A Method for Bottleneck Detection, Prediction, and Recommendation Using Process Mining Techniques

Abstract

Bottlenecks arise in many processes, often negatively impacting performance. Process mining can facilitate bottleneck analysis, but research has primarily focused on bottleneck detection and resolution, with limited attention given to the prediction of bottlenecks and recommendations for improving process performance. As a result, operational support for bottleneck resolution is often partially or not realized. The aim of this paper is to propose a method for Bottleneck Detection, Prediction, and Recommendation (BDPR) using process mining techniques to achieve operational support. A design science research methodology is adopted to design, develop, and demonstrate the BDPR method. A systematic literature review and a developed classification model provide theoretical support for the BDPR method and offer scholarly in the field of process mining a starting point for research. The BDPR method extends the utility of the classification model and aims to provide guidance to scholars and practitioners for assessing, selecting, evaluating, and implementing process mining techniques to realize operational support. A case study at a logistics service provider demonstrates the use of the proposed BDPR method.