Poor coordination of interacting processes in the construction industry is largely responsible for a low workflow reliability of 30-60% that results in costly delays and repeated process coordination needs in this $700+ billion dollar US industry ($3.4 trillion globally). However, since the coordination needs are information intensive, new computer science practices supported by detailed understanding of the application domain holds the promise to improve practice. This project objectives are to generate new theory and understandings to enable rapid configuration of computerized links - process connectors - between firms to manage distributed process coordination tasks. A key limitation of existing models and methods for integrating distributed services (e.g., Web Services) is that they do not provide the necessary tools to rapidly configure specific implementations (particularly for complex services). This limits their usability in projects where firms assemble and disassemble in an ad hoc manner. To support the research objectives, two specific outcomes are anticipated: (1) Development of an ontology that can accommodate interactions between processes, in particular interactions where there are different levels of detail and constraints that must be propagated across processes. (2) Development of a specification language to flexibly model heterogeneous processes and process connectors in an executable format. To guide the development and testing the research team will collaborate with industry throughout the project. Specific focus will be placed on residential construction, a large $300+ billion subset of the US construction industry. Collaboration with industry provides directed opportunities to foster the educational objectives of the grant to involve students in the research and to develop examples and systems for classroom pedagogy.
While focusing on residential construction, the theory and methods developed under this proposal are applicable to a wide range of project industries such as commercial and industrial construction, shipbuilding, custom manufacturing, etc. More broadly, the proposed research will provide the foundation to enable dynamic process modeling by firms and individuals. It is expected that the proposed research will enable a step change in firms' capabilities to share process knowledge and coordinate their collective actions, allowing them to generate, evaluate and refine process alternatives more rapidly and reliably than is currently possible.
