Model discovery of parallel business processes using modified Heuristic Miner

Process Mining or Process Discovery is a method to automatically discover process models from event log data. Since the process discovery is gaining attention among researchers as well as practitioners, the quality of the resulted process models is required. Business process model contains sequence and parallel traces. Many algorithms have been employed for process discovery, such as Alpha, Alpha++ and Heuristic Miner. Both Alpha ++ and existing Heuristic Miner cannot discover processes containing parallel OR. In this paper we propose the modified Heuristic Miner which utilizes the threshold intervals to discover parallel XOR, AND, and OR. The threshold intervals are determined based on average dependency measure in dependency graph. The results show that the modified Heuristic Miner can discover OR split and join which cannot be discovered by Alpha ++ as well as the existing Heuristic Miner.