PREPRINT: Subject to revision. Per­ mission to publish this paper, in full or in part, after its presentation and with credit to the author and the Society may be ob­ tained upon request. The Society is not re­ sponsible for statements or opinions advanced in pa­ pers or discussions at its Meetings. CHASSIS AND TOTAL CAR RELIABILITY By J. R. GRETZINGER Director of Reliability and Quality Control BUICK MOTOR DIVISION Flint, Michigan For presentation at the SAE DETROIT SECTION October 3, 1960 SOCIETY of AUTOMOTIVE ENGINEERS, lnc., 485 Lexington Avenue, New York 17, N. Y. Downloaded from SAE International by University of Birmingham, Tuesday, August 21, 2018Downloaded from SAE International by University of Birmingham, Tuesday, August 21, 2018CHASSIS AMD TOTAL CAR RELIABILITY Reliability is really the business of everyone associated with design, manufacture, and servicing of automobiles. It is not one department's job. To discuss the subject of reliability intelligently, we. should have a common under­ standing of some of the fundamental terms. To those of you who are familiar with reliability programs, this will be a repetition to some degree. First is the definition of reliability — what are we talking about? The reliability we are talking about is a quantitative thing, not a qualitative thing. The definition used by most is — "Reliability is the mathematical prob­ ability that a device will perform satisfactorily for a specified time under specified conditions of usage." For example, assume we had a test lot of 100 automobiles. If 90 of them perform without failures or malfunctions for a specified time, and 10 failed for some reason, the reliability of this automobile under this test con­ dition would be 90%. The reliability concept that we are using originated on military pro­ grams a few years ago. It is true that there is very little really new about reliability, but the fact that it is an organized, planned and scientifically executed program by people who have no other primary responsibility is new. Let's take a look at some of the things that caused military reliability pro­ grams to be put into effect: 1. Quality control organizations are not able to control field failures adequately. 2. Product complexity is increasing. 3. We have a shorter development time for our programs. 4. The customer is demanding higher reliability. 5. The cost of unreliability is increasing. 6. The importance of flight safety. 7. We have no way to test "one-shot" equipment. 8. There was some very unfavorable publicity concerning the reliability of our important weapons systems. These are some of the reasons we need an organized effort for relia­ bility, and the result has been that the military is now making the requirements for reliability a formal part of some of their contracts. The items on this chart that are underlined also apply equally to the automobile industry. In addition, there are definite advantages in applying a reliability program to an automobile as compared to military products. 1. We have higher volume production, which gives us a larger statistical sample. 2. We normally have more prototypes to test. Downloaded from SAE International by University of Birmingham, Tuesday, August 21, 2018- 2 - 3. Our design changes are much less radical than in military- equipment. 4. We have more failure data on previous designs and the cur­ rent design. 5. We have "better control of our servicing of our product. 6. We can obtain faster decisions to correct a reliability problem. There are, no doubt, more reasons, but these are very definite pluses for reliability in the automobile industry. The 1927 Buick had 238 critical parts, and if you assume the complete car reliability was 90%, then each criti­ cal part had a reliability of 99.956%. Now, the 1960 Buick is more complex and has 696 critical parts. If each part had the same reliability, then the auto­ mobile reliability would drop to 73