Forecasting future process models with concurrency aware causal nets
Concurrency-Aware Process Model Forecasting with Causal Nets
Machine LearningDatabases
Summary
Predicting how processes will behave in the future often ignores tasks that happen at the same time. The authors extend forecasting methods to capture these simultaneous actions by using a representation called causal nets, which show how different tasks relate and combine. They tested their approach on real data and found it predicts future process behavior nearly as well as models built from actual future observations. This method also performs better than older approaches that assume fixed process structures and struggle with complex changes.
What this means in practice
- •For business process analysts: Forecast evolving business workflows by predicting future models that include concurrent tasks for better process improvement planning.
- •For workflow automation teams: Anticipate changes in process concurrency to adjust automation logic before actual process changes occur, reducing downtime or errors.
Authors
Yongbo Yu, Jari Peeperkorn, Johannes De Smedt, Jochen De Weerdt
Abstract
Process model forecasting (PMF) aims to predict the process model that will characterize a future period, thereby providing a process-level view of how behavior is expected to evolve. Existing PMF methods, however, forecast directly-follows graphs, which cannot explicitly represent concurrency. We extend PMF to causal nets by forecasting time series of relation and binding counts and using these forecasts to reconstruct future process models with AND/XOR semantics. To evaluate the resulting models, we introduce a protocol that accounts for partial traces and constructs the workflow nets required for conformance checking. Experiments on four event logs show that the forecasted models achieve conformance levels close to those of models re-mined from observations in the corresponding future windows. They also outperform static discovery baselines, which retain high precision on the structurally stable log but exhibit substantial precision losses on the other three logs. Filtering infrequent bindings improves most conformance metrics, although it also removes much of the concurrent behavior captured by the models.