Main contents


Optimizing the work-efficiency of machine operators


In man-machine processes, such as working machines, the human operator plays a significant role in terms of productivity, fuel economy, and quality of the end product. Therefore, the overall performance of the system depends, on one hand, on the technical performance and on the other hand on the performance of man-machine interaction (HMI). The focus of OTTO-project is on the latter. The performance of the HMI includes the performance and skills of the human operator and the performance of the control system responding to the operator's commands. It has been recognized that the performance variations between two operators working with similar machines can be tens of percents. In order to utilize the whole capability of the machine both the technical performance and the human skills need to be optimized.

This project aims to develop methods for improving the work-efficiency of machine operators.


In this project the objectives are (i) to develop metrics to evaluate the skills of the operators during normal work, (ii) to develop methods to recognize the working style and skill improvement areas, and (iii) to develop methods to give educative feedback to the operator to achieve better performance. The research results are applied in working machines such as container cranes and forestry machinery.


The research methods include advanced signal processing methods such as hidden Markov models (HMM), clustering methods, fuzzy inference systems (FIS), and dimension reduction methods. The theoretical methods around various disciplines such as systems technology, control engineering, data mining, signal processing, estimation theory, statistic etc., are utilized in a novel and open-minded manner yet retaining the scientific criticism. The results are validated and tested in real world industrial applications.

Results and potential opportunities

The research results can be used to utilize novel intelligent services in working machines. The quantification of the operators’ skills, and recognition of the need for training enable to increase the proficiency of the operators. Currently the operators of the working machines do not have possibilities to get personal feedback of their work. The intelligent coaching system gives criticism about the operator’s actions regarding to the efficiency of the process. Moreover, the recognition of the variation in the operator’s performance enables also the better recognition of the machines technical performance degradation. Thus, the operator performance evaluation system can be attached to the overall process health management (PHM) system.

Partners and funding

Our industrial partners are:

Our research is financed by the 100 Year foundation of The Federation of Finnish Technology Industries.

Technical information

  • Keywords: Skill evaluation, human factors, hidden Markov models, human adaptive mechatronics, intelligent coaching systems, intelligent machines, intelligent services
  • Duration: 2008-2010
  • Research area: Process Control, Mechatronics, Operations & Maintenance
  • Related projects: LATU
  • cURL error: name lookup timed out