Joint repair sourcing and stocking policies for repairables using Erlang-A and Erlang-B queueing models

This research focuses on minimizing the life cycle cost of a fleet of aircraft. We consider two categories of repairable parts; upon failure of a first category part (No-Go part), its aircraft becomes non-operational but when a second category part (Go part) fails, the aircraft can still operate for a predetermined period of time before it becomes non-operational. In either case, to minimize aircraft downtime, the failed part has to be replaced with one from the good part inventory, returned from the repair facility or through exchange from a supplier-an emergency sourcing mechanism which is common in the airline industry. Motivated by the observation that a modern aircraft contains a significant fraction of Go parts (estimated at 50% of all repairable parts), we develop a strategic model to decide on stocking and sourcing policies using Erlang-A and Erlang-B queueing models. The suggested model provides an alternative to existing models that typically consider only failed parts that immediately cause a system to be non-operational, and do not consider an emergency sourcing mechanism. A realistic implementation of the model for a fleet of Boeing 737 aircraft, based on a list of 2,805 part types, demonstrates that significant cost savings may be achieved by explicitly modeling Go parts.