Process-Based Performance Analysis of Construction Tendering in Indonesia Using Real-World Event Logs

The tender process plays an important role in the pre-construction phase, but its implementation with the e-tendering system has not yet fully reduced its operational burden, especially regarding time inefficiencies and process complexity. Research on the operational performance analysis of the tender process using real execution data is very limited, even though the adoption of e-tendering has actually become widespread. Process mining has emerged as a promising approach to uncover the actual process flow using event data while also identifying inefficiencies in that flow. Using event log data from the Indonesian Ministry of Public Works, we analyzed 1,246 tenders that employed pre-qualification and post-qualification methods. Process discovery and performance analysis were conducted using Disco software to visualize workflows, assess the duration patterns of tender activities, and detect potential bottlenecks. The results show that the evaluation activity is the most critical and time-consuming stage in both tender methods. Moreover, differences in process structure and the presence of parallel activities further complicate the assessment of process performance more accurately. The findings emphasize the importance of specific and detailed process analysis for each method in e-tendering. The recommendations generated for performance improvement are to automate the evaluation activities, enhance the evaluators' competencies, and redesign the parallel workflow.