Solving the Aircraft Disassembly Scheduling Problem | AI Research

Solving the Aircraft Disassembly Scheduling Problem addresses the logistical challenge of dismantling end-of-life aircraft efficiently. As the aviation industry grows, so does the number of retired planes. While recycling these aircraft is environmentally necessary, it often yields thin profit margins. This research focuses on optimizing the disassembly process—specifically the removal of parts and materials—to reduce the time required to complete the work, thereby improving economic viability.

The Complexity of Disassembly

Dismantling an aircraft is a massive scheduling task involving up to 1,500 individual operations. These tasks are not straightforward; they are governed by a complex web of requirements. Technicians must possess specific certifications to handle valuable parts, and certain tasks must be performed in a strict sequence. Furthermore, the physical environment is constrained: some areas of the aircraft have limited space, restricting how many technicians can work simultaneously. Finally, the aircraft must remain balanced throughout the process to ensure safety, meaning the order of mass removal is strictly regulated.

Modeling the Solution

To solve this, the authors propose two mathematical approaches: a Constraint Programming (CP) model and a Mixed Integer Programming (MIP) model. These models are designed to handle the unique constraints of the problem, such as technician availability, skill requirements, and the physical balance of the aircraft. By treating the disassembly as a resource-constrained project scheduling problem, the models aim to minimize the "makespan," or the total time elapsed from the start of the first task to the completion of the last.

Testing with Real-World Data

The researchers tested their models using operational data from a Boeing B737, provided by an industrial partner. By running simulations on instances involving up to 1,450 tasks, the team evaluated how different constraints—such as spatial limitations and mass-balance requirements—impact the overall schedule. This approach allows companies to better plan their labor and resource allocation, turning a complex, manual-heavy process into a more predictable and profitable operation.

Key Considerations

The research highlights that while many scheduling problems exist in manufacturing, aircraft disassembly is unique because it requires the simultaneous management of skills, physical space, and structural balance. Unlike standard assembly line problems, the disassembly process is highly sensitive to the order in which parts are removed. The study provides a robust framework for handling these variables, offering a practical tool for the growing aircraft recycling market as it prepares for an expected increase in retired aircraft over the coming decade.