Aircraft spare parts inventory management represents a critical challenge for airlines due to high capital costs, long repair lead times and strict service-level requirements. When the number of components increases, the resulting large-scale optimisation problem becomes difficult to solve using classical optimisation techniques, mainly because of the curse of dimensionality. The purpose of this paper is to investigate the applicability of a cooperative coevolutionary genetic algorithm to large-scale aircraft spare parts inventory optimisation and to evaluate its effectiveness in reducing total inventory cost while maintaining required service levels.
To address this problem, a cooperative coevolutionary genetic algorithm based on a divide-and-conquer strategy is developed. The proposed approach decomposes the original large-scale optimisation problem into smaller subproblems, which are optimised iteratively. The model minimises total inventory cost subject to service-level constraints derived from the minimum equipment list (MEL) criticality classifications. The methodology is tested using a real-world data set consisting of 940 aircraft spare parts, enabling direct comparison with classical Poisson-based approaches and linear programming (LP) methods.
The results demonstrate that the proposed cooperative coevolutionary genetic algorithm achieves a 24.10% cost reduction compared with the manufacturer-recommended policy while fully satisfying all service-level constraints. When evaluated against the LP results reported in the benchmark study under the same data set, the genetic algorithm-based solution produces a comparable total inventory cost and improved computational tractability for the large-scale integer decision space.
This study provides a novel application of cooperative coevolutionary genetic algorithms to large-scale aircraft spare parts inventory optimisation. The results highlight the potential of cooperative coevolution as a scalable and effective optimisation framework for complex inventory management problems in the aviation maintenance domain.
