A short time before the problem surfaced, the engineer in charge of the pad-liner project had attended a DOE workshop. The introductory workshop, Experiment Design Made Easy, covers the practical aspects of DOE. The trainee learns all about simple but very powerful two-level factorial designs. The workshop follows the standard approach outlined in Box, Hunter and Hunter’s classic, Statistics for Experimenters. During this workshop, the student discovers how to effectively:
The supplier’s manufacturing process included a dryer and an adhesive-curing oven. After consulting Design-Expert®, a software program for design of experiments with a built-in design builder to determine the best type of DOE to conduct (Figure 2), eight factors affecting these two operations were input into the software.
Three factors were discovered that significantly affected pad thickness:
These process factors produced such a significant change that the desired pad-release specification could be accomplished without changing any of the materials. This was good news because a material change meant requalifying an entire assembly – a time-consuming and costly process.
The new process has been a complete success. In the last year and a half since implementation, the automaker is reporting nearly 100% pad peel-off — and there have been no complaints or rejects.
Cost savings are substantial. If each factor had been tested separately, the total expense would have reached $200,000. Instead, the entire redesign process cost $10,000 in material. There was also a savings in the number of work hours spent on the process, since testing each individual factor would have consumed more time in each phase.
Today, DOE is playing a much-needed role at the supplier’s facility by improving existing material, developing new products, and optimizing throughput, material and process costs.
A far more effective method is to apply a systematic approach to experimentation that considers all factors simultaneously. That approach is DOE. Companies worldwide are adopting DOE as a cost-effective method for solving problems plaguing their operations.
DOE provides information about factor interaction and the way a total system works, something not obtainable when testing one factor at a time while holding other factors constant. DOE shows how interconnected factors respond over a wide range of values, or levels, without requiring the testing of all possible values directly. It is done by fitting response data to mathematical equations.
Collectively, the equations serve as models that predict what will happen for any given combination of values. Using these models, engineers optimize critical responses and find the best combination of values.