Causal process mining for temporal deviations in event flows

Read full text Summarise with VIRAT
Authors: Fernando Montoya Cáceres
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

In the management of modern business processes, characterized by dynamism and competitiveness, temporal deviations constitute a critical challenge for operational efficiency. Although current analytical tools enable descriptive monitoring, they lack the ability to identify, interpret, and explain the causal relationships underlying such inefficiencies, limiting effective decision-making during the planning and execution phases. This research bridges the existing gap in the analysis of temporal deviations through the development of a methodology that integrates Process Mining with Causal Inference and Discovery. The proposed approach not only detects anomalies but also provides the system with interpretability and explainability regarding their root causes. This makes it possible to optimize role dynamics and enables monitoring based on structural causal relationships, offering a superior understanding of the process compared to traditional metrics. The methodology was validated through a rigorous dual approach: first, using synthetic data from the scientific community for controlled testing; and subsequently, through an in-field implementation within a real business environment. This procedure demonstrated both the theoretical soundness of the model and its practical applicability for causal analysis in industry. As applied research, this work consolidates a comprehensive approach that combines causal intelligence with process management to govern temporal deviations. The main contribution is a framework that facilitates early decision-making grounded in causal reasoning, enabling the evaluation of targeted actions and the ex-ante estimation of their impact on process improvement, covering the full cycle from planning to execution.

Fields of Study & Subjects

Business Process Modeling and Analysis Data Mining Algorithms and Applications Business and Management Studies

Publication details

Journal / source
CRIS
Publication date
25 March 2026
Language
EN