Joseph Awoamim Yacim and Douw Gert Brand Boshoff
The paper aims to investigate the application of particle swarm optimisation and back propagation in weights optimisation and training of artificial neural networks within the…
Abstract
Purpose
The paper aims to investigate the application of particle swarm optimisation and back propagation in weights optimisation and training of artificial neural networks within the mass appraisal industry and to compare the performance with standalone back propagation, genetic algorithm with back propagation and regression models.
Design/methodology/approach
The study utilised linear regression modelling before the semi-log and log-log models with a sample of 3,242 single-family dwellings. This was followed by the hybrid systems in the selection of optimal attribute weights and training of the artificial neural networks. Also, the standalone back propagation algorithm was used for the network training, and finally, the performance of each model was evaluated using accuracy test statistics.
Findings
The study found that combining particle swarm optimisation with back propagation in global and local search for attribute weights enhances the predictive accuracy of artificial neural networks. This also enhances transparency of the process, because it shows relative importance of attributes.
Research limitations/implications
A robust assessment of the models’ predictive accuracy was inhibited by fewer accuracy test statistics found in the software. The research demonstrates the efficacy of combining two models in the assessment of property values.
Originality/value
This work demonstrated the practicability of combining particle swarm optimisation with back propagation algorithms in finding optimal weights and training of the artificial neural networks within the mass appraisal environment.
Details
Keywords
P. McCluskey, R. Munamarty and M. Pecht
Plastic ball grid arrays (PBGAs) have been gaining increasing industry acceptanceas potentially the lowest cost packages for high pin count applications. The main factors for…
Abstract
Plastic ball grid arrays (PBGAs) have been gaining increasing industry acceptance as potentially the lowest cost packages for high pin count applications. The main factors for their success include the inherent low cost of plastic packages, the high yields achievable with existing assembly equipment, the density advantages of an area array and, in many cases, superior electrical performance. The reliability of these packages with respect to solder joint fatigue has been extensively studied; however, little has been published on their reliability with respect to moisture‐related failure mechanisms during assembly, such as popcorning. In this study, two types of plastic ball grid array packages, a 225‐lead full matrix array and a 256‐lead perimeter array, which were preconditioned by baking at 125°C for 24 hours followed by exposure to 85°C and 30% relative humidity for 168 hours, were then subjected to three passes of simulated infra‐red reflow at 1°C/s and 0.67°C/s. Subsequent examination using scanning acoustic microscopy and environmental scanning electron microscopy revealed considerable delamination and cracking in both package types reflowed at the higher ramp rate. This damage originated in the die attach and propagated along the weakest interfaces in the package. At the lower ramp rate, a much smaller amount of delamination was observed. This suggests that there is a critical ramp rate below which popcorn cracking can be inhibited.
Details
Keywords
R. Munamarty, P. McCluskey, M. Pecht and L. Yip
Two types of Plastic ball Grid array packages, a225‐lead full matrix array and a 256‐lead perimeter array, were subjected to 168 hours of moisturepreconditioning at 85°C and 30…
Abstract
Two types of Plastic ball Grid array packages, a 225‐lead full matrix array and a 256‐lead perimeter array, were subjected to 168 hours of moisture preconditioning at 85°C and 30% relative humidity followed by simulated infra‐red reflow at temperature ramp rates of 1°C/s and 0.67°C/s. The packages were subsequently examined for delamination and cracking using scanning acoustic microscopy and environmental scanning electron microscopy. At the higher ramp rate, delamination and cracking were observed in both package types, originating in the die attach and propagating along the weakest interfaces. At the lower ramp rate, a small amount of delamination was observed. This suggests that there is a critical ramp rate below which popcorn cracking does not occur.
Details
Keywords
N. Geren, M. Çakirca and M. Bayramoğlu
To design a flexible integrated robotic assembly and rework (remanufacturing) cell for assembly, selective assembly and rework of advanced surface mount components (SMCs) using…
Abstract
Purpose
To design a flexible integrated robotic assembly and rework (remanufacturing) cell for assembly, selective assembly and rework of advanced surface mount components (SMCs) using the generic methodology developed in this paper.
Design/methodology/approach
Manual rework procedures are investigated for all advanced SMCs. General and specific component‐related rework considerations are obtained and necessary tooling candidates for automation are determined. This is followed by determination of the specific automated rework procedure and selection of suitable tooling for automated robotic rework and generation and evaluation of design concepts.
Findings
The developed methodology, which considers the reflow tool at the centre of the development process, has worked well in designing a flexible integrated robotic assembly and rework cell.
Practical implications
This study identified the rework requirements for advanced SMCs, the essential features for rework reflow tools, criteria for comparing reflow tools, and a generic procedure for design and concept selection.
Originality/value
It provides valuable knowledge for designers of flexible integrated robotic assembly and rework cells for assembly, selective assembly and rework of advanced SMCs.
