Skip to content
Conference

Forecast-driven Dual Sourcing Inventory Management using Simulation-Optimization

· IISE Annual Conference & Expo 2025 · 0 citations

Abstract

Supply chain networks today are complex webs of siloed actors operating globally, functioning within a highly volatile, uncertain, and disruption-prone environment. The COVID-19 pandemic and its repercussions have severely impacted all major industries with unforeseen shifts in demand patterns, supply unavailability, shipment delays, and supplier unreliability. Ensuring customer satisfaction and growth in such an environment requires consistent product availability, coordination among the actors, and capabilities to proactively assess and smartly react to the upcoming uncertainties.Dynamic supplier management has proven effective in supply chain management, as recognized by both researchers and practitioners. Dual sourcing, in particular, has been extensively studied, considering various uncertainties, constraints, inventory control policies, and sourcing mechanisms. However, a gap exists due to varying assumptions about network, demand, supplier, and supply transportation characteristics. Furthermore, there exists an industry need for dynamic dual-sourcing and, consequently, multi-sourcing strategies that are proven and applicable in real-world scenarios. Our work addresses this gap, towards facilitating seamless and quasi-autonomous day-to-day decision-making.Our contribution builds on our data-intensive study of dual-mode forecast-driven inventory management to address challenges faced by a major North American e-commerce-based furniture manufacturer with whom we collaborate in action research mode. We first propose a simulation-optimization based dynamic dual sourcing framework, and then compare the performance of three dual-sourcing policies: single-index, dual-index, and tailored base-surge. Finally, we investigate the sensitivity of the policies and their performance to supplier characteristics and constraints.

View source

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.