Multi-dimensional Pokayoke-Stoplight Control system in unmanned manufacturing cells

Abstract

A multi-dimensional Pokayoke-Stoplight Control (M-PSC) system has been developed that uses an on-line, real-time, and fuzzy-logic-based approach. By using various non-contact or contact sensors, multi-dimensional measurements have been taken on a real-time basis. By processing in a multi-input and single-output fuzzy logic controller, the measurements are interpreted to represent the status of the manufacturing process. The fuzzy logic controller serves as a decision-making device to turn on a stoplight, e.g. green, yellow, or red, not only to indicate the acceptance of the product but also to warn the monitor of the UMC when a change in the process occurs. The fuzzy logic system has been pre-defined based on the statistical process control technique and 100% inspection methodology. This approach enables the M-PSC system to provide on-line, real-time inspection to prevent defects before they occur in a UMC. The performance of the M-PSC system has been evaluated by experimental tests on a CNC lathe. The results demonstrate that the M-PSC system provides a UMC with a “make one, check one, and pass one on” capability.