Logistics process

This paper investigates the application of Process Mining (PM) techniques to redesign and optimize logistics processes within an Iranian bank. The primary aim is to identify inefficiencies, bottlenecks, and process deviations using real-world event log data and to provide data-driven recommendations for process improvement. Data comprising 35,642 event reports related to 16,490 logistics process workflows were extracted from the bank's automation and correspondence systems over six months in 2022. Disco 2.14 was used for data analysis. Results revealed that only 3.6% of product demands conformed to the predefined process model, indicating high process variability and improvement potential. Analyses also showed the average process duration was 5.7 days, exceeding the bank's internal benchmark (three to five days), and the process fulfillment ratio was 83.3%, falling short of the desired target of 95%. Key inefficiencies identified included excessive waiting times for unfulfilled demands (averaging 315.7 days) and bottlenecks in the "Registering the purchase invoice" and "Registering the warehouse receipt" activities. Drawing on these findings, suggestions were proposed to optimize the procurement process, automate manual efforts, and improve alignment with the defined process model. This study contributes to the existing knowledge by providing an empirical case study of PM application in a specific context within the banking industry. The findings underscore the importance of monitoring and managing process conformance, as well as addressing excessive waiting times to improve customer satisfaction and operational efficiency. Limitations of this study include reliance on data from a single bank and a focus on logistics processes. Future research could focus on investigating root causes of process deviations, using PM for predictive analysis, and evaluating the impact of process improvements on key performance indicators.