Enhancing Process Mining on Sparse Event Logs: A Simulation-Driven Academic Case Study

Many business processes involve complex, partly automated workflows, with limited timestamp data in the process event log. Sparse event logs limit the opportunity for process analysis and optimization. This paper describes the use of Discrete Event Simulation (DES) to model such processes and to augment sparse event logs with richer, synthetic data, enabling the application of process mining techniques for process enhancement. The approach is illustrated by application to an administrative process in an academic institution, namely an assessment extension request process. The study combines data-driven and knowledge-based approaches to understanding and modeling the existing process. Preliminary data analysis is used to obtain high-level process insights. The paper shows how this lays the foundation for the use of domain knowledge to construct and verify a discrete event simulation model. By integrating process mining with simulation, research can provide actionable insights to improve the efficiency and effectiveness of the current assessment extension request process. Observations are made on how this approach can be applied to other business processes with sparse event logs.