A Fair OR-ML Framework for Resource Substitution in Large-Scale Networks

2025-11-22

AI Summary

Large-scale logistics networks suffer from persistent node-level imbalance due to spatiotemporal mismatches between resource supply and demand. Method: This paper proposes a decentralized resource substitution optimization framework integrating operations research and machine learning. It is the first to explicitly incorporate fairness into resource substitution modeling; employs supervised learning to characterize dispatcher preferences; and introduces a dynamic κ-resource selection mechanism to prune the decision space, generating efficient, fair, and interpretable substitution solution sets within an integer programming framework. Contribution/Results: The method balances global coordination with individual preferences. Evaluated on a leading global express delivery company, it reduces model size by 80% and accelerates solving time by 90%, while preserving solution optimality. This significantly enhances both the efficiency and stakeholder acceptability of dynamic multi-resource allocation.

Technology Category

• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Mixed Discrete/Continuous Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Ensuring that the right resource is available at the right location and time remains a major challenge for organizations operating large-scale logistics networks. The challenge comes from uneven demand patterns and the resulting asymmetric flow of resources across the arcs, which create persistent imbalances at the network nodes. Resource substitution among multiple, potentially composite and interchangeable, resource types is a cost-effective way to mitigate these imbalances. This leads to the resource substitution problem, which aims at determining the minimum number of resource substitutions from an initial assignment to minimize the overall network imbalance. In decentralized settings, achieving globally coordinated solutions becomes even more difficult. When substitution entails costs, effective prescriptions must also incorporate fairness and account for the individual preferences of schedulers. This paper presents a generic framework that combines operations research (OR) and machine learning (ML) to enable fair resource substitution in large networks. The OR component models and solves the resource substitution problem under a fairness lens. The ML component leverages historical data to learn schedulers' preferences, guide intelligent exploration of the decision space, and enhance computational efficiency by dynamically selecting the top-$κ$ resources for each arc in the network. The framework produces a portfolio of high-quality solutions from which schedulers can select satisfactory trade-offs. The proposed framework is applied to the network of one of the largest package delivery companies in the world, which serves as the primary motivation for this research. Computational results demonstrate substantial improvements over state-of-the-art methods, including an 80% reduction in model size and a 90% decrease in execution time while preserving optimality.

Problem

• Optimizing resource substitution to minimize network imbalances in logistics
• Incorporating fairness and scheduler preferences in decentralized resource allocation
• Developing OR-ML framework for efficient resource substitution in large networks

Innovation

• Combines operations research and machine learning
• Learns scheduler preferences from historical data
• Dynamically selects top resources for efficiency