A review of business process mining: state‐of‐the‐art and future trends

Abstract

PurposeThis paper seeks to examine the area of business process mining, providing an overview of state‐of‐the‐art techniques. An outline of the main problems experienced in the practice of process mining is given along with reference to work that addresses the most challenging issues experienced in this field. This paper also aims to examine the application of soft computing techniques to process‐mining problems.Design/methodology/approachThis paper is based on a comprehensive review of literature covering more than 50 research papers. These papers are analysed to identify current trends and future research directions in the field.FindingsProcess‐mining techniques are now becoming available as graphical interface‐driven software tools, where flow diagram representations of processes may be manipulated as part of the mining task. A significant number of papers employ mining heuristics to aid in the task of process discovery. Soft computing algorithms are increasingly being investigated to aid the accuracy and speed of mining algorithms. Many papers exist that address common mining problems such as noise and mining loops. However, problems such as duplicate tasks, mining perspectives and delta analysis require further research.Originality/valueThe contribution of this paper is to provide a summary of the current trends in process‐mining practice and point out future research directions. A review of the work in this new and expanding area has been provided in the form of illustrative graphs and tables that identify the trends in this area. This is the most comprehensive and up‐to‐date review of business process‐mining literature.

A Principled Approach to the Analysis of Process Mining Algorithms

“…As a process is executed, the systems involved will record information in log files. Process mining [7] uses these logs to discover and analyse models of business processes.…”

A stage-based model for enabling decision support in process mining

“…For example, since business process management, workflow management, and process-aware information systems do fundamentally consider the concept of the process, process mining can be related and linked to them [89]. However, process mining can be classified as an extension and a descendant of data mining [85], since they share many technical aspects in terms of knowledge extraction.…”

“…Subsequently, process mining at the beginning of this century started to attract academics and industry, who became aware of the feasibility of process mining and its applications. Therefore, much research has been conducted and algorithms created to overcome the challenges [85,106]. However, process mining or business process mining is still a nascent discipline [11,64,92] and complicated issues have not been resolved in the existing studies [99].…”

“…Process mining can be classified as an extension and descendant of data mining [85]. The ideas that have been developed in the field of data mining are widely applicable for assessing the outcomes of process mining.…”

“…As a knowledge discovery process, it typically involves data cleaning, data integration, data selection, data transformation, pattern discovery, pattern evaluation and knowledge presentation" [34]. Tiwari et al [85] state that data mining works by extracting knowledge through identifying the patterns from huge data sets.…”