Abstract

This report describes an evaluation of a new lighting system designed for the final paint inspection deck for Ford Michigan Truck plant (MTP). The design of this new lighting system was based on the principles of effective paint inspection lighting as described by Lloyd and Boyce (1996) and as specified in the Ford specification (Ford, 1997). In this evaluation the new system was compared with the existing system which was installed approximately four years prior. The existing system was similar to the lighting systems installed in six other U.S. Ford assembly plants.

The new lighting system was installed on the South spill out inspection line during the plant shutdown in December 1996 and the data reported here were collected over a six day period in March, 1997. Two objective measures of system performance were used in the evaluation: Number of defects missed, and number of vehicles sent to rework. Inspectors and managers were also interviewed to determine how they liked the new system relative to the old.

The number of defects missed was measured using a down-stream audit area where vehicles randomly selected from the North and South lines were sent for a thorough re-inspection. The re-inspection results indicated that the mean number of defects missed on the South line was 25% lower than the mean number of defects missed on the North line. The results also showed that the number of "zero defect" vehicles passing through the South line was double that of the North line.

Prior to the lighting change there were no differences between the North and South lines in the number of vehicles sent to rework. After the lighting change, an average of 45 percent more vehicles were sent to rework for the South line than for the North line for the A and B crews. There was no difference between the lines in the number of vehicles sent to rework for the C crew.

Results of the inspector interviews indicated that the seek and repair operators overwhelmingly preferred to work on the North line, even though 9 of the 12 inspectors felt defect visibility was better on the South line.

These results indicate there was a large, statistically-reliable, improvement in inspection performance (for small topographical defects) on the line where the new lighting system was installed. These results are consistent with the results of similar in-plant evaluations conducted at the Chicago and Twin-Cities Assembly plants and provide support for the validity of the lighting system design as specified. The finding that inspectors do not prefer the new lighting system is consistent with the results of inspector interviews at the Twin-Cities Assembly plant. The reliable discrepancy between inspection performance and inspector preference data suggests that inspector preferences should not be the only criteria used in the design of new inspection lighting systems.