Bridging Theory and Practice: A Stochastic Learning-Optimization Model for Resilient Automotive Supply Chains

TL;DR

The paper targets resilience in JIT automotive supply chains by integrating Bayesian learning with stochastic inventory optimization in a two-echelon setting. The proposed closed-loop framework updates demand and disruption beliefs with conjugate priors and re-optimizes policy parameters every seven periods using a sample-average approach over 1000 posterior draws. In simulations across stationary, demand-shock, and disruption scenarios, the adaptive policy achieves a 7.4% cost reduction in stable environments and a 5.7% reduction during disruptions, but exhibits notable degradation under abrupt shocks due to conservative Bayesian updating. The work provides quantitative validation of AI-enhanced SCM, offers practical implementation guidance, and identifies critical boundary conditions and future research directions for robust, real-time resilience.

Abstract

Supply chain disruptions and volatile demand pose significant challenges to the UK automotive industry, which relies heavily on Just-In-Time (JIT) manufacturing. While qualitative studies highlight the potential of integrating Artificial Intelligence (AI) with traditional optimization, a formal, quantitative demonstration of this synergy is lacking. This paper introduces a novel stochastic learning-optimization framework that integrates Bayesian inference with inventory optimization for supply chain management (SCM). We model a two-echelon inventory system subject to stochastic demand and supply disruptions, comparing a traditional static optimization policy against an adaptive policy where Bayesian learning continuously updates parameter estimates to inform stochastic optimization. Our simulations over 365 periods across three operational scenarios demonstrate that the integrated approach achieves 7.4\% cost reduction in stable environments and 5.7\% improvement during supply disruptions, while revealing important limitations during sudden demand shocks due to the inherent conservatism of Bayesian updating. This work provides mathematical validation for practitioner observations and establishes a formal framework for understanding AI-driven supply chain resilience, while identifying critical boundary conditions for successful implementation.

Figures (5)

• Figure 1: Integrated Learning-Optimization Framework. The closed-loop system continuously adapts to changing conditions: Bayesian learning updates parameter estimates from observed data, which inform stochastic optimization to update inventory policy parameters, creating an adaptive control system.
• Figure 2: Bayesian parameter convergence in stationary environment. The learning algorithm successfully converges to true parameter values ($\lambda=10$, $\alpha=0.02$) within 50 periods.
• Figure 3: Policy parameter adaptation during demand shock scenario. The slow adaptation of Bayesian updating leads to inadequate inventory levels post-shock (period 183).
• Figure 4: Performance degradation during demand shock. Significant stockout costs accumulate during the adaptation period.
• Figure 5: Bayesian learning and inventory adaptation during supply disruption scenario. The system successfully learns increased disruption probability and adjusts safety stock accordingly.