Analyzing Administrative Process Sequences Using PM4Py: A Case Study in an Albanian Municipality

• Published in: European Journal of Computer Science and Information Technology Vol. 13; no. 51; pp. 119 - 137
• Main Authors: Kosta, Anxhela, Lili, Ilma, Xhina, Endrit
• Format: Journal Article
• Language: English
• Published: 26.08.2025
• ISSN: 2054-0957; 2054-0965
• DOI: 10.37745/ejcsit.2013/vol13n51119137

Inefficiency, delays, and procedural redundancies are often the hindering elements in administrative processes in local government institutions, especially in municipalities in Albania. When citizens make a specific request to the municipality for a specific process, there are many cases when they never receive a response and do not know what happens to their request, at what step it is, what is missing, when it will be approved, etc. For this reason, this paper presents a focused approach to analyze the sequences of work processes that occur on the Albania administrative side with the aim of identifying inefficiencies and opportunities for improvement through Process Mining. Initially, it was necessary to construct a real dataset of events capturing the execution of municipal services. We developed a SQL Server database structured in five main tables to generate event logs, comprising approximately 366,000 cases, 620,205 events, and 33 unique activities across multiple municipal workflows. Through Pm4Py we analyze and visualize all work process sequences. The goal was to identify inefficiencies and understand how we can optimize the workflow sequences. Through Heuristic and Alpha miner algorithms we analyse and model the processes to build a structured model in the form of a graph such as Petri Net and Heuristic, providing a clear and consistent visualization. Petri facilitates the identification of delays, interruptions and sequences of repeated actions. Pm4py plays a crucial role in accelerating and increasing the efficiency of model building at this stage of the process. Future work, the study will focus on clustering the SNA(social network analysis) to identify relevant anomalies for each process. This clustering analysis aims to identify and classify potential anomalies in the structure and performance of local administrative workflows, providing insights for further optimization efforts.