Case studies on vibration based damage identification : multi-criteria approach

Vibration based damage identification methods examine the changes in primary modal parameters or quantities derived from modal parameters. As one method may have advantages over the other under some circumstances, a multi-criteria approach is proposed. Case studies are conducted separately on beam, plate and plate-on-beam structures. Using the numerically simulated modal data obtained through finite element analysis software, algorithms based on flexibility and strain energy changes before and after damage are obtained and used as the indices for the assessment of the state of structural health. Results show that the proposed multi-criteria method is effective in damage identification in these structures.

Identification and analysis of business and software services : A consolidated approach

Although the benefits of service orientation are prevalent in literature, a review, analysis, and evaluation of the 30 existing service analysis approaches presented in this paper have shown that a comprehensive approach to the identification and analysis of both business and supporting software services is missing. Based on this evaluation of existing approaches and additional sources, we close this gap by proposing an integrated, consolidated approach to business and software service analysis that combines and extends the strengths of the examined methodologies.

Identification of critical input variables for risk-based cost estimates for road maintenance and rehabilitation

An estimation of costs for maintenance and rehabilitation is subject to variation due to the uncertainties of input parameters. This paper presents the results of an analysis to identify input parameters that affect the prediction of variation in road deterioration. Road data obtained from 1688 km of a national highway located in the tropical northeast of Queensland in Australia were used in the analysis. Data were analysed using a probability-based method, the Monte Carlo simulation technique and HDM-4’s roughness prediction model. The results of the analysis indicated that among the input parameters the variability of pavement strength, rut depth, annual equivalent axle load and initial roughness affected the variability of the predicted roughness. The second part of the paper presents an analysis to assess the variation in cost estimates due to the variability of the overall identified critical input parameters.

Identification of key environmental issues for building materials

Manufacture, construction and use of buildings and building materials make a significant environmental impact internally (inside the building), locally (neighbourhood) and globally. Life cycle assessment (LCA) methodology is being applied for evaluating the environmental impact of building/or building materials. One of the major applications of LCA is to identify key issues of a product system from cradle to grave. Key issues identified in an LCA lead one to the right direction in assessing the environmental aspects of a product system and help to identify the areas for improvement of the environmental performance of a product as well. The purpose of this paper is to suggest two methods for identifying key issues using an integrated tool (LCADesign), which has been developed to provide a method of determining the best alternative for reducing environmental impacts from a building or building materials, and compare both methods in the case study. This paper assists the designers or marketers related to building or building materials in their decision making by giving information on activities or alternatives which are identified as key issues for environmental impacts.

Combination strategies for a factor analysis phone-conditioned speaker verification system

This work aims to take advantage of recent developments in joint factor analysis (JFA) in the context of a phonetically conditioned GMM speaker verification system. Previous work has shown performance advantages through phonetic conditioning, but this has not been shown to date with the JFA framework. Our focus is particularly on strategies for combining the phone-conditioned systems. We show that the classic fusion of the scores is suboptimal when using multiple GMM systems. We investigate several combination strategies in the model space, and demonstrate improvement over score-level combination as well as over a non-phonetic baseline system. This work was conducted during the 2008 CLSP Workshop at Johns Hopkins University.

Student success : the identification and support of first year university students at risk of attrition

The engagement behaviour of 1,524 student-enrolments (“students”) in five first year units was monitored and 608 (39.9%) were classified as “at risk” using the criterion of not submitting or failing their first assignment. Of these, 327 (53.8%) were successfully contacted (i.e., spoken to by phone) and provided with advice and/or referral to learning and personal support services while the remaining 281 (46.2%) could not be contacted. Nine hundred and sixteen students (60.1%) were classified as “not at risk.” Overall, the at risk group who were contacted achieved significantly higher end-of-semester final grades than, and persisted (completed the unit) at more than twice the rate of, the at risk group who were not contacted. There were variations among the units which were explained by the timing of the first assignment, specific teaching-learning processes and the structure of the curriculum. Implications for curriculum design and supporting first year students within a personal, social and academic framework are discussed.

Identification of barriers to and facilitators for the implementation of occupational road safety initiatives

To explore potential barriers to and facilitators for implementing occupational road safety initiatives, in-depth interviews were conducted with personnel from four major Australian organizations. Twenty-four participants were involved in the interviews comprising 16 front line employees and eight managers. The interviews identified that employees perceived six organizational characteristics as potential barriers to implementing occupational road safety initiatives. These included: prioritisation of production over safety; complacency towards occupational road risks; insufficient resources; diversity; limited employee input in safety decisions; and a perception that road safety initiatives were an unnecessary burden. Of these organizational characteristics, prioritisation of production over safety and complacency were the most frequently cited barriers. In regards to facilitators, participants perceived three organizational characteristics as potential facilitators to implementing occupational road safety initiatives. These included: management commitment; the presence of existing systems that could support the implementation of initiatives; and supportive relationships. Of these organizational characteristics, management commitment was the most frequently cited facilitator.

Improved GMM-based speaker verification using SVM-driven impostor dataset selection

The problem of impostor dataset selection for GMM-based speaker verification is addressed through the recently proposed data-driven background dataset refinement technique. The SVM-based refinement technique selects from a candidate impostor dataset those examples that are most frequently selected as support vectors when training a set of SVMs on a development corpus. This study demonstrates the versatility of dataset refinement in the task of selecting suitable impostor datasets for use in GMM-based speaker verification. The use of refined Z- and T-norm datasets provided performance gains of 15% in EER in the NIST 2006 SRE over the use of heuristically selected datasets. The refined datasets were shown to generalise well to the unseen data of the NIST 2008 SRE.

Data-driven impostor selection for T-norm score normalisation and the background dataset in SVM-based speaker verification

A data-driven background dataset refinement technique was recently proposed for SVM based speaker verification. This method selects a refined SVM background dataset from a set of candidate impostor examples after individually ranking examples by their relevance. This paper extends this technique to the refinement of the T-norm dataset for SVM-based speaker verification. The independent refinement of the background and T-norm datasets provides a means of investigating the sensitivity of SVM-based speaker verification performance to the selection of each of these datasets. Using refined datasets provided improvements of 13% in min. DCF and 9% in EER over the full set of impostor examples on the 2006 SRE corpus with the majority of these gains due to refinement of the T-norm dataset. Similar trends were observed for the unseen data of the NIST 2008 SRE.