Object-centric process predictive analytics. (1st March 2023)

• Record Type: Journal Article
• Authors: Galanti, Riccardo; de Leoni, Massimiliano; Navarin, Nicolò; Marazzi, Alan
• Is Part Of: Expert systems with applications. Volume 213:Part C(2023)
• Publication Date: 2023-03-01
• Subjects: Predictive analytics -- Object-centric process -- Gradient boosting -- Artifact-centric process -- Process mining -- Explainable AI; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2022.119173
• Languages: English
• ISSNs: 0957-4174

Abstract

Abstract: Object-centric processes (also known as Artifact-centric processes) are implementations of a paradigm where an instance of one process is not executed in isolation but interacts with other instances of the same or other processes. Interactions take place through bridging events where instances exchange data. Object-centric processes are recently gaining popularity in academia and industry, because their nature is observed in many application scenarios. This poses significant challenges in predictive analytics due to the complex intricacy of the process instances that relate to each other via many-to-many associations. Existing research is unable to directly exploit the benefits of these interactions, thus limiting the prediction quality. This paper proposes an approach to incorporate the information about the object interactions into the predictive models. The approach is assessed on real-life object-centric process event data, using different Key Performance Indicators (KPIs). The results are compared with a naïve approach that overlooks the object interactions, thus illustrating the benefits of their use on the prediction quality. Highlights: Object-centric processes as a new emerging paradigm in industry. Traditional approaches are unable to exploit the benefits of object-interaction. Considering processes interactions when predicting improves prediction quality. Integration of Explainable AI techniques to highlight relevant features. Shapley values proved the importance of object-interaction and aggregated features.