Papers by Keyword: Multiple Linear Regression

Abstract: An efficient grounding system design for safe fault current discharge requires the knowledge of soil resistivity. This study analysed the dependency of soil resistivity (ρ_s) on some soil factors such as soil texture (ST), moisture content (MC), temperature (T) and depth (D) for optimal grounding system operation. Two seasonal measurements of ρ_s were conducted during rainy season (July 2023) and dry season (January 2024) at six different sites in Federal University of Agriculture, Abeokuta (FUNAAB), Ogun State, Nigeria as a case study. Using Herojat Rhomega-smart resistivity meter, the ρ_s at the sites was measured via Wenner method. The ST and MC were determined via laboratory analysis of five 5 kg samples of soil from each site at the D values generated by the meter while T was measured using a thermometer. The ρ_s dependency on MC, T and D at each site was modelled using multiple linear regression (MLR). Coefficient of determination (R²) was used to determine MC, T and D contributions to ρ_s. The obtained results revealed that the measured ρ_s at the six sites over the study period was a function of ST, MC, T and D. The developed MLR models for the sites for both rainy and dry seasons showed that MC, T and D collectively influenced the ρ_s value better than the individual factors owing to higher value of multiple R² observed. The outcomes of this study could be adopted as good reference points for further soil resistivity analysis and grounding system installation for FUNAAB.

Abstract: Corrosion of steel structures in marine environments is a critical issue affecting infrastructure integrity and maintenance costs worldwide. Generally, inhibitors have proven to reduce the corrosion rate to the barest minimum than other methods. The inhibitors are produced using the experimental method which is time consuming and costly. This necessitate the development of models for the quick assessment of the efficiency of the inhibitor. This research focused on the prediction of corrosion inhibitory efficiency of water hyacinth on mild steel in marine environment using multiple linear regression (MLR) method. Various concentrations (5 ml, 10 ml, 15 ml, 20 ml and 25 ml) were added to the samples immersed in seawater and a sample without the addition of the inhibitor was used as the control for a period of 30 days. The study was carried out using weight loss method and the corrosion rate as well as the inhibition efficiency were calculated. Phytochemical analysis and atomic absorption spectroscopic were carried out on the inhibitor while Scanning Electron Microscopy and Energy Display X-ray Spectroscopy were used to analyze the steel sample. The analysis of the result showed that the best inhibition efficiency obtained was 90% and this was achieved with 15% concentration of the inhibitor. Multiple linear regression model was developed to predict the inhibitor’s efficiency. The predicted efficiency with the MLR model was compared with that of the experimentally obtained efficiency and the outcome shows a conformity between the experimental and the predicted value. It would therefore be recommended to rely on multiple linear regression in predicting the efficiency of water hyacinth for corrosion control of mild steel in marine environment based on the closeness of the predicted values to the experimental values.

Abstract: Traditional taxis have an important place in urban mobility because taxis provide flexible, comfortable and door-to-door service to passengers. However, for the continuity of traditional taxi services, profitability analyzes were needed despite the sharing economy. For this reason, the focus of this study is traditional taxis. This study aims to analyze the profitability of traditional taxi services in the urban arteries of Istanbul. For this purpose, a survey was conducted with 35 taxis and 70 taxi drivers. Then a model was then developed consisting of the independent variables number of trips (TRP), total trip distance (DST) and efficiency (EFF) that affect the profitability of taxi services. Additionally, contour plots were used to more accurately evaluate the effect of independent variables. As a result, it was concluded that the most important variable affecting the profitability of traditional taxi services is the efficiency (EFF) independent variable.

Abstract: Environmental contamination might result from sawdust waste that has not been adequately managed. However, waste has a high economic value. This study aimed to analyze the characteristic model of sawdust after the carbonization process. The research method used the L9(3)4 Orthogonal Array experiment. The research variables included: drying temperature (X₁), drying time (X₂), carbonization temperature (X₃), and carbonization time (X₄), each with three levels of factors. The research response variables were moisture content (Y₁), volatile matter (Y₂), ash (Y₃), and fixed carbon (Y₄) of sawdust charcoal. The results showed that the average moisture content was 0.9%, volatile matter 8.3%, ash content 8.29%, and fixed carbon content 82.51%. According to the outcomes of multiple linear regression analysis, the correlation coefficients (R) of the four were very significant for moisture content, volatile matter, ash, and fixed carbon of 0.865, 0.929, 0.987, and 0.935, respectively. The optimum conditions obtained were water content X_1-2X_2-3X_3-3X_4-3, volatile material X_1-1X_2-1X_3-1X_4-1, ash content X_1-1X_2-1X_3-1X_4-1, and carbon content X_1-2X_2-1X_3-1X_4-1. The outcomes of the sawdust charcoal proximate analysis model validation test were normally distributed, and there was no homoscedasticity, multicollinearity, or negative autocorrelation. Thus, the four models produced in this study were feasible and valid so that they could use them to predict the physical material characteristics of teak sawdust.

