PGTNet: A Process Graph Transformer Network for Remaining Time Prediction of Business Process Instances
This work addresses a domain-specific problem in business process management by improving prediction accuracy for process analysts, though it is incremental as it builds on existing deep learning methods.
The paper tackles the problem of predicting remaining time in business processes by introducing PGTNet, which transforms event logs into graphs and uses a transformer network, achieving state-of-the-art performance across 20 real-world datasets with notable gains in complex scenarios.
We present PGTNet, an approach that transforms event logs into graph datasets and leverages graph-oriented data for training Process Graph Transformer Networks to predict the remaining time of business process instances. PGTNet consistently outperforms state-of-the-art deep learning approaches across a diverse range of 20 publicly available real-world event logs. Notably, our approach is most promising for highly complex processes, where existing deep learning approaches encounter difficulties stemming from their limited ability to learn control-flow relationships among process activities and capture long-range dependencies. PGTNet addresses these challenges, while also being able to consider multiple process perspectives during the learning process.