The WHY in Business Processes: Unification of Causal Process Models

TL;DR

This paper introduces a novel method to unify multiple causal process variants into a consistent model that captures causal-flow alternations, enhancing understanding of causal relationships in business process event logs.

Contribution

It presents a formal, proven method for unifying causal process models that explicitly represent causal-flow alternations across variants, addressing previous limitations.

Findings

Successfully unified multiple causal variants in datasets

Preserved correctness of original causal models

Open-source implementation available

Abstract

Causal reasoning is essential for business process interventions and improvement, requiring a clear understanding of causal relationships among activity execution times in an event log. Recent work introduced a method for discovering causal process models but lacked the ability to capture alternating causal conditions across multiple variants. This raises the challenges of handling missing values and expressing the alternating conditions among log splits when blending traces with varying activities. We propose a novel method to unify multiple causal process variants into a consistent model that preserves the correctness of the original causal models, while explicitly representing their causal-flow alternations. The method is formally defined, proved, evaluated on three open and two proprietary datasets, and released as an open-source implementation.