0018-9162/98/$10.00 © 1998 IEEE August 1998 35 Increasing Productivity at Saturn T o remain competitive, manufacturing enterprises must increase throughput and simultaneously reduce costs. This re- quires daily and long-term examination and analysis of a plant’s functions and operations. Using this data, an enterprise can identify production flow bottlenecks and analyze capacity and other factors, which in turn helps identify improve- ment opportunities. While such information is critical, it is often a challenge to obtain in a complex manu- facturing environment. The automotive industry provides some prime examples of complex manufacturing environments. At Saturn Corp., we turned to academic research— work involving model-integrated computing (MIC) 1 — as a framework to organize the diverse types of data our information system must deal with. In MIC, you build domain-specific models to capture information relevant to a system under design. Tools developed to work with MIC then automatically and quickly gen- erate applications. Because MIC depends on domain-specific models, we must first explain some basics about the process Saturn uses to build cars. THE BUSINESS OF BUILDING CARS Manufacturing automobiles involves many dis- parate operations: stamping, molding, fabrication, casting, machining, assembly, and so on. Saturn brings these operations together in an integrated manufac- turing system designed around just-in-time principles. The Saturn manufacturing site, in essence, is a net- work of processes and buffers. Processes represent the operations required to build a car. An operation, in turn, can involve actions (such as the casting, machining, and welding of car parts) or assemblies (such as for transmissions, engines, and the final car). Each process has associated measurements that indicate its productivity. These measurements— critical information in making business decisions— include entities such as cycle time, production count, work in process, and downtime. Buffers (or banks) lie between processes. A buffer holds parts and/or subassemblies produced by an upstream process for consumption by a downstream process. In different sections of the plant, buffers take on different physical forms: They may be parts con- tained in portable containers or on conveyors, for example. Despite their different physical manifesta- tions, buffers have common measurements that are pertinent to production: • bank count (the number of parts or subassemblies in the buffer), and • minimum and maximum buffer capacities. The interconnectivity of processes and buffers cap- tures the sequence of operations required to produce a car. It also defines the interdependence of processes on each other and on buffer capacities, and how one Computing Practices Earl Long Saturn Corp. Amit Misra Janos Sztipanovits Vanderbilt University Automaker Saturn drew from academic research to integrate information across disparate systems within its manufacturing site. The new information system gave team members across the plant access to real- time and historical throughput data. This information helped Saturn identify changes that led to a 10 percent increase in plant throughput. .