Situation-aware adaptation of choreographies—The DiStOPT approach

Abstract

The rise of the IoT and Industry 4.0 has increased the complexity of collaborating business processes, i. e., choreographies, as more partners and assets are involved. However, maintaining and executing business choreographies are complex tasks. Moreover, enabling robust and reliable execution is important, as failures or delays cause high costs among partners. For example, manufacturing companies usually depend on different suppliers, and it is crucial to be up-to-date about possible delays in shipments as this leads to delays in the manufacturing of their products. In this case, a choreography needs to be designed and operated in a way that it can adapt to cope with such problems. This requires i) timely recognition and tamper-resistent logging of problems that occur at each involved partner, which are referred to as situations in the scope of this article, and ii) an approach for a timely adaptation of choreographies based on occurring situations. Therefore, in this article, we introduce DiStOPT, an approach to i) model and recognize situations in a distributed and timely manner, and ii) model and execute situation-aware choreographies based on the recognized situations. The contributions are evaluated in a manufacturing scenario and validated by a prototypical implementation.