cs.LGNov 7, 2026

Differentiable Agent-Based Modeling for Multi-Echelon Supply Chains: Gradient-Based Optimization of Heterogeneous Policies at Scale

Authors: Bulent Soykan, Murtadha Alsayegh, Sean Mondesire, Ghaith Rabadi

Organizations: Institute for Simulation and Training, University of Central Florida, Orlando, USA · School of Modeling, Simulation, and Training, University of Central Florida, Orlando, USA

Abstract

Global supply chains rely on the interactions of thousands of heterogeneous agents, posing significant challenges for optimization. While traditional optimization often assumes homogeneity to ensure tractability, Reinforcement Learning struggles with sample inefficiency when scaling to high-dimensional state spaces. We propose a Differentiable Agent-Based Model for multi-echelon inventory management. By implementing stochastic dynamics via differentiable tensor operations, we enable direct gradient propagation through the simulation. We validate the framework by recovering theoretical Newsvendor optima without supervision. Crucially, we demonstrate ``Mass Personalization'' by optimizing 10,000 distinct agent policies simultaneously in seconds, achieving a 53% cost reduction compared to scalar baselines. Furthermore, we utilize gradient sensitivity analysis to generate causal heatmaps, accurately recovering inherent lead-time delays from the computation graph. This approach bridges the gap between simulation and optimization, enabling the efficient training of self-correcting Digital Twins.

Figures & tables

Explore similar work

CardsList
  1. OR-Transformer: Scaling Real-Time Decision-Making to 1,000 Items

    Sep 1, 2026Shuze Daniel Liu, David Simchi-Levi, Claire Chen +2Inventory ControlTransformer

  2. AgentGrad: Intervention-guided Prompt Optimization for Multi-Agent Systems

    Sep 8, 2026Jaewon Chu, Jinwoo Seo, Jaewon Cho +4Multi-Agent LLM SystemsMulti-Agent System Optimization

  3. Reliability and Effectiveness of Autonomous AI Agents in Supply Chain Management

    May 16, 2026Carol Xuan Long, David Simchi-Levi, Feng Zhu +3AI Agent ReliabilityMulti-Agent Systems