Data-Driven Decision Support for Sustainable Engineering Change Management Using Prescriptive Process Mining

Abstract

The digital transformation of manufacturing enterprises necessitates more efficient engineering change management through data-driven decision support and process optimization. Although process mining has advanced considerably in recent years, most existing studies focus on descriptive or predictive process analysis, while prescriptive optimization integrated with sustainability considerations remains insufficiently explored. This study aims to develop and evaluate a prescriptive optimization model for Engineering Change Management based on process mining methods and Environmental, Social, and Governance considerations. The proposed model combines rule-based case routing, resource profiling, and risk-oriented alerts and is evaluated using a real-world event log from a manufacturing and engineering company. The methodological framework integrates process discovery using the Inductive Miner algorithm, conformance checking, performance and bottleneck analysis, deterministic rule-based case classification, resource profiling, and prescriptive analytics. The proposed model was evaluated through retrospective validation on a test dataset using both parametric and non-parametric statistical methods. The retrospective analysis estimated a 36.8% reduction in average active case processing time, from 6.31 to 3.98 h, while maintaining the modeled satisfaction level (p < 0.001). These results represent model-based estimates rather than effects observed after prospective implementation. The proposed approach extends existing process mining applications by combining prescriptive analytics with sustainability-oriented process governance. It provides a basis for integration into Product Lifecycle Management, Enterprise Resource Planning, and Engineering Change Management systems as a decision-support module, with the potential to improve operational efficiency and support sustainable manufacturing practices.