Solving Dual Sourcing Problems with Supply Mode Dependent Failure Rates

TL;DR

Downtime-prone assets incur enormous costs, motivating dual sourcing of spare parts from additive and conventional manufacturers with supply-mode–dependent failure rates. The authors develop an iterative heuristic (IWA) and three RL families (value-based, policy-based, actor-critic), augmented by Endogenously Parameterised Learning (EPL) to train a single policy applicable to many SKUs. In synthetic tests, DCL and its EPL variant achieve near-optimal performance (average gaps as low as $0.4\%$), while EPL dramatically reduces training time and enables cross-SKU generalisation; in a real energy-case, dual sourcing saves up to $22.6\%$ over single sourcing, with IWA and DCL performing strongly and EPL offering substantial computational savings. The results demonstrate scalable, robust inventory policies that explicitly account for endogenous demand caused by differing AM/CM failure rates and provide actionable guidance on when to prefer AM, CM, or dual sourcing in practice.

Abstract

This paper investigates dual sourcing problems with supply mode dependent failure rates, particularly relevant in managing spare parts for downtime-critical assets. To enhance resilience, businesses increasingly adopt dual sourcing strategies using both conventional and additive manufacturing techniques. This paper explores how these strategies can optimise sourcing by addressing variations in part properties and failure rates. A significant challenge is the distinct failure characteristics of parts produced by these methods, which influence future demand. To tackle this, we propose a new iterative heuristic and several reinforcement learning techniques combined with an endogenous parameterised learning (EPL) approach. This EPL approach - compatible with any learning method - allows a single policy to handle various input parameters for multiple items. In a stylised setting, our best policy achieves an average optimality gap of 0.4%. In a case study within the energy sector, our policies outperform the baseline in 91.1% of instances, yielding average cost savings up to 22.6%.

Figures (9)

• Figure 1: Failure process for an exemplary installed base.
• Figure 2: High-level overview of the algorithm.
• Figure 3: Overview of a general RL training procedure embedded in EPL.
• Figure 4: Cost analysis for Instance 3.
• Figure 5: Boxplots of savings of the policies compared to each other.