Quality prediction for reconfigurable manufacturing systems via human error modelling

The early assessment of the performance of responsive manufacturing systems contributes significantly to achieving their potential effectiveness. In most manufacturing scenarios, the human involvement is considered one of the critical elements affecting the system performance. In reconfigurable manufacturing systems (RMSs), the workers' tasks are expected to change frequently as the system is reconfigured. The ability to predict the probability of errors caused by human involvement can provide the system designer with insights as to the required skill levels, training programmes, job design, tasks assignment, work organization as well as options for modifying the system design to achieve better quality results. A model for assessing the probability of human errors in RMSs, based on tasks characteristics, work environment, as well as workers capabilities has been developed using the multi-attribute utility analysis. Application of the model to an industrial case study demonstrated its ability to assess, early in the system design and development stage, the probability of errors resulting from human involvement. This is critical in investigating different improvement opportunities to achieve lower levels of errors owing to human involvement, and hence higher quality.