Prediction of remaining time to a business processes conclusion using deep learning

• Ronildo Oliveira da Silva Federal University of Ceará (UFC)
• Regis Pires Magalhães Federal University of Ceará (UFC)
• Lívia Almada Cruz Federal University of Ceará (UFC)
• Criston Pereira de Souza Federal University of Ceará (UFC)
• Davi Romero de Vasconcelos Federal University of Ceará (UFC)
• José Antônio Fernandes de Macêdo Federal University of Ceará (UFC)

DOI: https://doi.org/10.5753/sbbd.2023.231707

Abstract

This work aims to predict the remaining time to complete a business process instance using deep learning models. Efficiently predicting the remaining time to complete a process instance contributes to preventing uncertain waits, discovering bottlenecks in a processes, and assist alert systems. This paper uses deep learning architectures to predict the remaining time to conclusion a business process, which surpass state-of-the-art solutions. The architectures used are validated with two sets of public data, facilitating the reproducibility of the experiments.

Keywords: Prediction of remaining time to conclusion, Business Process, Deep Learning

References

Bukhsh, Z. A., Saeed, A., and Dijkman, R. M. (2021). Processtransformer: Predictive business process monitoring with transformer network. arXiv preprint arXiv:2104.00721.

Castro, M. A., Souza Jr, N., Escovedo, T., Lopes, H., and Kalinowski, M. (2022). Mineração de processos aplicada à auditoria interna na marinha do brasil. In Anais do XXXVII Simpósio Brasileiro de Bancos de Dados, pages 241–253. SBC.

Kalenkova, A., Ageev, A., Lomazova, I. A., and van der Aalst, W. M. (2017). E-government services: Comparing real and expected user behavior. In International Conference on Business Process Management, pages 484–496. Springer.

Mello, P., Santoro, F., and Revoredo, K. (2020). It incident solving domain experiment on business process failure prediction. Journal of Information and Data Management, 11(1).

Mello, P. O., Revoredo, K., and Santoro, F. (2019). Business process failure prediction: a case study. In Anais do VII Symposium on Knowledge Discovery, Mining and Learning, pages 89–96. SBC.

Navarin, N., Vincenzi, B., Polato, M., and Sperduti, A. (2017). Lstm networks for data-aware remaining time prediction of business process instances. In 2017 IEEE Symposium Series on Computational Intelligence (SSCI), pages 1–7. IEEE.

Park, G. and Song, M. (2020). Predicting performances in business processes using deep neural networks. Decision Support Systems, 129:113191.

Paschek, D., Luminosu, C. T., and Draghici, A. (2017). Automated business process management–in times of digital transformation using machine learning or artificial intelligence. In MATEC Web of Conferences, volume 121, page 04007. EDP Sciences.

Polato, M. (2017). Dataset belonging to the help desk log of an italian company.

Ponsard, C. and Darimont, R. (2019). Towards goal-oriented analysis and redesign of bpmn models. In MODELSWARD, pages 527–533.

Reijers, H. A. (2021). Business process management: The evolution of a discipline. Computers in Industry, 126:103404.

Stjepić, A.-M., Ivančić, L., and Vugec, D. S. (2020). Mastering digital transformation through business process management: Investigating alignments, goals, orchestration, and roles. Journal of entrepreneurship, management and innovation, 16(1):41–74.

Tax, N., Verenich, I., Rosa, M. L., and Dumas, M. (2017). Predictive business process monitoring with lstm neural networks. In International Conference on Advanced Information Systems Engineering, pages 477–492. Springer.

van Dongen, B. (2012). Bpi challenge 2012.

Venkateswaran, P., Muthusamy, V., Isahagian, V., and Venkatasubramanian, N. (2021). Robust and generalizable predictive models for business processes. In Business Process Management: 19th International Conference, BPM 2021, Rome, Italy, September 06–10, 2021, Proceedings, pages 105–122. Springer.

Venugopal, I., Töllich, J., Fairbank, M., and Scherp, A. (2021). A comparison of deep-learning methods for analysing and predicting business processes. In 2021 International Joint Conference on Neural Networks (IJCNN), pages 1–8. IEEE.