Random Forest Based Bus Operation States Classification Using Vehicle Sensor Data

In bus companies, it is important for an operation manager to grasp operation states of vehicles from a viewpoint of safety management and improving an operation efficiency. Currently, for allowing operation managers to grasp operation states of vehicles, drivers should record operation states by manually operating a recorder called "Digital-tachograph." However, operating the digital tachograph is a heavy burden to the driver. In addition, the records may have driver's human error. In order to solve these problems and to realize efficient operation, we propose a method for automatic classification of operation states using sensor data obtained from buses. We implemented a classifier using Random Forest with the sensor data. As a results of experiments, the correct answer rate was 0.92 or more in each condition unless it was irregular operation.