Spare parts network design improved with graph neural network and optimization
Learning-Augmented Optimization for Strategic Two-Echelon Spare Parts Network Design
Machine Learning
Summary
Managing spare parts supply across many locations is complex and slow to optimize. The authors developed a method that uses machine learning, specifically graph neural networks, to predict the benefits of different network designs faster. They combine this with careful optimization techniques to avoid being overly optimistic. Tested on a large Amazon network, their approach achieved about 30% more savings while keeping nearly perfect service levels. This method helps find better designs more efficiently compared to traditional exact evaluations alone.
What this means in practice
- •For logistics planners: Design large spare parts networks for distribution centers with improved cost savings and reliable service.
- •For supply chain optimization teams: Accelerate evaluation of network design candidates by combining graph-based predictions with optimization.
Authors
Donato Maragno, Marco Caserta, Alberto Sinigaglia, Komlanvi Ametana, David Corredor Montenegro, Luca D'Angelo
Abstract
We study the strategic design of a two-echelon spare-parts inventory network where evaluating each candidate topology requires an expensive inventory optimization model. The design partitions hundreds of sites into feasible clusters and selects a central replenishment site for each cluster to reduce costs while maintaining service levels. Because the optimizer favors candidates with high predicted savings, it can exploit optimistic surrogate errors. We develop a conservative framework combining a graph neural network ensemble, variable neighborhood search, and set-partitioning recombination. The surrogate is trained on exact cluster evaluations, while a lower quantile of ensemble-predicted savings guides the search to limit optimism. Clusters found during the search are recombined through set partitioning using surrogate-based objective coefficients. The resulting network is evaluated with the exact inventory model, and only this evaluation is used to report performance. In a case study of 246 fulfillment centers in Amazon's North American network, the framework improves combined savings by 30.5% over an optimization baseline based entirely on exact cluster evaluations, while maintaining approximately 99.8% service across six independent replications. Under equal computational budgets, graph-surrogate-guided search achieves higher mean exact savings than a tabular alternative under both scoring schemes. Conservative scoring improves mean savings for both surrogate classes and reduces the share of final-network clusters overestimated by the graph surrogate from 68% to 28%. Predictive and ranking accuracy deteriorate among search-generated candidates with high surrogate scores, indicating that random holdout performance can incompletely characterize surrogate quality during optimization.