Hybrid Particle Swarm Optimization method for process mining

This paper presents a bio-inspired hybrid method that extracts the optimal or a near-optimal business process model from an event log. The proposed method combines Particle Swarm Optimization with Simulated Annealing to optimize the mining process in terms of execution time and model quality. To evaluate a candidate business process model we use a fitness function that considers as evaluation criteria the model completeness and preciseness according to the cases in the event log. The bio-inspired hybrid method has been integrated in the PROM framework and evaluated on a set of event logs.