The evaluation of an elementary virtual training system for manual assembly

Due to the low volume high variety strategies of manufacturing companies, manual assembly operators have a much larger cognitive load than before. The expertise of the operators must be kept up to date at any time. Since the high investment and low flexibility of a real setting to perform a manual assembly training, a virtual replica is introduced in many cases. The aim of this paper is to study the effect of an elementary virtual training for manual assembly tasks. In literature, different studies on the topic can be found; nevertheless, a comparison between the different studies is not possible due to diverse evaluation methods and descriptions. A benchmark for a uniform evaluation of virtual training systems is presented and applied to this experiment. Two groups were submitted to a number of manual assembly tasks. The test group got a virtual training period in advance. A significant learning transfer during that training period was observed. When the first assembly of the reference group is counted as a real training, no significant difference can be found between the virtual and real training. The outcomes of this experiment will be used in future work to compare different virtual training systems and influential factors such as the assembly complexity. Furthermore, the application of virtual training to manual assembly in a mixed-model environment and its industrial usability are topics that still need to be studied.