Model Integrated Computing-Based Software Design and Evolution

Among the most significant technological developments of the past 20 years are computer-based systems (CBSs), where functional, performance, and reliability requirements demand the tight integration of physical processes and information processing. Because complex component interactions exist in these systems, we must construct the software and its associated hardware such that they can evolve together. Model integrated computing (MIC) is an effective and efficient method for developing, maintaining, and evolving large-scale, domain-specific CBS applications. MIC is modelbased, allowing the synthesis of application programs from models created using customized, domain-specific, multi-aspect model integrated program synthesis (MIPS) environments. Integrated models explicitly represent dependencies and constraints among various design views. Because engineers can input design information at appropriate levels in the design hierarchy, and are freed from low-level implementation details, true end-user programmability is achieved. This paper discusses MIC technology and presents two large-scale MIC applications currently in use—the Saturn Site Production Flow (SSPF) system and the Integrated Test Information System (ITIS). The SSPF is a manufacturing execution system used by General Motors’ Saturn division to model, monitor, and analyze throughput characteristics of the plant. SSPF has been deployed in two Saturn plants and has helped Saturn to increase throughput in the Spring Hill, TN plant by nearly 10%. The ITIS, used to support Department of Defense aerospace testing at Arnold Engineering Development Center, integrates diverse sets of information from distributed, heterogeneous data sources into a seamless real-time, on-demand data system. The ITIS allows for rapid generation and customization of test information systems that track changing user requirements, allowing secure, uniform access to search metadata and retrieve test data from geographically distributed engineering teams. Using MIC technology, the system can respond to rapidly changing requirements for the interconnection of a wide variety of data sources and analyses.