CAN Bus Joystick Data to Assess Operator Workload

Modern forest harvesting machinery generate an abundance of underutilised data in their control systems. The Controller Area Network (CAN) bus data stream offers the opportunity to investigate the operation of the machinery in detail while in real-world harvesting scenarios. This study uses CAN data to assess a component of operator workload in forwarder operations, by introducing a method to interpret forwarder joystick movements. The data was captured in a clearfell logging operation case study in Canterbury, New Zealand. The joystick data was then analysed to determine the time and number of operator input movements per load cycle (»grab«) totalling 418 grabs. This, combined with video analysis, identified independent variables that describe the grab cycle such as time of day, number of logs per grab, log grade, and the activities of »pencilling« (vertical drop of logs in grab to align large ends) and »dropping« (releasing logs from the grab before loading). Factors that significantly affected the operators' time taken to complete the grab and the number of required joystick movements included number of logs, pencilling and dropping. For example, the average load cycle was 18-seconds for four logs, and this increased by 6.1-seconds and 14.4-seconds per grab when pencilling or dropping, respectively. Average total joystick movements were ~108 per grab. This case study demonstrated that CAN bus data can be used to improve our understanding of the operation of harvesting equipment such as forwarders. An example use of the result is to share and compare this with the harvester operator (who crosscuts and sets out the logs for forwarder-collection) as the need for pencilling and dropping is a consequence of misalignment or debris caught in the piles. The method used also presents an opportunity for human factors research, particularly in operator fatigue management and training through the measurement of joystick movements with a genuine possibility of real-time performance feedback.