This is the fourth edition of my welcome to the RAMS® 2013 web site prior messages are archived on this web site). This is the July update to my message, and I am happy to say that I can identify our Keynote and Banquet speakers for RAMS® 2013. As I have also been doing in these messages, I will also talk about a topic in Reliability – this time the use, misuse, and failure to use Confidence limits (or degree of belief for the Bayesians) when talking about probabilities.
I am actually thrilled to introduce our Keynote Speaker. I have long been a proponent of applying Reliability techniques in improving the delivery of Medical Care. Our Keynote speaker is Dr. Donald Lighter, Director of The Institute for Healthcare Quality Research and Education and Corporate Medical Director for Healthcare Quality and Safety at Shriners Hospitals for Children. He is also the author of two text books on Health Care Quality Improvement. Health Care Quality is of interest to all of the attendees at RAMS®. It doesn’t matter whether or not we are in the Health Care Business, we are all customers of Health Care Practitioners and Institutions and the Quality of the care given is very important to us. The term Quality of Care includes the Reliability of the care given – it’s a terminology issue. I am hoping to be enlightened and learn more about this very important subject from Dr. Lighter. Dr. Lighter’s bio has been posted on our web site.
Regarding our Banquet Speaker who I teased in my last message, we have her under contract and I can tell you her name is Dava Sobel. She is the author of Galileo’s Daughter and Longitude, both New York Times bestsellers, as well as The Planets, and A More Perfect Heaven. She is also a former New York Times Science writer, and the most famous ex-classmate of mine – we attended elementary school together in the Bronx, NY and we are both graduates of the Bronx High School of Science. She will be talking about her books and there will be a book signing after the Banquet. Here is a link to Dava’s web site for those who want to know more about her: www.davasobel.com.
I was attending another conference; yes there are other conferences, but RAMS® is the best and historically most prominent conference in the R&M disciplines, where someone presented: “The Problem with Probability”. This presentation dealt with the subjective nature of probabilities used by many reliability practitioners. The problem was the presenter was unfamiliar with confidence limits, as are many Reliability Practitioners. Another presenter noted that he never had a Probability or Statistics course as an undergraduate. Confidence limits are as important as the Probability value, but require some in-depth knowledge of Statistics to be properly understood. A probability should be presented with its confidence limits to allow proper assessment of its impact. Opinion polls use margin of error to present this concept, usually at 90% confidence. When assessing sample data, high confidence limits (90%+) without a wide spread of values requires a lot of data points. Dr. Ralph Evans, who for many years edited the IEEE Transactions on Reliability, and was the Publications Chair for RAMS®, used the term “genuine imitation statistician” to describe the approach to statistics and probability of many reliability engineers. Great care needs to be taken when discussing probabilities or we start looking like the charts on the front page of USA Today where predicted doom is just around the corner, rather than a peer reviewed technical publication. I will be happy to discuss this or any other topics with you in January at RAMS® 2013.
I look forward to seeing you in January at the Rosen Shingle Creek Resort.
Aron Brall, CRE