Skill Evaluation of Human Operators in Partly Automated Mobile Working Machines

The performance of a mobile working machine is strongly dependent on the skills of the operator. It is commonly accepted that among operators, productivity may vary tens of percents. This research considers the human skill evaluation in working machines during normal work. A general framework for skill evaluation via task sequence recognition by a hidden Markov model (HMM) is described. The definitions of skill metrics, based on task resource consumption, task completion frequency, task difficulty, and ability to plan and make decisions, are given and justified through an example. The skill evaluation methods are illustrated by utilizing them in two industrial applications.