Archive for April 4th, 2011
Monday, April 4th, 2011
Due to the recent events in Japan, many companies, like Boeing, are concerned. The practice of keeping inventories lean in order to boost profits is proving to be problematic and Japanese suppliers are unable to keep up with demand. An article in Bloomberg Businessweek discusses the risks associated with just-in-time inventory. Smeal’s Susan Xu, professor of management science and supply chain management, shares her thoughts on what supply chains need to do to avoid these risks.
Today’s supply chains are operating under lean manufacturing principles which embrace low inventory and just-in-time (JIT) delivery. However, a tightly coupled supply chain removes inventory redundancy and makes the network as a whole more vulnerable in the face of disruption, as illustrated by the significant impact of Japan’s earthquake on part deliveries of Boeing’s 787 Dreamliner, which is already more than three years behind schedule due to perpetual production delays. To avoid the downside of JIT, companies need to increase safety stock to certain extend to isolate the impact of disruption. However, stockpiling inventory in anticipation of low probability, high impact disruptive events is also not a cost-effective strategy. Companies need to create reliable and cost efficient supply chains by using other risk mitigation strategies.
First, a move toward modularization and mass customization in recent years has allowed many companies to offer customers with a wide variety of products. These products tend to have flexible configurations and it is often the case that a part in shortage can be replaced by another part with supply. Flexibility in product design and product variety can be an effective strategy for supply chain risk management. In times of crisis, a company can dynamically influence consumer choices by offering reconfigured products or alternative products via dynamic pricing, incentives and promotions. Dell used this tactic effectively to manage its part shortage problem during the Taiwan earthquake in 1999. In contrast, Boeing was unable to alter the configuration of 787 Dreamliner due to its unique product design.
Second, relying on a single source for a critical part of a product can leave a company in a highly vulnerable position when the supply chain is interrupted. Boeing single-sourced most parts of 787 Dreamliner, as few suppliers were capable of producing these sophisticated parts. It has been shown in academic research and practice that risk diversification via multi-sourcing can help companies maintain a steady supply of critical parts. However, risk diversification may not be achieved by simply lining up multiple suppliers. There have been many real world examples, including ones illustrated in this article, where a single catastrophic event simultaneously degraded the capabilities of suppliers located in the same region. In a recent project, partly supported by my NSF grant “Risk Management of Supply Chain Networks with Dependent Disruptions,” we show that companies not only need to multi-source, but also need to select the suppliers who are not subject to common-cause disruptions (e.g., the suppliers located in different geographical locations).
Third, in the assembly phase of supply chain operations, the deliveries of subassemblies must be highly balanced and coordinated, to prevent the delay of the final assembly due to shortages of certain subassemblies. Boeing suffered from repeated delays because it outsourced subassembly operations to a large number of suppliers worldwide, many of them faced with their own interruptions such as natural disasters, labor unrest, and economic downturns. The results from our recent NSF project show that, while risk diversification is desirable at the component level, risk concentration is preferable at the assembly stage (e.g., selecting suppliers located in the same geographical region). In retrospect, if Boeing had decided to use its in-house capacity for the subassemblies of 787 Dreamliner, it might have been able to avoid some significant delays.