Search published articles


Showing 8 results for Regression Model

Khaki A.m., Moayedfar R.,
Volume 2, Issue 2 (6-2004)
Abstract

The purpose of the present study, is proposing a more flexible model comparing with linear regression model, to estimate the rate of household trip production and its prediction in the�project horizon. For this purpose, a combined model composed of poission distribution and the possible distribution of A. in the form of negative binominal distribution are used. Then the proposed model was conducted on a real case (Karaj City). Then the result of model processinghas been compared to .the real observation in the peak hours in Karaj city.
S.n. Moghaddas Tafreshi, Gh. Tavakoli Mehrjardi, S.m. Moghaddas Tafreshi,
Volume 5, Issue 2 (6-2007)
Abstract

The safety of buried pipes under repeated load has been a challenging task in geotechnical engineering. In this paper artificial neural network and regression model for predicting the vertical deformation of high-density polyethylene (HDPE), small diameter flexible pipes buried in reinforced trenches, which were subjected to repeated loadings to simulate the heavy vehicle loads, are proposed. The experimental data from tests show that the vertical diametric strain (VDS) of pipe embedded in reinforced sand depends on relative density of sand, number of reinforced layers and height of embedment depth of pipe significantly. Therefore in this investigation, the value of VDS is related to above pointed parameters. A database of 72 experiments from laboratory tests were utilized to train, validate and test the developed neural network and regression model. The results show that the predicted of the vertical diametric strain (VDS) using the trained neural network and regression model are in good agreement with the experimental results but the predictions obtained from the neural network are better than regression model as the maximum percentage of error for training data is less than 1.56% and 27.4%, for neural network and regression model, respectively. Also the additional set of 24 data was used for validation of the model as 90% of predicted results have less than 7% and 21.5% error for neural network and regression model, respectively. A parametric study has been conducted using the trained neural network to study the important parameters on the vertical diametric strain.
H. Behbahani, S.a. Sahaf,
Volume 5, Issue 3 (9-2007)
Abstract

The available methods for predicting mechanical characteristics of pavement layers are categorized into two general groups, Destructive and Non-destructive. In destructive method, using coring and pavement subgrade and performing necessary experiments on them, the quantities of layers properties will be identified. In Non-destructive method, the attained deflection is measured by applying the loading on pavement surface using equipments such as FWD which charges the impact dynamic load, and the mechanical characteristics of pavement layers are determined using back calculations. The procedure of conducting these calculations is that by knowing the thickness of the pavement layers and assuming the initial amounts for mechanical characteristics of the layer, the attained deflection at the desired points on the pavement surface will be calculated. Then, new figures are assumed for the characteristics of layers in a reattempt and calculations are repeated again. This trial and error is continued until the produced basin deformations from the calculations with true value, differs in an acceptable range. Using this method may have no accurate and single answer, since the various compositions of layers characteristics can produce similar deformations in different points of pavement surface. In this article, using an innovative method, a measurement is taken in constructing and introducing a mathematical model for determining the elastic module of surface layer using deflections attained from FWD loading equipment. The procedure is such that by using dynamic analysis software of finite elements like ABAQUS and ANSYS, the deformation of corresponding points on the surface of the pavement will be attained by FWD loading equipment. This analysis will be performed on a number of pavements with different thicknesses and different layers properties. The susceptibility analysis of different points deformations show, which will be performed as a result of the change of properties and layers thicknesses. Using this artificial data base as well as deflection basin parameters (DBP), a measurement will be taken toward constructing a regression model for determination of asphalt layer model, i.e. Eac =f(DBP) function shall be attained. To achieve the maximum correlation coefficient, an attempt is made to use the parameters of deformations basin which has the most susceptibility in changing asphalt layer module.
K. J. Tu, Y. W. Huang,
Volume 11, Issue 4 (12-2013)
Abstract

The decisions made in the planning phase of a building project greatly affect its future operation and maintenance (O&M) cost. Recognizing the O&M cost of condominiums’ common facilities as a critical issue for home owners, this research aims to develop an artificial neural network (ANN) O&M cost prediction model to assist developers and architects in effectively assessing the impacts of their decisions made in the planning phase of condominium projects on future O&M costs. A regression cost prediction model was also developed as a benchmark model for testing the predictive accuracy of the ANN model. Six critical building design attributes (building age, number of apartment units, number of floors, average sale price, total floor area, and common facility floor area) which are usually available in the project planning phase, were identified as the input factors to both models and average monthly O&M cost as the output factor. 55 of the 65 existing condominium properties randomly selected were treated as the training samples whose data were used to develop the ANN and regression models the other ten as the test samples to compare and verify the predictive performance of both models. The study results revealed that the ANN model delivers more accurate and reliable cost prediction results, with lower average absolute error around 7.2% and maximum absolute error around 16.7%, as compared with the regression model. This study shows that ANN is an effective method in predicting building O&M costs in the project planning phase. Keywords: Project management, Facility management, Common facilities, Cost modeling.
H. Shahnazari, M. A. Shahin, M. A. Tutunchian,
Volume 12, Issue 1 (1-2014)
Abstract

