FOCUS: Decrease cycle time and build a predictive model
A leading manufacturer of automotive parts was using a chemical absorption process. There was insufficient understanding of the key input factors controlling the chemical weight from this process with no predictive model available. The current feasible solution was not optimum and had a Cpk of 0.40 and a defective rate of 23%.
The Black Belt used the DMAIC methodolgy to tackle this problem. During the measure phase, the opportunity for significant improvement could be seen from the baseline data. A measurement system analysis was conducted and showed their measurement process in control and very capable. The team built a cause and effect diagram describing their process. Several noise factors were changed to control factors with standard operating procedures developed by the project team. Two factors were selected for a full factorial design of experiment. A regression model was developed and the process was optimized for performance on their response variable while simultaneously reducing the cycle time requirement. In addition to the revised standard operating procedures, the control phase incorporated control charts and statistical process control methods to ensure the process and associated sub-systems remained at the improved performance settings.
The regression model produced was optimized to shorten the process cycle time by a factor of 2, resulting in a 100% improvement over the historical setting. The control phase Cpk had improved to 1.0 with a 0.26% defective rate. This represents an improvement nearing 2 full orders of magnitude from the baseline performance. The success of this project shows the power of the Six Sigma methodology. The project met all goals with estimated future savings of $100,000. This enabling technology will help this company remain ahead of their competitors and win future customer programs.
* Due to nondisclosure agreements, the organization referenced in this example cannot be disclosed.