Understanding human variability to improve manufacturing system design

Abstract

The way in which a manufacturing system is designed is a crucial determinant of its operational efficiency and profitability. System design is, therefore, a major consideration for organisations. However, the efficacy of current system design techniques could be improved. Designers still do not often construct assembly line layouts that provide the high level of efficiency and flexibility that is required by today's highly competitive organisations. One aspect of system design processes that has been found particularly deficient is the way in which human labour is considered. Current system design practices tend to disregard that workers may vary as they perform production tasks and the impact that this may have on the assembly line. In particular, system designers appear to have little understanding of how production work may be affected by workers' personal attributes. The research presented in this thesis sought to identify the nature and impact of human work performance to inform design stage practices. In particular, the research aimed to establish the extent to which the personal attributes of workers influence variations in their performance of production tasks. The research involved parallel data collection studies in a real manufacturing system where workers' performance of production tasks and their personal attributes in respect of work-related attitudes were measured and analysed. Overall, this research did not nd evidence of relationships between the particular work- related attitudes and production task performance that were measured in the study. However, indications were found that suggest production task performance variations may be produced by interactions between task characteristics and workers personal attributes. This evidence is an important development in understanding worker behaviour and informing manufacturing system designers that their neglect of worker performance variation in design stage evaluations may be a major cause of current design weaknesses.