Details
Keywords
The purpose of the study is to examine the efficiency of linear, nonlinear and artificial neural networks (ANNs), in predicting property prices.
Abstract
Purpose
The purpose of the study is to examine the efficiency of linear, nonlinear and artificial neural networks (ANNs), in predicting property prices.
Design/methodology/approach
The present study uses a dataset of 1,468 real estate transactions from 2020 to 2022, obtained from the Department of Property Taxes of Republic of Kosovo. Beginning with a fundamental linear regression model, the study tackles the question of overlooked nonlinearity, employing a similar strategy like Peterson and Flanagan (2009) and McCluskey et al. (2012), whereby ANN's predictions are incorporated as an additional regressor within the ordinary least squares (OLS) model.
Findings
The research findings underscore the superior fit of semi-log and double-log models over the OLS model, while the ANN model shows moderate performance, contrary to the conventional conviction of ANN's superior predictive power. This is notably divergent from the prevailing belief about ANN's superior predictive power, shedding light on the potential overestimation of ANN's efficacy.
Practical implications
The study accentuates the importance of embracing diverse models in property price prediction, debunking the notion of the ubiquitous applicability of ANN models. The research outcomes carry substantial ramifications for both scholars and professionals engaged in property valuation.
Originality/value
Distinctively, this research pioneers the comparative analysis of diverse models, including ANN, in the setting of a developing country's capital, hence providing a fresh perspective to their effectiveness in property price prediction.
Details
Keywords
Joseph Awoamim Yacim and Douw Gert Brand Boshoff
The paper introduced the use of a hybrid system of neural networks support vector machines (NNSVMs) consisting of artificial neural networks (ANNs) and support vector machines…
Abstract
Purpose
The paper introduced the use of a hybrid system of neural networks support vector machines (NNSVMs) consisting of artificial neural networks (ANNs) and support vector machines (SVMs) to price single-family properties.
Design/methodology/approach
The mechanism of the hybrid system is such that its output is given by the SVMs which utilise the results of the ANNs as their input. The results are compared to other property pricing modelling techniques including the standalone ANNs, SVMs, geographically weighted regression (GWR), spatial error model (SEM), spatial lag model (SLM) and the ordinary least squares (OLS). The techniques were applied to a dataset of 3,225 properties sold during the period, January 2012 to May 2014 in Cape Town, South Africa.
Findings
The results demonstrate that the hybrid system performed better than ANNs, SVMs and the OLS. However, in comparison to the spatial models (GWR, SEM and SLM) the hybrid system performed abysmally under with SEM favoured as the best pricing technique.
Originality/value
The findings extend the debate in the body of knowledge that the results of the OLS can significantly be improved through the use of spatial models that correct bias estimates and vary prices across the different property locations. Additionally, utilising the result of the hybrid system is thus affected by the black-box nature of the ANNs and SVMs limiting its use to purposes of checks on estimates predicted by the regression-based models.
Details
Keywords
Mohammad A. Gharaibeh and Jürgen Wilde
The purpose of this paper is to investigate the thermomechanical response of four well-known lead-free die attach materials: sintered silver, sintered nano-copper particles…
Abstract
Purpose
The purpose of this paper is to investigate the thermomechanical response of four well-known lead-free die attach materials: sintered silver, sintered nano-copper particles, gold-tin solders and silver-tin transient liquid phase (TLP) bonds.
Design/methodology/approach
This examination is conducted through finite element analysis. The mechanical properties of all die attach systems, including elastic and Anand creep parameters, are obtained from relevant literature and incorporated into the numerical analysis. Consequently, the bond stress-strain relationships, stored inelastic strain energies and equivalent plastic strains are thoroughly examined.
Findings
The results indicate that silver-tin TLP bonds are prone to exhibiting higher inelastic strain energy densities, while sintered silver and copper interconnects tend to possess higher levels of plastic strains and deformations. This suggests a higher susceptibility to damage in these metallic die attachments. On the other hand, the more expensive gold-based solders exhibit lower inelastic strain energy densities and plastic strains, implying an improved fatigue performance compared to other bonding configurations.
Originality/value
The utilization of different metallic material systems as die attachments in power electronics necessitates a comprehensive understanding of their thermomechanical behavior. Therefore, the results of the present paper can be useful in the die attach material selection in power electronics.
Details
Keywords
Marcelo Luis Barbosa dos Santos
When a concept is diffusely defined or, as this article argues, “taken for granted”, it becomes very difficult to track such concept on the literature and have some continuity as…
Abstract
Purpose
When a concept is diffusely defined or, as this article argues, “taken for granted”, it becomes very difficult to track such concept on the literature and have some continuity as researchers build on top of previous results. This article proposes a definition for user-generated content, a term that though has lost some saliency, stands in the center or the social media phenomenon, so it should not be disregarded as an object of study.
