The evolution of the manufacturing industry may be viewed as proceeding from Dedicated Manufacturing Systems (DMS) to Reconfigurable Manufacturing Systems (RMS). Customer requirements change unpredictably, and so DMS are no longer able to meet modern manufacturing requirements. RMS are designed with the focus of providing rapid response to a change in product design, within specified part families. The movement from DMS to RMS facilitates mass-production of custom products. Custom parts require inspection routines that can facilitate variations in product parameters such as dimensions, shape, and throughputs. Quality control and part inspection are key processes in the lifecycle of a product. These processes are able to verify product quality; and can provide essential feedback for enhancing other processes. Mass-producing custom parts requires more complex and frequent quality control and inspection routines, than were implemented previously. Complex, and higher frequencies of inspection negatively impact inspection times, and inherently, production rates. For manufacturers to successfully mass-produce custom parts, processes which can perform complex and varying quality control operations need to be employed. Furthermore, such processes should perform inspections without significantly impacting production rates. A method of reducing the impact of high frequency inspection of customized parts on production rates is needed. This dissertation focuses on the research, design, construction, assembly, and testing of a Non- Contact Automated Inspection System (NCAIS). The NCAIS was focused on performing quality control operations whilst maintaining the maximum production rate of a particular Computer Integrated Manufacturing (CIM) cell. The CIM cell formed part of a research project in the School of Mechanical Engineering, University of KwaZulu-Natal; and was used to simulate mass-production of custom parts. Two methods of maintaining the maximum production rate were explored. The first method was the automated visual inspection of moving custom parts. The second method was to inspect only specified Regions of Interest (ROIs). Mechatronic engineering principles were used to integrate sensor articulation, image acquisition, and image processing systems. A specified maximum production rate was maintained during inspection, without stoppage of parts along the production line occurring. The results obtained may be expanded to specific manufacturing industries.

Thesis (M.Sc.Eng.)-University of KwaZulu-Natal, Durban, 2009.