5 Matching Results

Search Results

Advanced search parameters have been applied.

Characterizing reliability in a product/process design-assurance program

Description: Over the years many advancing techniques in the area of reliability engineering have surfaced in the military sphere of influence, and one of these techniques is Reliability Growth Testing (RGT). Private industry has reviewed RGT as part of the solution to their reliability concerns, but many practical considerations have slowed its implementation. It`s objective is to demonstrate the reliability requirement of a new product with a specified confidence. This paper speaks directly to that objective but discusses a somewhat different approach to achieving it. Rather than conducting testing as a continuum and developing statistical confidence bands around the results, this Bayesian updating approach starts with a reliability estimate characterized by large uncertainty and then proceeds to reduce the uncertainty by folding in fresh information in a Bayesian framework.
Date: October 1, 1997
Creator: Kerscher, W.J. III; Booker, J.M.; Bement, T.R. & Meyer, M.A.
Partner: UNT Libraries Government Documents Department

Applications of quantitative whole body autoradiographic technique in radiopharmaceutical research

Description: The routine evaluation of radiopharmaceuticals involves dissecting tissue distribution studies (DTDS) and gamma or positron imaging. DTDS have the following disadvantages: since not all tissues can always be sampled, sites of radiopharmaceutical uptake may be missed and because the procedure involves weighing of dissected tissue samples, the spatial resolution of this method is low and determined by the smallest amount that can be weighed accurately. Gamma camera imaging and positron emission tomography though more comprehensive in evaluating the global distribution of a compound, have relative low spatial resolution. Whole body autoradiography of small animals has a much higher spatial resolution as compared to the above and depicts the global distribution of radiopharmaceuticals. A computer-assisted quantification method of WBARG applied to positron, beta, and gamma emitters will complement the method by producing quantitative values comparable to those obtained by dissection and direct tissue counting, with the advantages of depicting the global distribution at high spatial resolution.
Date: January 1, 1982
Creator: Som, P.; Oster, Z.H.; Yonekura, Y.; Meyer, M.A.; Fand, I. & Brill, A.B.
Partner: UNT Libraries Government Documents Department

Data-gathering method for use in modeling energy research, development, and demonstration programs

Description: The development and testing of a data-gathering method for use in a computer program designed to model energy research, development, and demonstration programs for decisionmakers are described. The data-gathering method consists of face-to-face interviews with the scientists working on the projects that will be modeled by the computer program. The basic information gained from an interview includes time estimates for reaching certain project goals and the probability of achieving those goals within the times estimated. The interview method is based on decision analysis techniques. The Magnetic Fusion Energy program of the US Department of Energy was selected as the test case for this project. The data gathering method was used at five fusion projects to determine whether it could meet its design criteria. Extensive statistical analysis was performed to learn how much the expert's answers agreed, what factors were likely to enter into their estimates, and how their estimates corresponded to the officially scheduled dates and to the dates that the project goals were actually achieved. The interview method was considered to have met its design criteria and to be a valid tool for planning.
Date: January 1, 1981
Creator: Meyer, M.A.; Booker, J.M.; Cullingford, H.S. & Peaslee, A.T. Jr.
Partner: UNT Libraries Government Documents Department

The use of fuzzy control system methods for characterizing expert judgment uncertainty distributions

Description: Fuzzy logic methods permit experts to assess parameters affecting performance of components/systems in natural language terms more familiar to them (e.g., high, good, etc.). Recognizing that there is a cost associated with obtaining more precise information, the authors particular interest is in cases where the relationship between the condition of the system and its performance is not well understood, especially for some sets of possible operating conditions, and where developing a better understanding is very difficult and/or expensive. The methods allow the experts to make use of the level of precision with which they understand the underlying process. The authors consider and compare various methods of formulating the process just described, with an application in reliability analysis where expert information forms a significant (if not sole) source of data for reliability analysis. The flow of information through the fuzzy-control-systems based analysis is studied using a simple, hypothetical problem which mimics the structure used to elicit expert information in Parse. They also characterize the effect of using progressively more refined information and examine the use of fuzzy-based methods as data pooling/fusion mechanisms.
Date: December 1998
Creator: Smith, R. E.; Booker, J. M.; Bement, T. R.; Parkinson, W. J.; Meyer, M. A. & Jamshidi, M.
Partner: UNT Libraries Government Documents Department