An overview of Process Mining and its applicability to complex, real-life scenarios

Abstract

Process models are considered to be the cornerstone of Business Process Management since they enable the documentation, analysis and improvement of business processes. Traditionally, these process models are subjective and depend strongly on perceptions.Process mining remedies the subjectivity issue by utilizing trails left by operational processes (event logs). This novel research field tightly couples event data and process models. Thus, it offers a comprehensive set of tools capable of providing fact-based insights into actual end-to-end processes.The different approaches to process mining and the plethora of tools it offers, make it applicable to a large spectrum of domains. However, blindly applying process mining to real-life, complex and highly flexible processes, proves to be inefficient. In fact, it is necessary to adapt the adopted approach to the particularities of each case study, and to frame its application by the understanding the problem at hand.This paper illustrates, through studies documented in scientific literature, that the applicability and efficiency of process mining in real-life scenarios relies on adapting the approach to their inherent challenges: Some processes might be too complex (e.g. healthcare processes), or too flexible in the absence of a system that enforces a normative model (e.g. loan applications, interaction with radiology equipment...). Event logs might present some challenges as well: Some might be extremely large, heterogenous, related to different processes or suffer from missing data. Plus, in some scenarios, we might need to account for some other aspects that traditional process mining doesn't consider (e.g. the financial aspect of processes).The papers reviewed, try to adapt to the challenges faced by defining a clear set of questions and combining process mining techniques with other analysis approaches. The papers are related to two application domains. The first is healthcare since medical procedures are too complex and sometimes involve different intertwined processes. The second is banking, finance and audit, since it is highly flexible, and in some cases, it requires the integration of the financial dimension.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 9 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.