Design/methodology/approach
Celebrating 20 years of the concept, this research performs a multidisciplinary literature review of 61 academic articles on UGC. Through deconstruction of the acronym UGC, it builds on the present converging, conflicting and diverging definitions and/or approaches to UGC on an attempt to consolidate a broader definition that encompasses the complexities of the phenomenon in a context of consolidation of social media, to be applied to social sciences.
Findings
Following the present analysis, UGC is defined as any kind of text, data or action performed by online digital systems users, published and disseminated by the same user through independent channels, that incur an expressive or communicative effect either on an individual manner or combined with other contributions from the same or other sources.
Originality/value
This is the first academic effort that aims to create an in-depth dialogue over the different approaches to UGC across disciplines on the social sciences field. It should help reignite interest in the acronym, which got somehow eclipsed by the broader field of social media; whilst without UGC, social media would not exist or would not have the same social impact it does in its current form. Analogously, UGC as a topic of research has been deeply affected by the emergence and consolidation of Social Media. As this debate evolves, this contribution should be helpful as a reference to operationalize UGC on future research.
Peer review
The peer-review history for this article is available at: https://publons.com/publon/10.1108/OIR-06-2020-0258
Details
Keywords
Rotimi Boluwatife Abidoye and Albert P.C. Chan
The predictive accuracy and reliability of artificial intelligence models, such as the artificial neural network (ANN), has led to its application in property valuation studies…
Abstract
Purpose
The predictive accuracy and reliability of artificial intelligence models, such as the artificial neural network (ANN), has led to its application in property valuation studies. However, a large percentage of such previous studies have focused on the property markets in developed economies, and at the same time, effort has not been put into documenting its research trend in the real estate domain. The purpose of this paper is to critically review the studies that adopted ANN for property valuation in order to present an application guide for researchers and practitioners, and also establish the trend in this research area.
Design/methodology/approach
Relevant articles were retrieved from online databases and search engines and were systematically analyzed. First, the background, the construction and the strengths and weaknesses of the technique were highlighted. In addition, the trend in this research area was established in terms of the country of origin of the articles, the year of publication, the affiliations of the authors, the sample size of the data, the number of the variables used to develop the models, the training and testing ratio, the model architecture and the software used to develop the models.
Findings
The analysis of the retrieved articles shows that the first study that applied ANN in property valuation was published in 1991. Thereafter, the technique received more attention from 2000. While a quarter of the articles reviewed emanated from the USA, the rest were conducted in mostly developed countries. Most of the studies were conducted by universities scholars, while very few industry practitioners participated in the research works. Also, the predictive accuracy of the ANN technique was reported in most of the papers reviewed, but a few reported otherwise.
Research limitations/implications
The articles that are not indexed in the search engines and databases searched and also not available in the public domain might not have been captured in this study.
Practical implications
The findings of this study reveal a gap between the valuation practice in developed and developing property markets and also the contributions of real estate practitioners and universities scholars to real estate research. A paradigm shift in the valuation practice in developing nations could lead to achieving a sustainable international valuation practice.
Originality/value
This paper presents the trend in this research area that could be useful to real estate researchers and practitioners in different property markets around the world. The findings of this study could also encourage collaboration between industry professionals and researchers domiciled in both developed and developing countries.
Details
Keywords
Mohammad A. Gharaibeh and Jürgen Wilde
In power electronics, there are various metallic material systems used as die attachments. The complete understanding of the thermomechanical behavior of such interconnections is…
Abstract
Purpose
In power electronics, there are various metallic material systems used as die attachments. The complete understanding of the thermomechanical behavior of such interconnections is very important. Therefore, this paper aims to examine the thermomechanical response of four famous die attach materials, including sintered silver, sintered nano-copper particles, gold-tin solders and silver-tin transient liquid phase (TLP) bonds, using nonlinear finite element analysis.
Design/methodology/approach
During the study, the mechanical properties of all die attach systems, including elastic and viscoplasticity parameters, are obtained from literature studies and hence incorporated into the numerical analysis. Subsequently, the bond stress–strain relationships, stored inelastic strain energies and equivalent plastic strains are thoroughly examined.
Findings
The results showed that the silver-tin TLP bonds are more likely to develop higher inelastic strain energy densities, while the sintered silver and copper interconnects would possess higher plastic strains and deformations. Suggesting higher damage to such metallic die attachments. The expensive gold-based solders have developed least inelastic strain energy densities and least plastic strains as well. Thus, they are expected to have improved fatigue performance compared to other bonding configurations.
Originality/value
This paper extensively investigates and compares the mechanical and thermal response of various metallic die attachments. In fact, there are no available research studies that discuss the behavior of such important die attachments of power electronics when exposed to mechanical and thermomechanical loads.