Evaluating Predictive Power of Multivariate Event Logs for Business Process Performance Forecasting

Dewi J. Putri

Evaluating Predictive Power of Multivariate Event Logs for Business Process Performance Forecasting

Authors

Dewi J. Putri

Process Mining Specialist , Indonesia

Keywords:

Predictive Process Monitoring, Event Logs, LSTM, Temporal Convolutional Network, Gradient Boosting, Cycle-Time Prediction, SLA Forecasting, Calibration

Synopsis

We investigate how multivariate event logs—combining control-flow, temporal, resource, and case attributes—support short-horizon forecasting of process performance (cycle time, SLA breaches, throughput, and cost). We compare tree-based learners (Gradient Boosting), deep sequential models (LSTM/TCN), and simple statistical baselines under a unified pipeline with rolling-origin evaluation and calibration. On three public process-mining benchmarks, enriched multivariate logs consistently improve accuracy over control-flow-only data, with the best gains for early-stage predictions and SLA-breach classification. We discuss feature design, leakage pitfalls, and deployment considerations for real-time monitoring.

References

(1) van der Aalst, W. M. P. (2016). Process Mining: Data Science in Action (2nd ed.). Springer.

(2) Senderovich, A., Weidlich, M., Gal, A., & Mandelbaum, A. (2015). Queue mining for delay prediction in service processes. Information Systems, 53, 278–295.

(3) Rogge-Solti, A., & Weske, M. (2015). Prediction of remaining service execution time using stochastic Petri nets with arbitrary firing delays. Information Systems, 54, 1–14.

(4) Anugula Sethupathy, U.K. (2020). Cloud-Native Architectures for Real-Time Retail Inventory and Analytics Platforms. International Journal of Novel Research and Development, 5(6), 339–355. https://doi.org/10.56975/ijnrd.v5i6.309063

(5) Tax, N., Verenich, I., La Rosa, M., & Dumas, M. (2017). Predictive business process monitoring with LSTM neural networks. In CAiSE Workshops (pp. 477–492). Springer.

(6) Teinemaa, I., Dumas, M., Rosa, M. L., & Maggi, F. M. (2019). Outcome-oriented predictive process monitoring: Review and benchmark. IEEE Transactions on Knowledge and Data Engineering (TKDE), 31(1), 165–183.

(7) Evermann, J., Rehse, J. R., & Fettke, P. (2017). Predicting process behaviour using deep learning. Decision Support Systems, 100, 129–140.

(8) Camargo, M., Dumas, M., & González-Rojas, O. (2019). Learning accurate LSTM models of business processes. In Business Process Management (BPM) (pp. 286–302). Springer.

(9) Di Francescomarino, C., Dumas, M., Federici, M., Ghidini, C., Maggi, F. M., & Rizzi, W. (2018). Clustering-based predictive process monitoring. ACM Transactions on Knowledge Discovery from Data (TKDD), 12(2), 1–41.

(10) Mori, U., Mendiburu, A., Dasgupta, S., & Lozano, J. A. (2017). Early classification of time series: A review. Data Mining and Knowledge Discovery, 31(1), 138–172.

(11) Anugula Sethupathy, U.K. (2019). Real-time inventory visibility using event streaming and analytics in retail systems. International Journal of Novel Research and Development, 4(4), 23–33. https://doi.org/10.56975/ijnrd.v4i4.309064

(12) Yan, Z., Han, J., & Liu, H. (2019). Predictive process monitoring with uncertainty estimation. In Proceedings of the International Conference on Process Mining (ICPM) (pp. 289–296). IEEE.

(13) Kumar, A., Suri, K., & Kumar, P. (2020). Quantile regression for predictive monitoring of business processes. In Proceedings of the International Conference on Business Process Management Workshops (pp. 321–333). Springer.

(14) Márquez-Chamorro, A. E., Resinas, M., & Ruiz-Cortés, A. (2018). Predictive monitoring of business processes: A survey. IEEE Transactions on Services Computing, 11(6), 962–977.

(15) Pasquadibisceglie, V., Appice, A., Malerba, D., & Ceci, M. (2019). Using ensembles of predictors to improve the accuracy of predictive process monitoring. Proceedings of the IEEE International Conference on Business Informatics (CBI), 1, 136–145.

(16) Galanti, R., Comuzzi, M., & Ter Hofstede, A. H. M. (2021). Explainable predictive process monitoring. Decision Support Systems, 142, 113467.

Published

January 21, 2022

Series

Research paper

License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.