Case Study

Manufacturing

How a scientific test uncovered an insight that reduced steps in the finishing process, achieved cost savings, and increased productivity.

Background: A major global automotive manufacturer engaged Predictum to review its manufacturing operation with the goal of realizing cost savings and increasing productivity.

The Problem

During a plant tour, our engineers came across a labour-intensive activity involved manually buffing car body panels by a dozen workers to improve the look of the part. When we enquired about the the cost-effectiveness of this activity, we raised a simple, yet important question, which is common in many industries, including manufacturing: "How do you know that this activity improves part quality?" On further investigation, it was confirmed there was no data about this process to substantiate the work effort and the manufacturer required a systematic approach to determine the benefit.

The Solution

In working with the manufacturing team, we came up with a strategy to conduct 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. In the experiment, each part's defects were noted. They randomly applied the treatment to about 50% the affected area and left the remaining 50% untouched. This took some coordination and effort which engaged staff of all levels, who were interested in the learning about the results, from across the plant.

Through statistical analysis, Predictum revealed measurable results on how to significantly improve productivity, quality, and yields while streamlining operations and decreasing unnecessary costs.

The Benefit

In reviewing the experiment, the results were astounding. This step in the process, which had cost $1.5 million annually, had no measurable benefit. The data showed that approximately 50% of the time both the treated side and the non-treated side showed defects through subsequent steps. Stopping the treatment did not reduce yield, but it did save the manufacturer $1.5 million per year and allowed them to focus on the root causes of these defects. As a result of this insight, process decisions were made quickly and effectively. This new approach had led the manufacturer to review of other areas of its manufacturing methods and continue to make significant improvements in its operations.

The Problem

No data to support the effectiveness of a process

The Solution

A critical, scientific review process to determine the effectiveness of a process

The Benefit

$1.5 million in annual cost savings