Industry Implementation of Six Sigma Techniques
If you have ever seen an industrial manufacturing environment, it may appear that rugged-physical labor is the key to success. While there is a fair amount of tough tasks that must be muscled through, the fine technical details are what ensures that processes stay within the desired parameters. If a process gets out of control, the product will not meet the customers expectations and risk revenue for the company or even a customer. To remedy a process that is out of control, Lean Six Sigma tools can be utilized to ensure the success of the company and satisfaction for the customer.
Imagine that you are a worker for Hodges Fiberboard, a wood composite panel manufacturer. The company manufacturers high-density fiberboard for the use in flooring and furniture. The press operators are having an issue with the fiber entering the press being too wet/ dry. A moisture meter, located just before the press, indicates no sign of sporadic readings outside of set parameters during production. It is up to you to determine the root cause and address the problem.
Luckily for you, the tools learned during your Lean Six Sigma Green Belt class will be sufficient to solve this problem. You have already completed the first step, define the problem. In this scenario, the problem is the moisture meter appears to be out of calibration. The next step is to measure and acquire data. Ten fiber samples are taken from the production line at certain times and weighed using the exact same method. The samples are then placed in an oven for 24 hours to dry. Samples are removed from the oven and weighed again. By using a simple calculation you can determine the content of the fiber samples.
Next, you gather data from the computer to determine the moisture meter reading at the time you gathered the fiber samples form the line. The samples are plotted against one another on a scatter plot to form a correlation.
Now that you have a graph of the correlation between meter moisture and lab moisture, it is time to analyze the data. As you can see from the scatter plot, the regression line slope is y=0.8676x which is not ideal. A slope of y=1x would be the best case for this scenario as a slope of 1x would indicate that the values from the moisture meter and the lab measurements are equal. The R2 value is 0.69, which indicates that the data does not closely fit the regression line.
After learning that the moisture meter is not reading accurately, it is now time to improve the process. The new regression is entered into the moisture meter to update the calibration. Now the moisture meter should be reading accurate moisture of the fiber and the press operator will not have any more issues with moisture. The last thing to do is control this solution by conducting routine checks of the moisture meter to ensure that the regression has not changed.