of the maintenance bases and compete for that resource. Since that competition can result in avoidable waiting times, the maintenance costs are extended by running costs for the bases and opportunity costs for missed revenue during these waiting periods. Opportunity costs are cost drivers. To reduce the waiting times, two operational approaches are examined: Extending the opening hours of the maintenance facilities and checking the aircraft earlier to reduce simultaneous maintenance demand. While an extension of operating hours reduces the overall maintenance costs, the adjustment of tasks is more effective to lower waiting times. Thus, an improved system needs to use a combined approach. That combination results in overall maintenance costs of approximately $ 58 per flight hour of which about seven percent account for the opportunity costs.