Nondestructive Evaluation (NDE) Technology Initiatives (NTIP). Delivery Order 0039: Statistical Comparison of Competing Material Models

Abstract

The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. REPORT DATE (DD-MM-YY)2. REPORT TYPE 3. DATES COVERED (From -To) SPONSORING/MONITORING AGENCY REPORT NUMBER(S) AFRL-ML-WP-TR-2003-4096 DISTRIBUTION/AVAILABILITY STATEMENTApproved for public release; distribution is unlimited. SUPPLEMENTARY NOTESThis report contains color. ABSTRACT (Maximum 200 Words)The ability to decide objectively between competing material models has always been an important engineering task. For a variety of reasons, simple inspection of the general "fit" of various models to the appropriate data does not always result in an obvious choice for the superior model, and heretofore there has been no statistical measure that can be used to choose between models (except in the simple case of "nested" models). Bayes factors overcome these previous limitations and can provide a valid objective statistical measure for choosing between competing material models. This study explores the use of Bayes factors as an objective measure in choosing between competing material models. The focus is on the use of this method by engineers who may or may not be well versed in statistical methods. In addition, the statistical background necessary to gain an in-depth understanding of the Bayes method is included as appendices. Also discussed in some detail is the Random Fatigue Limit s-N model proposed by Pascal and Meeker, which holds considerable promise for describing HCF behavior, as well as the applicability of applying Bayes factors to this model. SUBJECT TERMSBayes Factors, modeling, random fatigue limit ................................................................................................. 1) The ability to decide objectively between competing material models has always been an important engineering task. For a variety of reasons, simple inspection of the general "fit" of various models to the appropriate data does not always result in an obvious choice for the superior model, and heretofore there has been no statistical measure that can be used to choose between models (except in the simple case of "nested" models). Bayes factors overcome these previous limit...