How a scientific test uncovered an insight that reduced steps in the finishing process, achieving cost savings and increased productivity.
Background: A major global automotive manufacturer engaged Predictum® to review it’s manufacturing operation with the purpose of realizing cost savings and increasing productivity.
During a plant tour our engineers came across a labour intensive activity where about a dozen people were manually buffing car body panels to improve the look of the part. Inquiring about the the cost-effectiveness of this activity we raised a simple but important question common in many industries including manufacturing: "How do you know that this activity improves part quality?". On further discussion it was confirmed there was no data on this process and the manufacturer required a systematic approach to determine the effectiveness of all this work.
Working with the manufacturing team we put together a strategy to conduct what is called a "paired comparison" experiment. A paired comparison generally is any process of comparing entities in pairs to judge which of each entity is preferred, or has a greater amount of some quantitative property, or whether or not the two entities are identical. Here each part's defects were noted. They randomly applied the treatment to about 50% the affected area and left the other 50% untouched. This took some coordination and effort which had the positive effect of drawing in staff from across the plant at all levels. Everyone was interested in the results.
Through statistical analysis Predictum revealed opportunities readily apparent to improve productivity, quality and yields. Streamlining operations while significantly decreasing unnecessary costs
Reviewing the experiment the results were stunning. This step in the process, which cost $1.5 million annually, had no effect. The data showed that approximately 50% time both treated side AND the non-treated side showed defects through subsequent steps. Stopping the treatment did not reduce yield but it did save $1.5 million/year and allowed them to focus on the root causes of these defects. As a result of this insight decisions were made quickly and effectively. This new approach had led to review of other areas of the manufacturing methods to continue to make significant improvements.