Collaborative Research: Multi-Product Assemble-to-Order Systems: Performance Optimization and Supply Chain Applications

Duke University, Durham NC

Investigators

Abstract

The objective of this research is to study the assemble-to-order (ATO) system, which is a hybrid of make-to-stock at the component (subassembly) level and assemble-to-order (or, configure-to-order) at the end-product (final assembly) level. The usual inventory-service tradeoff, a key factor in supply chain management, becomes even more prominent in the ATO system, as each customer order typically involves a large number of components, and the stockout of any component will cause a delay in supplying the order. Quantitative analysis of the ATO system, in general, is difficult as the production-inventory dynamics among the components are highly correlated, driven by a common demand stream. The research will focus on developing bounds and approximations based on various notions of stochastic order, and use these as surrogate performance measures, such that the resulting optimization problems possess special structures, which in turn lead to efficient algorithms. The ATO system is an appealing business process that provides both mass customization and quick response to order fulfillment, and hence plays an important role in a wide range of industries, internet commerce in particular. The research is expected to generate new knowledge and insights to the mathematical foundation and basic properties of ATO systems, and to enhance supply chain management technologies in the design and analysis of such systems.