Description

Coordinating complex process structures is a fundamental task for enterprises. Car development, for example, is based on long-running, concurrently executed, and dependent processes. The increased use of electronic systems for realizing advanced multimedia and safety functions, results in numerous processes to be coordinated for the integrated development, testing and release of these systems. The coordination of hundreds up to thousands processes forming large process structures, is a complex and error-prone task. In order to facilitate process management, IT support for the modeling and the flexible enactment of these process structures is urgently needed.

 In the COREPRO project, we have been elaborating fundamental requirements for the IT support of development processes (e.g., release management). As one first result, we have observed that these processes are often aligned according to the product to be developed. In particular, every subcomponent of the product has related processes that have to be mapped to the overall development process structure and to be synchronized according to the dependencies between the subcomponents.

The underlying idea of COREPRO is to utilize this information in order to enable the automated coordination data­-driven (i.e., product-driven) process structures. We enhance this idea in order to support the full process life cycle comprising modeling, enactment and change of process structure. In particular, we provide solutions for

• data-driven modeling and automated generation of large process structure
• data-driven enactment and dynamic adaptation of large process structures
• exception handling for data-driven process structures.

Our approach enables the automated, data-driven generation of large process structures and thereby reduces modeling efforts significantly. We further provide techniques for the dynamic adaptation of large process structures, which enable process structure changes on a high level of abstraction. Finally, we develop concepts for advanced exception handling to prevent inconsistencies when dynamically applying complex changes.