The Ethics of MTBF
The National Society of Professional Engineers (NSPE) has a code of ethics. As expected. Other professional groups also have similar code of ethics that provide guidance to their members and set expectations for everyone.
One of the Fundamental Canons of the NSPE code of ethics is:
When should we be ethical?
An underlying element of this canon is the person acting deliberately to deceive others primarily for personal gain. While important and not an expected behavior of someone from any field, it implies knowingly committing the deception.
A contrary expectation is the professional training would include basic and advanced knowledge of practices that provides the foundation of behavior that avoids any need for deception. In other words someone working in a profession should know how to perform their duties and not have to resort to any deceptive acts.
We should not expect any deception and the professional person should know better. What happens if the professional training and body of knowledge inadequately prepares the professional?
The act in accordance with their understanding of appropriate behavior, using there ‘flawed’ knowledge to provide analysis, summaries, and recommendations.
In short, they do not know better
One may argue that the body of knowledge mastered by those deemed a professional in that field is the basis for professional behavior. Committees representing the profession validate the BoK supports the professional and public.
The BoK may be slow to change, it may lag advances in techniques and processes. Yet, when an element of the BoK is proven to have a fault or inadequate technique, it should be changed.
Members of the profession then agree to update the guiding documents to reflect the improvements. At what point does change become required?
I would suggest waiting for mistakes based on faulty elements of the BoK leading to death or major loss is unethical.
If we know better, it is in my opinion our obligation to take action to reduce the inappropriate action based on the faulty technique and encourage adoption and use of superior techniques. Is it the ethical thing to do?
Is the Use of MTBF Ethical?
I would say it is not.
That statement should not be a surprise to you. As a community of professional reliability engineers we generally agree MTBF and related reliability measures are widely mis-understood.
We also may agree that MTBF is rarely useful in describing product reliability.
Part of purpose of this site, this essay, and many other writers and speakers in our profession have conveyed the unsuitability of MTBF as a meaningful measure of reliability.
It is hard to argue within the profession that we believe MTBF is both meaningful and understood. Yet, customers request MTBF. Contracts require reports using MTBF. And, we comply.
We claim ease of use, first order approximation, or it actually is useful in some narrow circumstance.
Do we have an obligation to calculate reliability accurately and present the data in clear and understandable metrics? Do we have the professional obligation to avoid deceiving our customers and clients?
Is providing an unwitting person a measure of reliability that is misunderstood not a form of deception? Especially when they believe the product is more reliable.
As a professional what is our obligation when we see MTBF claims being misused by our design team?
It is not to stand by and allow decision to lead to premature product failures. If we hear of claims from vendors based on MTBF, we should ask more questions (basis of calculations, supporting testing, etc.)
And, if we receive a request from a customer asking for MTBF, what is our obligation? What is your ethical responsibility?
We stand alone. In too many cases a professional has unique knowledge and an obligation to use that knowledge for the betterment of society. We have the ability to minimize confusion, facilitate meaningful decisions and enable product reliability improvements.
I say there is no ethical reason to use MTBF. What do you say?
5 thoughts on “Ethics and MTBF”
The vast cases of the application of MTBF are absolutely unethical. MTBF not only represents a useless metric but also is highly misleading. MTBF has cost the oil and gas industry (the industry I work in) millions of dollars and millions of lost hours chasing the wrong failure modes and failure causes. The continued use of MTBF and the high validity and pressure being forced upon the industry to use MTBF is definitely unethical. Thank you for bringing this issue to the fore.
and as professionals we have an obligation to do something about it.
I agree… MTBF is, as most (“means”/”Averages”), deceptive a great deal of the time. The better measure is the actual failure distribution. If a failure distribution is not available for use, then he “lessons learned” library of previous failures and failure failure modes for similar parts should be analyzed to better describe the failure distribution of the current part under study.
MTBF or its corrolary for non-repairable systems MTTF (Mean Time To Failure) are flawed in calculation and implementation. The direction that I have taken is to provide the raw data, assumptions made, and calculation methodology with any reliability calculations provided to customers.
Excellent practice and one that we all should adopt.
thanks for the comment.