A Token-Replay-Based Conformance Checking Method for Dataflow in IoT System

This paper proposes a method of conformance checking to diagnose differences between an IoT system and its actual use. IoT is a mechanism that connects the physical and cyber spaces, circulates data between them, and creates new value. Dataflow is a paramount perspective of IoT systems. Once the proposed method takes the dataflow model of the IoT system and the execution log of the system, it diagnoses and outputs the fitness of the log to the dataflow model based on the token-replay technique. It was implemented using PM4Py, a process mining package for Python. The implementation was applied to an IoT system for telemedicine and numerically showed the differences between the IoT system and its actual use.