Home About us Contact
|
Home About us Contact | ||
|
|
||
Proposed Criterion (proposed + criterion)
Selected AbstractsA Multicriteria Decision Support Methodology for Annual Rehabilitation Programs of Water NetworksCOMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING, Issue 7 2007Pascal Le Gauffre A first set of criteria is formulated for the purpose of comparing and ranking rehabilitation projects. Each proposed criterion is a measure of a particular impact of the condition of a pipe. The ELECTRE TRI method is implemented for defining rehabilitation priorities. Two reference profiles are used to define the limits of three categories associated with three increasing priority levels. With these two reference profiles, applying the ELECTRE TRI method to an asset stock (a set of pipes that are candidates for rehabilitation) means assigning each pipe to one of six possible priority groups. A second set of criteria, based on the concept of efficiency, is proposed for comparing alternative rehabilitation programs (subsets of the asset stock). [source] Strain field measurements of rubber by image analysis and design criteria for laminated rubber bearings (LRB)EARTHQUAKE ENGINEERING AND STRUCTURAL DYNAMICS, Issue 4 2004Chamindalal Sujeewa Lewangamage Abstract Although seismic isolation rubber bearings in bridges and buildings have proven to be a very effective passive method for reducing earthquake-induced forces, a detailed mechanical modeling of the rubber that is used in bearings under large strains has not been established. Therefore, a 3D model of failure behavior and the design criteria for the safety evaluation of seismic isolation bearings have not yet been developed. This paper presents: (1) correlation-based template-matching algorithms to measure large strain fields of continua; (2) a failure criterion for rubber; and (3) the design criteria for the safety evaluation of laminated algorithms, data-validation algorithms were developed and implemented to eliminate possible unrealistic displacement vectors present in the measured displacement field. The algorithms were successfully employed in the strain field measurement of LRB and rubber materials that are subjected to failure. The measured local strains for rubber material at failure were used to develop a failure criterion for rubber. The validity of the proposed criterion was evaluated by applying it to the LRB; the criterion was introduced into a 3D finite element model of LRB, compared with the experimental results of bearings failure, and verified. Finally, design criteria are proposed for LRB for the safety evaluation. Copyright © 2003 John Wiley & Sons, Ltd. [source] Numerical modeling of brittle fracture in porous CFC-materialsPROCEEDINGS IN APPLIED MATHEMATICS & MECHANICS, Issue 1 2006Romana Piat Both energy and stress criteria are necessary conditions for fracture but neither the one nor the other is sufficient. A combination of these criteria is proposed in [1]. This combined criterion is used for numerical simulation of crack propagation by the 4-point bending test in porous materials. Examples of such materials are carbon-carbon composites (CFC) [2, 3]. Micrographs of the cross-sections of these materials are used for FEM modeling of the crack propagation on the basis of the proposed criterion. Results of the numerical modeling are compared with experimental results. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim) [source] Model selection for generalized linear models with factor-augmented predictorsAPPLIED STOCHASTIC MODELS IN BUSINESS AND INDUSTRY, Issue 3 2009Tomohiro Ando Abstract This paper considers generalized linear models in a data-rich environment in which a large number of potentially useful explanatory variables are available. In particular, it deals with the case that the sample size and the number of explanatory variables are of similar sizes. We adopt the idea that the relevant information of explanatory variables concerning the dependent variable can be represented by a small number of common factors and investigate the issue of selecting the number of common factors while taking into account the effect of estimated regressors. We develop an information criterion under model mis-specification for both the distributional and structural assumptions and show that the proposed criterion is a natural extension of the Akaike information criterion (AIC). Simulations and empirical data analysis demonstrate that the proposed new criterion outperforms the AIC and Bayesian information criterion. Copyright © 2009 John Wiley & Sons, Ltd. [source] | ||||||||||