Leveraging process mining and event log enrichment in European public procurement analysis: a case study

This article explores the application of knowledge management and artificial intelligence techniques to refine the examination of administrative procedures, particularly within the realm of public procurement, to enhance the quality and efficiency of public administration. Key challenges in legal procedural studies include managing complexity, ensuring adherence to mandatory timelines, and maintaining regulatory compliance at every procedure stage. Automated process analysis provides a means to address these challenges by automatically extracting reliable of actual processes, offering valuable insights into how legal workflows are executed in practice—insights that are often difficult to obtain through conventional methods. Our re- search focuses on extracting pertinent information from extensive datasets, specifically legal event logs from public procurement procedures. We leverage process mining to analyze temporal events within administrative workflows and propose augmenting the corresponding logs using large language models for event and date extraction from legal texts. Legal experts oversee this methodology to ensure the successful integration of technology into the legal domain. We present a multinational case study applying this knowledge management framework to the Tender Electronic Daily dataset, spanning five European countries from 2016 to 2022. The findings demonstrate that techniques such as information extraction, the use of large language models, and process discovery significantly enhance legal knowledge management. Two domain experts evaluated the methodological approach and discussed the results, confirming its potential to improve compliance monitoring, control flow, and timeliness, thereby bolstering the efficiency of legal procedures.