Incentives and commonality in a decentralized multiproduct assembly system.

In this paper, we explore the impact of decentralized decision making on the behavior of multiproduct assembly systems. Specifically, we consider a system where three components (two product specific and one common) are used to produce two end products to satisfy stochastic customer demands. We study the system under both centralized and decentralized decision making. In the decentralized system, we prove that for any set of wholesale prices, there exists a unique Pareto-optimal equilibrium in the suppliers' capacity game. We show that the assembler's optimal wholesale prices lie in one of two regions--one leads to capacity imbalance and one does not. We use these results to derive insights regarding the inefficiencies that decentralization can cause in such systems. In particular, several of our findings indicate that outsourcing the management of component supplies may inhibit the use of operational hedging approaches for managing uncertainty.

Subject classifications: games, group decisions: noncooperative; inventory, production: multi-item, echelon, stage; uncertainty: stochastic.

Area of review: Manufacturing, Service, and Supply Chain Operations.

1. Introduction

Over the years, firms producing multiple finished products have explored a variety of ways to provide responsive service to customers, while keeping inventories as low as possible. One approach widely adopted in recent years has been the assemble-to-order approach, with Dell Computer being perhaps the best-known example. By using many of its components across multiple product lines, Dell can offer tremendous product variety while holding inventory of only a limited number of components. This use of component commonality is just one example of how operational flexibility can be leveraged to improve supply chain responsiveness, while avoiding excessive inventories. This approach and others based on the same principle have received significant attention in the academic literature in recent years, and have been successfully implemented in a number of companies. Some examples include operational hedging using flexible production capacity (e.g., Seagate Technologies; see Van Mieghem 1998b) and postponement (e.g., Hewlett-Packard; see Kopczak and Lee 1996).

In addition to their inventory/production strategy (achieving flexibility by stocking only components and producing finished goods only once an order is received), most assemble-to-order systems also involve some level of decentralization. For example, most of the component inventories that Dell uses in its assembly process are actually held by its suppliers. Similarly, in the automobile industry, component production capacity required to feed the final assembly plant is typically owned and controlled by the suppliers. Shortages in the supply of components can have a serious negative impact on the performance of supply chains. In 2000, basic memory chips and some high-frequency transistors used in cellular phone manufacturing were in short supply. Building capacity for some of these components can take up to 18 months, making it difficult for end-product assemblers to bring new products to market in a timely fashion (see Hilsenrath 2000).

While there has been significant research in the operation of assemble-to-order systems, very little attention has been paid to the impact of decentralized decision making in such systems. This paper focuses on that impact, particularly the impact of decentralization on the use of commonality and hedging strategies in such systems.