Saturday, November 12, 2005

Process Capability and System Capability


If you have worked with suppliers, you know how frustrating it is to be responsible for one that can't prevent defectives from reaching your back door. You throw all your experience at them. You encourage and insult them. Anything to make them better.

Zoom out and think about what is happening. First, an analogy. When you are releasing a new process, you conduct a process capability study. This entails collecting and measuring output (ex. a critical dimension) from the process over a period of time. The results are summarized and compared to the customer's requirements (ex. specification limits). If the output doesn't meet the customer's requirements, you have four choices:
1. Change the customer's requirements so that the process output fits
2. Improve the process to meet the customer's requirements
3. Change the process to make the output fit the customer's requirements
4. Keep the process intact and 100% inspect the output to the customer's requirements.

The first one is typically impossible. Number two should be tried but, depending on the degree of incapability, may yield marginal results. Number three is difficult and costly. Unfortunately, number three is often implemented to much consternation.

Now, think about that supplier that delivers defectives to you month after month. His quality system is not capable of preventing defectives from reaching your warehouse. You have four choices as a supplier development professional:
1. Change your expectations to fit the supplier's performance
2. Improve the supplier's quality system to meet your requirements
3. Change suppliers
4. Keep the supplier and make sure his output is 100% inspected (or more) prior to you.

Number one should not be listed. It should not be considered under any circumstances. Number two should be tried but will take much time. This is what supplier development really means. Number three is the last straw. It is difficult and costly. Number four is not done enough but must be a focus to protect the customer.

5 comments:

Anonymous said...

Hi Stephen,

As a result of reading your blog "Process Capability and System Capability", I think you will discover my site on Highest Wages In All States will be a great help.

To illustrate our diverse range of wages information, here are some of the latest search terms that located our site ...

Photogrammetrists Wages
Chemical Engineers Wages
Civil Engineering Technicians Wages
Computer Hardware Engineers Wages
Electrical Engineers Wages
Electrical Drafters Wages
Electronics Drafters Wages
Electricall Engineering Technicians Wages
Electro Mechanical Technicians Wages.

We have over 200 "have to read" career and wages articles in addition to many other interesting subjects in our Wages For All Occupations site.

Best Regards
Emily

Anonymous said...

I agree 100% with the number four tactic, but people do not always practice what they preach! If we have a supplier who is constantly providing us with various defect, concentrating on past issues will not fix the big problem, but force us to continue in firefighting mode until we reach the last straw.
Companies need to concentrate on those parts of the program that are broken and implement a plan of action that will begin to fix the issues one at a time to prevent the system from continually breaking down.
I don't want to be a fire fighter. Only we can prevent future issues!

Anonymous said...

If the supplier is out of control in many areas how do you convince the "system" that what you are doing is correct? Can you pick and choose what to look at if there is a lack of consistant qality?
When do you decide to go 100% inspect?

Stephen said...

Tactic number four is hard to do from a logistics and commercial standpoint but I am convinced that it is the only way to get out of the reactive mode. If you constantly react to problems,you can't visit the supplier to help them develop. If the supplier constantly receives calls from you, he can't focus on improving his quality system.

Reaction is a tremendous drain on resources.

Controlled shipping is intended to stop the flow of defectives from the supplier to you. Simply put, it gets the customer off the supplier's back. The supply chain can stabilize.

I am going through this with three suppliers right now. Based on our data, we concluded that these suppliers have quality systems incapable of supplying defect free product. So, the supplier quality group informed purchasing that the suppliers must use a third party to 100% inspect their product prior to receipt at our plant.

The suppliers, as expected, balked at this and countered with their own plan to 100% product. Yes, it wasn't what we wanted but we obviously got the supplier's attention.

We are now agreeing on what must be checked. This is somewhat negotiable as well.

Purchasing must be an active part in this activity. More than anything, this is a commercial issue. We are telling the supplier to 100% inspect multiple characteristics at their own expense.

It is important to send the message that no defectives are allowed. We are sick and tired of poor quality. Actions speak louder than words and controlled shipping is a strong action.

Stephen said...

When do you decide to 100% inspect? Great question. I'll tell you how we decided.

We have over one hundred suppliers. We have a well established process for measuring supplier performance. So, we have an abundance of data.

We measure our suppliers using four factors: the number of descrepancy reports written in a month, the number of corrective actions requests issued in a month, whether or not the supplier met the PPM target, and whether or not we had a customer return due to a supplier defect.

In analyzing the data, six suppliers stood above the rest in terms of poor quality performance. So, we are focusing our efforts on these suppliers.

Looking at the six supplier's data, though the performance is bad, it is very predictable. Meaning, you know that the supplier will deliver bad product every month. You don't know when but you know it will happen. It is analogous to a control chart. Think of an xbar, r chart that runs for a long period of time within the control limits. The behavior is stable and predictable. But just as a process can be in control but not capable, a supplier's performance can be in control but not capable of supplying defect free product.

These suppliers are always in reactive mode. The supplier quality engineers responsible for them are always reacting. The consequence is that nothing gets better.

The only way to get proactive is to stop defectives (NOW!) from getting to your plant. Until the supplier fixes his quality system (with your help), the product must be 100% inspected.

This is no different than a process capability study. If the process is not capable, you must 100% inspect the product until the capability is improved.