Deep reinforcement learning has been coined as a promising research avenue to solve sequential decisionmaking problems, especially if few is known about the optimal policy structure. We apply the proximal policy optimization algorithm to the intractable joint replenishment problem. We demonstrate how the algorithm approaches the optimal policy structure and outperforms two other heuristics. Its deployment in supply chain control towers can orchestrate and facilitate collaborative shipping in the Physical Internet.
A fter decades of offshoring production across the world, companies are rethinking their global networks.Local sourcing is receiving more attention, but it remains challenging to balance the offshore sourcing cost advantage against the increased inventories, because of its longer lead time, and against the cost and (volume) flexibility of each source's capacity. To guide strategic allocation in this global network decision, this paper establishes reasonably simple prescriptions that capture the key drivers. We adopt a conventional discrete-time inventory model with a linear control rule that smoothes orders and allows an exact and analytically tractable analysis of single-and dual-sourcing policies under normal demand. Distinguishing features of our model are that it captures each source's lead time, capacity cost, and flexibility to work overtime. We use Lagrange's inversion theorem to provide exact and simple square-root bound formulae for the strategic sourcing allocations and the value of dual sourcing. The formulae provide structural insight on the impact of financial, operational, and demand parameters, and a starting point for quantitative decision making. We investigate the robustness of our results by comparing the smoothing policy with existing single-and dual-sourcing models in a simulation study that relaxes model assumptions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.