Abstract: The weight percentage of food waste, plastics and rubber, paper, textile, weed and wood, and leather were measured for dry-base municipal solid wastes in a city of west China respectively. The dry higher heating value, wet higher heating value, and wet lower heating value of municipal solid wastes were also measured respectively. Based on the measured physical compositions data of wastes, three models were developed to predict three kinds of heating values respectively by the multiple linear regression method. The prediction results were compared with three predictive models from different regions in the world, and the predictive results of the developed models are the most accurate.

Abstract: Waste generated in construction sites has recently increased and has become an uncontrollable cause of environmental problems and profit loss to contractors. The lack of real data or research on such wastes is due to the lack of suitable policies regarding this issue. The actions of contractors are not controlled by rules on this issue. This situation leads to the lack of action or awareness on the side of the contractor. Concrete waste is also part of the waste generated in construction sites. We determine the concrete waste generated in construction stages and conduct multiple linear regression analysis of the amount of column waste generated. The methodology employed in this study involves site observations, interviews with site personnel, and sampling at housing construction sites. The estimation method is utilized for the sampling of concrete waste. Results show that the average percentage of column waste is 13.93% and that of slab waste is 0.34%. These percentage values are derived from the total order of the concrete. The difference is due to the sizes of structures and method of handling. The regression model obtained from the sample data on column waste resulted in an adjusted R² value of 0.895. Therefore, the model predicts approximately 89.5% of the factors involved in concrete waste generation.

Abstract: In this paper, a new procedure for the identification of constitutive elastoplastic models coupled with an isotropic damage variable under large strains is presented. It is a statistical approach used for experimental characterization and identification based on a design of experiments of numerical simulations of mechanical characterization tests. The parameters for a reference material are determined by multiple linear regression as a function of shape indices.The material of reference is mild steel E24; it was characterized in a series of tensile tests of thin plate specimens. The Swift hardening law coupled with an isotropic damage variable that was identified by introducing in the established formulations shape indexes extracted from the experimental tension/elongation curves.The number of simulations required for the identification of the parameters of the reference material is roughly 18% of the number required by the inverse method (simplex).

Abstract: Considering the complexity and randomness of the karst groundwater systems, a multiple linear regression model was developed for groundwater-level prediction based on the R language. The Jinci Spring basin was taken as a case study. Results show that the established model can predict the dynamics of the karst groundwater levels with high accuracy at an annual time scale, which can be served for macroscopic groundwater management.

Abstract: Air quality has been deteriorated seriously in Shanghai as a result of urbanization and modernization. Air pollution concentrations were decreased during the period of 2008-2011. PM₁₀, SO₂ and NO₂ concentrations were higher in winter than in summer. Meteorological conditions affect air pollution levels in the urban atmosphere significantly due to their important role in transport and dilution of the pollutants. Multiple linear regression models were used to predict next day’s PM₁₀, SO₂ and NO₂ concentrations, respectively. The calculated R² values were 0.753, 0.800 and 0.861 for PM₁₀, SO₂ and NO₂ regression model, respectively. This result shows the multiple regression model analysis, providing simple and faster application facilities, is useful for modeling the impacts of meteorological factors on air pollutant levels.

Abstract: The main objective of time series analysis is to develop models that can establish the relation between variables,the paper ,An improved method of the RBF is proposed,which is a five-layered network structure comprising of an input layer,wavelet layer, product layer, output layer and polynomial regressive weight layer.which uses an online optimization approach, the method uses an offline learning method known as SNPOM,the polynomial weights are updated many times during the process of looking for the search direction to update the nonlinear parameters. The experiment showed that the optimization technique can speed up the convergence rate of nonlinear model during the learning process.