About the columnist
Larry Edson retired from General Motors as a Technical Fellow for Reliability and Test in 2008.
He now uses his 40+ years of experience to help alternative energy companies in the field of reliability and accelerated testing.
It is more expensive to do a job wrong than it is to do it right, and it is more expensive to make a product un-reliable than it is to make it reliable. What do you think about that? Do I have the words backwards or is this the real truth when one looks at the total life cycle costs?
The columnist, Larry Edson, on top of a wind turbine.
By Larry Edson
The automotive industry has learned the hard way that, in the long run, a reliable product will always cost less than an un-reliable product. The difference between reliable and not reliable is not about how much money you throw at the product, but rather how disciplined is the thought in design and test and, even more important, your reaction to what is learned in test. Companies that are vertically integrated are responsible unto themselves for efforts to produce a reliable product. However, many industries, like automotive and wind are system integrators, and as such depend heavily upon the actions of their suppliers.
The system integrator role is much like a teacher who will be financially rewarded based upon the academic improvement shown by his students. The teacher knows that he cannot go home with each student every night and help that student with their homework. The teacher must assemble intuitive lesson plans and be a good mentor to the students so that each student can progress at an optimum rate. The teacher must cultivate an enthusiasm in the student so that conscientious work will occur in the unattended moments. This can represent a very frustrating situation for teachers just as it does for system integrators.
The reputation of the system integration company depends upon the actions and attitude of the suppliers to that company. How do you manage this situation? The same way it is managed by the school teacher. Have good lesson plans and be engaged with the students in their work and always be a source of motivation as opposed to a source of discouragement. In the world of reliability, the good lesson plans become a Reliability Program Plan. But a plan by itself is not adequate; it must be accompanied with an insatiable level of motivation and enthusiasm for developing robustness into products.
Dr. Tatsuhiko Yoshimura, a retired Director of Quality and Reliability for Toyota, was recruited by General Motors (GM) in 2003 with the intention that he would introduce new methodology in GM that would lead to quantum leap improvements in quality and reliability. After one year of introducing new methods such as DRBFM (Design Review Based upon Failure Modes) and DRBTR (Design Review Based upon Test Results), Yoshimura-san declared that he needed to redirect his energy from methods to culture.
The cultural shift he spoke of was for all of the people at all levels in the company to be consistently focused upon reliability and quality with an unparalleled level of consistency. “All The People, All The Time” (APAT) became his primary focus for the next four years in GM before he retired in 2007. Beyond APAT, Yoshimura-san focused on three major concepts in his teachings at GM; Good Design, Good Discussion, and Good Dissection (GD3). The good design comes from good engineering. The good discussion comes from the culture of the company and the willingness to look deeply into a product and discuss the details. Good dissection is the desire to take a product apart that has “passed the test” just to see if there are buds of problems emerging.
The characteristic that distinguishes the very best companies from the rest is how well they maintain a “consistency of purpose” when it comes to reliability and quality. Reliability is not a topic well suited to cyclic attention. Studies have shown that the one engineering metric that most closely tracks market share is reliability. Reliability is, in a sense, a measure of the faith that a customer puts in a brand name. Reliability is the cornerstone for decision making. People always return to the basic need for a product to be reliable even when tempted by eye catching designs and claims of extreme performance. It is time for the mature industries to share their experiences with the next generation industries, like wind and photo-voltaic, so reliability may progress quickly in those areas where it serves the betterment of mankind.
The major elements in a good Reliability Program Plan are noted below.
• Quantification of realistic Reliability Requirements
• Application of Anticipation Methods early in the program to develop robustness
• Quantification of Design Margin needed as a result of the reliability requirement
• Qualitative Testing to uncover weaknesses
• Quantitative Testing to demonstrate reliability with statistical vigor.
These will be treated in later columns in OceanWise.