Repairing Errors of Inconsistent Data Based on WFT-Nets

Data consistency is crucial for ensuring the reliability and accuracy of workflow systems. Inconsistent data caused by improper data operations, if not promptly repaired, can compromise the correctness of workflow systems. Existing methods for repairing inconsistent data mostly focus on either the process data or the persistent data. However, it is difficult for such methods to comprehensively deal with the errors of data inconsistency in workflow systems. Because workflow systems contain not only the process data, but also the persistent data. Based on Workflow net with tables (WFT-nets), this paper fully considers the above two types of data and their interaction to repair inconsistent data. The causes of inconsistent data are first analyzed in depth based on WFT-nets, with a specific focus on improper data element/table operations. In addition, these operations are further classified, and corresponding repair strategies are proposed. And then, in the WFT-net, transitions attached with such improper operations are called abnormal ones. To repair the errors of data inconsistency, the reachable configuration graph (RC-Graph) of the WFT-net is traversed to locate all possible abnormal transitions, and a repair algorithm is presented according to the sequence of “process data first, persistent data later”. Finally, a set of comparative experiments with WFD-nets is done to demonstrate the effectiveness of the proposed repairing method.