Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

How has Japan become a major economic power, a world leader in the automotive and electronics industries? What is the secret of their success? The consensus has been that, though the Japanese are not particularly innovative, they are exceptionally skilful at imitation, at improving products that already exist. But now two leading Japanese business experts, Ikujiro Nonaka and Hiro Takeuchi, turn this conventional wisdom on its head: Japanese firms are successful, they contend, precisely because they are innovative, because they create new knowledge and use it to produce successful products and technologies. Examining case studies drawn from such firms as Honda, Canon, Matsushita, NEC, 3M, GE, and the U.S. Marines, this book reveals how Japanese companies translate tacit to explicit knowledge and use it to produce new processes, products, and services.

The knowledge-creating company : how Japanese companies create the dynamics of innovation

On Educational psychology: A cognitive view.

https://homepages.fhv.at/hgb/New-Papers/CMAME07_BS07b.pdf

Robust Optimization - A Comprehensive Survey

Sustainability in the construction industry: A review of recent developments based on LCA

This paper deals with passive vibration control using a bi-unit impact damper, also referred to as bi-unit impact vibration absorber. The main contribution of this work is to experimentally analyze the performance of a bi-unit impact vibration absorber using digital image processing. With the use of this technique, two or three bodies will be tracked simultaneously, namely, the primary system, and either the response of one or two impacting steel balls. The absorption pattern of the bi-unit is studied under free vibrations, constant frequency excitation, and sweep excitations. These results reveal that the use of digital image processing can eliminate the use of existing limitations which can sometimes change the dynamics of the system. The optical technique used captures reliably the detailed vibroimpact dynamics of the bodies. The detailed dynamic patterns of the particles during varying levels of energy absorption from the primary system are discussed in context of published theoretical results.

Performance of a Bi-unit Damper Using Digital Image Processing

Currently plastic gears are widely used in industry, and not only for lightly loaded applications like household appliances, tools, and toys, but also in the more demanding areas of machinery in automotive applications. However there is a need to investigate important properties such as load capacity, endurance, cost, life, stiffness and wear. Tooth wear is one of the major failure modes in plastic gears just like with steel gears. This paper focuses on the simulation of wear for standard and non-standard gears using an analytical approach. A numerical model for wear prediction of gear pairs is developed. A wear model based on Archard’s equation is employed to predict wear depth. The variation of the contact load generated by the cumulative tooth profile wear is simulated and examined. A MATLAB-based virtual tool is developed to analyze wear behavior of standard and non-standard spur gears depending on various gear parameters. In this paper, this virtual tool is introduced with numerical examples.Copyright © 2013 by ASME

A Virtual Tool for Wear Simulation of Plastic Gear Pairs

Failure of a thermally loaded bimaterial specimen was investigated in this research. A high-stress gradient due to thermal loading occurs around the free edge of the interface of the bimaterial specimen. This high-stress gradient plays an important role in the failure of the specimen. The Weibull failure theory has been shown to be unable to account for high-stress gradients in externally loaded bimaterial specimens. The objective of this work is to develop a weight function method to calculate the effective stress intensity factors in the vicinity of a high-stress gradient in a thermally loaded bimaterial specimen, and to develop a modified Weibull failure theory to handle the high-stress gradient. It was found that the modified Weibull failure theory generated monotonous trends for the probability of failure with respect to increasing Weibull moduli, as demonstrated in the literature.

Applying modified Weibull failure theory to bimaterial specimen under thermal loading

Despite the increasing demand for offshore energy, structural components of offshore wind turbines (OWT), such as the tower and foundation, are considered the most critical parts of the turbine. In fact, uncertainties regarding load conditions, soil and structural properties highly undermine the OWT structural reliability. In this scenario, in order to obtain more accurate results, rigorous probabilistic analyses are necessary. In this study, a probabilistic analysis of the dynamic response of a monopile OWT is conducted by using a systematic uncertainty quantification (UQ) framework to deal with the uncertainty assessment of the model input parameters. The proposed dynamic model computes the dynamic response of the turbine due to wind and waves loads on the monopile structure utilizing a simple cantilever beam analytical model. The distributions of the model input parameters are determined using (1) non-parametric statistics for a large dataset, and (2) the maximum entropy principle for a small dataset. Monte Carlo simulations are performed to propagate the uncertainties of the model inputs and to determine the system reliability expressed in terms of their probability of failure for the serviceability limit state design criterion. Finally, to demonstrate the shortcomings of traditional approaches that assume standard distributions to model uncertainties, a UQ approach modeling the uncertainties of the parameters using normal distributions is contrasted with our framework. From the results, significant differences between the distribution shape and values of the probability of failure can be observed; thus, it demonstrates the importance of developing probabilistic frameworks with systematic UQ to have more realistic approximations of the reliability of the OWT structure.

/pdf/probabilistic-design-and-uncertainty-quantification-of-the-2jph8c8c3y.pdf

Probabilistic Design and Uncertainty Quantification of the Structure of a Monopile Offshore Wind Turbine

In the capstone design courses, the instructor can choose among different strategies of assigning students to project teams. One of these strategies allows the students select the project they would like to work on from a list. This sometimes proceeds in an ad-hoc manner. Currently, literature offers limited research that looks at the reasons why students select certain projects. Furthermore, the work examining the impact of this selection on team performance is also limited. The objective of this study is to examine the impact of the students’ project selection on team performance. This is carried out by (1) investigating the rationalization students undertake during the selection process, and (2) analyzing data concerning the influence of the selection on team performance. Two instruments are presented in this paper that can be used to analyze the impact of student selection of design projects on the team performance. These instruments are also used to track the types of projects the students select, such as faculty projects, industry supported projects, student initiated projects, and projects for design competitions. A systematic methodology, based on the students’ rankings of all the projects for assigning students to their preferred choice of projects, is also presented. Whereas the data presented shows that students generally read carefully the project description, the majority of students prefer the project clients to make short presentations. To analyze the impact of project choice on team performance, four categories, based on the student project choice, were proposed. Teams whose majority did not get their first choice of project, showed the largest drop between the mid-semester peer ratings compared to end-of-semester peer ratings. This study was performed at two universities.

Stephen Ekwaro-Osire

Papers

Performance of a Bi-unit Damper Using Digital Image Processing

A Virtual Tool for Wear Simulation of Plastic Gear Pairs

Applying modified Weibull failure theory to bimaterial specimen under thermal loading

Probabilistic Design and Uncertainty Quantification of the Structure of a Monopile Offshore Wind Turbine

Impact Of Student Selection Of Design Projects On Team Performance