Papers by Keyword: Multiple Regression Analysis

Abstract: Conventionally, the sin²ψ method has been used as X-ray stress measurement. However, in recent years, the XRD² method and the cos α method have been put into practical use and spreading. In addition, the Fourier analysis method that shares the same measurement principle of the cos α method has been developed and is attracting attention. Therefore, in this paper, the Fourier analysis method is examined from the measurement theory and the measurement accuracy is investigated. It is reviewed that the basic equation is a finite Fourier series, and that stress can be determined from the Fourier coefficients by using coordinate transformations. Then, while comparing it with the multiple regression analysis, the accuracy of the Fourier analysis method is discussed by using numerical calculations.

Abstract: The estimation of the ultimate capacity of rectangular or circular shaped steel tubular members filled with concrete, such as columns, beams, and beam-column connections, requires a detailed structural study to be carried out. Therefore, identify the concrete strength the member subjected to axial-load only. Using the Levenberg-Marquardt artificial neural network, this paper investigates the concrete-filled steel tubular (CFT) members axial strength. 201 experimental specimens were collected from the literature to obtain the best results, and a wide range of geometric and material properties of CFT members were included. The proposed design and specimens illustrate the practicality and effectiveness of the chosen CFT column approach to classify real structural results.

Abstract: Semiconductor manufacturing management system was developed and grown up over the past decades. In order to increase the product yield and enhance the production productivity, cluster tools became the main stream in modern wafer fabrication factories which occupies over 50% of production equipment. Generally, cluster tools are integrated by several components including robots, vacuum chambers (Load locks) and single-wafer process chambers in a module and can be treated as a small factory. The throughput estimation before recipe release is very difficult. However, it is necessary and important for the planning activity. In this work, a throughput estimation model for cluster tools is proposed. The Multiple Regression Analysis is applied to develop a set of throughput estimation equations. A simulation model of cluster equipment including 3 single-wafer process chambers are built to get the historical throughput data for the regression analysis. From the Multiple Regression Analysis, it reveals that different numbers of recipes processed in the same time have to develop different regression model. The major factors in the regression model include numbers of load ports and process time of each recipe. Furthermore, a set of recipes are used to test the accuracy of estimation. Based on the testing results, they revealed that the MAPE is under 3% and the estimation model is accepted in practice to forecast the throughput of recipes for the planning activities.

Abstract: The blanking operation have several die design parameters which affect the quality of the blank and its productivity. The main input parameters are sheet thickness and the punch and die clearance and the dependent output parameters are tool life and the burr height. The selected values should be in optimal value. The optimum value is achieved by using the genetic algorithm. The genetic algorithm is an optimization process to find the better results as an output. Then the development of mathematical modeling by using the equations derived from the multiple regression analysis is performed. It is achieved by converting the linear equations into the matrix form and then solving it using mathematical relations. This output is compared with the genetic algorithm results, to get the better results.

Abstract: Construction industry is one of the most important sectors for the Malaysian economic growth where it remains as the fastest growing activity for several years. The issue of rapid country development which has positive influence of generating high construction waste is very commonly related to this industry. One of the major contributors of construction waste is steel. Steel is expensive and can contribute to overall project cost implication. Hence, the amount of steel waste needsto be quantifiedat the initial stage. Thus,steel waste quantification model is an important element to be considered for better waste management. By using the data collected in the Central region in Malaysia, an empirical model, Estimation Model of Steel Waste Material has been developed to estimate the amount of steel waste generation based on certain variables. This study aims to investigate and verify the effectiveness of the model by using the data collected from the Northern region in Malaysia. The value of the Mean Absolute Percentage Error (MAPE), 23.02% indicates that the model performance is common and acceptable. Thus, the model has good potential to be adopted by the industry practitioners to plan a better and more effective waste management.

Abstract: In recent years, the regional difference of the real estate price in western area gradually increases, making the effects of macroeconomic regulation and control is different among regions. In this paper, based on the theory of market supply and demand, first of all, the indicators are selected that can reflect the conditions of real estate market, then western provincial capital cities’ real estate markets are divided into three regions through cluster analysis. Secondly, the natural logarithmic model is established concerning the commercial housing price and the factors. Multiple regression analysis is performed for each area, in order to analyze the reasons of property price differences in three regions.

Abstract: The service quality of the online banking information security product has become the important bottleneck of every financial institution to further explore online market and development potential customers. Through study the domestic and overseas scholars about the quality of service fields putting the part of the research extends to the service quality of the online banking information security product measurement process, and trying to build the measurement model suit for the service quality of the online banking information security product. We survey the university student of shaanxi Xian and use the internet to investigation while using the factor analysis and multiple regression method to verify the evaluation model and assessment index of rationalities. We extract the main influence factors of the service quality of the online bank information security products, preceding a deeper analysis. Finally, we derive the main influence factor of the service quality of the online bank information security products.

Abstract: Risks and variables are frequently encountered when executing special projects. If these risks are handled effectively and the variables are controlled with definite and effective manners, the probability of success is enhanced. This study aims to explore and analyze these critical elements and find out the influential factors to success. Findings reveal that at the planning stage of the special project, cost-estimation is deemed the most important element by the overall interviewees; meanwhile, at the stage of deployment, construction costs and administrative management have positive effects on the overall satisfaction level, nevertheless, control-and-guarantee has a negative effective on the overall satisfaction.

Abstract: Artificial intelligent techniques are being actively applied in many applications. With their powerful learning capability of neural networks and reducing the optimizing search space by prior knowledge rules of Fuzzy systems have been proven to be rather efficiency. In this research, the hybrid Neuro-Fuzzy system (NF) is proposed to be utilized as a predictor of the Grade Point Average (GPA) of students for future planning where the Radial Basis Function (RBF) is implemented as a neuro-fuzzy system. The NFs parameters consisted of centre and width of the Gaussian membership function and weight between input layer and output layer are automatically tuned by using Genetic Algorithms (GA) referred as NF-GA. The collected data is then tested and trained through NF-GA system with Minimum Mean Square Error (MMSE) technique. It has been shown that our proposed model is capable of prediction GPA by accurately 93%.The performance comparison between the proposed NF-GA and Multiple Regression Analysis (MRA) gives performance significantly by reducing the average error of the prediction down to 10%. Keywords: Neuro-fuzzy system, Genetic Algorithms, Multiple Regression Analysis, Radial Basis Function.

Abstract: This study analyzes oil palm production from 2001 - 2008 in Edo State, with a view of establishing the future of Nigerian Agricultural oil palm industries. The specific objectives were to identify and quantify the factors influencing the production of oil palm, reveal the constraints to the production of oil palm; analyze the viability of oil palm production as an investment and predict the future of oil palm enterprise in Edo State. Three Local Government Areas of high oil palm concentration in Edo State were covered. Ninety selected oil palm producers from nine villages in the study area were selected. Frequency distributions and multiple regression analysis were used to analyze the results. The results showed that for the period under review (2001 - 2008) on a per hectare basis, seedlings cost, and labour cost were significant inputs, positively influencing the production of oil palm in the Edo State. A growth rate of 6.2% and 23.5% were projected for oil palm with reference to palm oil and fresh fruit bunches (FFB), respectively, up to the year 2020, an indication that oil palm production in the State is viable and the future is bright. The study revealed that inadequate storage facilities, poor planting materials and lack of government assistance, among others, are major bottlenecks in oil palm production in Edo State. However, since oil palm production in Edo State is viable, it is an indication that the future of Nigerian Agricultural Industries is bright.