Enabling flexible task compositions, orders and granularities for knowledge-intensive business processes

During the last decades, many approaches and tools were developed that enable a flexible enactment of repetitive business processes. By contrast, knowledge-intensive processes (KIPs) have not 10 been effectively supported by process-aware information systems yet. In particular, KIPs are unpredictable, meaning that their structure (i.e. tasks, as well as the control and data flow between them) gradually evolves during process enactment based on user decisions and emerging data. Although considerable efforts have been 15 made regarding the support of KIPs, many of their flexibility requirements have not been effectively met so far. This paper proposes a data-centric and goal-driven approach that enables flexible KIP enactment with a dynamically evolving task structure and a varying task granularity according to the needs of the 20 knowledge workers. To better guide the latter in performing the tasks required to reach a process goal, the approach utilises artificial intelligence planning techniques. On one hand, the latter enable recommendations regarding the next tasks to be enacted for a given KIP instance with respect to goal achievement. On the 25 other, recommended tasks and task sequences may be dynamically adapted when mismatches between the digital process and the real-world KIP emerge. Overall, the approach provides high flexibility and proper user guidance during KIP enactment