Two test statistics are suggested for discriminating between the exponential model and the more general Weibull or gamma models, and these are compared to some previously used test statistics by Monte Carlo methods. The results of estimating reliability under an exponential assumption when the true model is Weibull is also investigated. These results as well as the tests mentioned above indicate that the exponential model is often not adequate when the more general models hold. In contrast to this result it was found that the Weibull model was quite robust relative to the generalized gamma distribution with regard to reliability estimation. Some general pivotal function properties are presented for the maximum likelihood estimator of reliability for the generalized gamma distribution and similar results also hold for the Weibull procedure under a generalized gamma assumption. These results made a Monte Carlo study of this problem feasible. Since the maximum likelihood estimators are apparently ill-behaved for smaller sample sizes and since the Weibull model is robust it appears little is gained by using the generalized gamma distribution for samples of size less than 200 to 400.
An experimenter wishes to select the most reliable of K s-normal (or lognormal) populations. If he samples n items from each of the K populations, then how large should n be in order that he will have a desired level for the probability of correct selection? This paper gives tables which answer this sample-size question.
A statistically robust method was developed using the Weibull distribution to identify and eliminate outliers from the failure stress determinations. The method is applicable to any failure stress data set that follows the Weibull distribution; however, in this application, it was developed for the AASHTO standard test method for conducting the direct tension test (DTT). A large number of stress-at-failure measurements with the DTT were made in the course of instructing users of this device. These data, all for the same asphalt, provided the means for studying the nature of the distribution of the breaking strength of these asphalt specimens. The training database contains more than 900 data points. The current AASHTO practice of eliminating the lowest two stress values was found to be reasonable. However, it is an arbitrary method that may lead to problems in the future. On the basis of the results of this study, the procedure is recommended for use and implementation in the next AASHTO version of the DTT standard.
The Federal Highway Administration (FHWA) has begun a number of initiatives to improve signing on the Nation's roadways. These include workshops to obtain input from experts across the country, a review of each State's highway sign replacement and refurbishing program, and a training course. This report documents the proceedings of the workshops, one held in the Western United States and one in the Eastern United States. The workshops opened with a plenary session on Issues, Needs, and the FHWA Research Program, followed by a plenary session on Development of Minimum Requirements. The program continued with a panel discussion on Performance Standard Criteria, followed by a plenary session on Materials Selection. Breakout sessions on freeway signing, non-freeway signing, and motorist services and tourist-oriented signing completed the first day of the workshops. The second day opened with reports of the breakout sessions and continued with a plenary session on Field Assessment Techniques. A plenary session on Maintenance Procedures and Programs followed. The second day ended with breakout sessions on sign replacement methods, using contracting versus in-house and prison industries, and traffic control during sign replacement. The third day opened with reports on the previous day's breakout sessions, followed by a plenary session on Improved Inventory Techniques. The workshops closed with a Look to the Future plenary session. This report details the remarks made by the panelists and the comments and concerns of the participants on each of these issues.
All laboratories conducting tests for the Strategic Highway Research Program (SHRP's) Long-Term Pavement Performance (LTPP) program were required to be accredited by the American Association of State Highway and Transportation Officials (AASHTO's) Accreditation Program (AAP). AAP includes site inspections of equipment and procedures, and participation in applicable proficiency sample testing. A few critical LTPP tests were not addressed fully by the AAP, and LTPP staff decided to conduct supplemental testing. The Hot Mix Asphalt (HMA) Laboratory Molded Proficiency Sample Program is one of those supplemental tests. Round 1 testing provided within- and among-laboratory diametral resilient modulus data for tests performed in accordance with SHRP Test Protocol P07. The objectives included drafting single operator and multi-laboratory test precision statements in testing proficiency status for SHRP laboratories, and preserving test sample information for future analysis. Worksheets, supporting data, analyses, final comments, and conclusions are presented. A complete set of proficiency sample statements in AASHTO and American Society for Testing and Materials (ASTM) format are provided.
The log-normal and the Weibull are often considered for situations in which a skewed distribution for a non-negative random variable is needed. The ratio of maximized likelihoods provides a good test for selecting one of these. A table of the necesqary critical values is given. The table may also be used for discriminnting between the normal and the type 1 extreme value distributions.
Journal Article Inferences for the Cauchy distribution based on maximum likelihood estimators Get access GERALD HAAS, GERALD HAAS University of Missouri at Rolla Search for other works by this author on: Oxford Academic Google Scholar LEE BAIN, LEE BAIN University of Missouri at Rolla Search for other works by this author on: Oxford Academic Google Scholar CHARLES ANTLE CHARLES ANTLE Pennsylvania State University Search for other works by this author on: Oxford Academic Google Scholar Biometrika, Volume 57, Issue 2, August 1970, Pages 403–408, https://doi.org/10.1093/biomet/57.2.403 Published: 01 August 1970 Article history Received: 01 September 1969 Revision received: 01 January 1970 Published: 01 August 1970
Abstract This paper presents a method of constructing conservative joint confidence regions for the parameters of a linear model and the volumes of these regions are compared with the volume of an exact confidence region. Confidence regions on the entire surface are obtained from the confidence region on the parameters. The procedures are illustrated for the case of a simple linear model.
A study was conducted to determine whether test specimen produced using silicone (Si) rubber molds affect the Superpave low-temperature grading parameters, S(60) and m-value, determined using the bending beam rheometer (BBR; AASHTO-TP1). Six laboratories participated in this study. Four unmodified asphalt binders with low-temperature performance grades between -6°C and -30°C were tested. The thickness of each test specimen was measured using the “flip” technique. The results showed that the Si rubber molds produce significantly thinner beams than the required 6.35 ± 0.05-mm thickness. This affects the calculated S(60) values but not the m-value. Aluminum (Al) molds were also used in this study according to AASHTO-TP1. It was found that the stiffness values S(60) calculated in the BBR test are significantly lower when Si rubber molds are used to produce test specimen as compared with those test beams produced using Al molds. The cause is the peculiarity of the calculation process specified in AASHTO-TP1. AASHTO-TP1 requires that the thickness of the test beam must always be assumed to be 6.35 mm for computing S(60). This study suggests that the test beams produced using the Si rubber molds are significantly thinner; therefore, assuming a thickness of 6.35 mm to calculate S(60) for these beams produces a negative bias. It is shown that when thickness is measured using the flip technique and the measured thickness is then used to calculate S(60), there is no statistically significant difference between the S(60) values of the test beams produced using either the Al or Si rubber molds.
