Employing Data and Process Mining Techniques for Redundancy Detection and Analystics in Business Processes

Abstract

The detection, quantification, and scrutiny of redundancies within business processes is pivotal in achieving cost reduction, enhancing efficiency, and ensuring compliance. Redundancies, often leading to inefficiencies, result in escalated costs and errors, thereby detrimentally influencing an organization's overall performance. To counter these issues, data mining and process mining techniques offer promising solutions by identifying and analyzing process redundancies. Data mining, an approach devoted to the analysis of large datasets in order to discern patterns, relationships, and anomalies, has been applied to business processes. It provides insights into redundancies by scrutinizing process-related data, such as event logs, thereby revealing patterns in task executions that may indicate redundancies. In contrast, process mining employs event logs to generate a process model mirroring the actual execution of a process. This actual process model is subsequently contrasted against an expected process model, facilitating the identification of redundancies such as unnecessary activities or loops. Cluster analysis, a technique employed in both data mining and process mining, is exemplified for its capacity to group similar process instances or models based on specified attributes or characteristics. The application of cluster analysis aids in the identification of redundant process models or similar process patterns, thereby enabling further comparison and optimization.

Cross-industry frameworks for business process reengineering: Conceptual models and practical executions

“…Process analysis is a foundational step in the BPR journey, providing organizations with insights into their current workflows, identifying bottlenecks, and pinpointing areas for optimization (Trabelsi et al, 2023). The first step in process analysis involves mapping out the organization's current workflows in detail.…”