Abstract
Leveraging historical process execution data to predict the evolution of an ongoing process is a challenging task in Business Process Management. In recent years, transformer-based models have shown promising results in multi-attribute sequence prediction, which refers to the task of forecasting not only the next activity in a process but also its associated attributes (e.g., timestamp, resource, and data values). However, although the accuracy of these models is generally high, there are specific situations in which background knowledge about the process can be beneficial. In particular, such knowledge can improve prediction quality for outlier executions in the training data and in scenarios where process changes occur due to concept drift in the test data. In this paper, we propose a principled method for integrating prior knowledge into an autoregressive transformer-based model for multi-attribute sequence prediction at test time. This knowledge is expressed using Multi-Perspective Declare, a process modeling language capable of describing process behavior from multiple perspectives, including control flow (i.e., the expected sequence of activities executed in the process), data flow (i.e., the data attributes associated with these activities during process execution), resources, and timestamps. We evaluate our method on real-life event logs and demonstrate that incorporating background knowledge consistently improves the performance of multi-attribute sequence prediction, especially in rare and exceptional situations.