Due to the heterogeneous nature of granular soils and the involvement of many effective parameters in the geotechnical behavior of soil-foundation systems, the accurate prediction of shallow foundation settlements on cohesionless soils is a complex engineering problem. In this study, three new evolutionary-based techniques, including evolutionary polynomial regression (EPR), classical genetic programming (GP), and gene expression programming (GEP), are utilized to obtain more accurate predictive settlement models. The models are developed using a large databank of standard penetration test (SPT)-based case histories. The values obtained from the new models are compared with those of the most precise models that have been previously proposed by researchers. The results show that the new EPR and GP-based models are able to predict the foundation settlement on cohesionless soils under the described conditions with R2 values higher than 87%. The artificial neural networks (ANNs) and genetic programming (GP)-based models obtained from the literature, have R2 values of about 85% and 83%, respectively which are higher than 80% for the GEP-based model. A subsequent comprehensive parametric study is further carried out to evaluate the sensitivity of the foundation settlement to the effective input parameters. The comparison results prove that the new EPR and GP-based models are the most accurate models. In this study, the feasibility of the EPR, GP and GEP approaches in finding solutions for highly nonlinear problems such as settlement of shallow foundations on granular soils is also clearly illustrated. The developed models are quite simple and straightforward and can be used reliably for routine design practice.
Farzin Kalantary, Javad Sadoghi Yazdi, Hossein Bazazzadeh,
Volume 12, Issue 3 (7-2014)
Abstract

In comparison with other geomaterials, constitutive modeling of rockfill materials and its validation is more complicated. This is principally due to the existence of more intricate phenomena such as particle crushing, as well as laboratory test limitations. These issues have necessitated developing more complex constitutive models, with many parameters. Regardless of the type of model, the calibrations of the parameters in such models are considered as one of the most important and challenging steps in the application of the model. Therefore, the need for comprehensive and rapid methods for evaluation of optimum parameters of the models is deemed necessary. In this paper, a Neuro-Fuzzy model in conjunction with Particle Swarm Optimization (PSO) is used for calibration of the twelve parameters of Hierarchical Single Surface (HISS) constitutive model based on the Disturbed State Concept (DSC). The Neuro-fuzzy system is used to provide a high-degree nonlinear regression model between the deviatoric stress and volumetric strain versus axial strain that has been obtained from consolidated drained large scale tri-axial tests on rockfill materials. The model parameters are determined in an iterative optimized loop with PSO and ANFIS such that the equations of DSC/HISS are simultaneously satisfied. Material data used in this study are gathered from the results of large tri-axial tests for two rockfill dams in Iran. It is shown that the proposed method has higher accuracy and more importantly its robustness is exhibited through test predictions. The achieved improvement is substantiated in a comparison with the more widely used "Least-Square" method.
Mohammad Reza Saberi, Alireza Rahai, Masoud Sanayei ,
Volume 15, Issue 1 (1-2017)
Abstract

Steel bridges play a very important role in every country’s transportation system. To ensure that bridges perform reliably, engineers monitor their performance which is referred to as Structural Health Monitoring (SHM). An important element of SHM includes the prediction of service life. There is ample historical evidence that bridge damage is pervasive and their life time is decreasing. To manage costs and safety, service life prediction of bridges is necessary. We present a statistical method to predict service life for steel bridges. A nonparametric statistical model based on the bootstrap method for stress analysis for fatigue life prediction of steel girder bridges is proposed. The bootstrap provides a simple approach for reproduction of the probability distribution of measured strain data. The bootstrap is sensitive to the number of events in the verification sample (data), thus we introduce a stable survival distribution function (SDF). An index is presented in this paper for inferring the service life of steel bridges, which can be known as the Life Index (µ). The life index function shows variation of the age of steel bridges under daily traffic loads. A regression model is developed which relates the service life of steel bridges using a bridge life index based on measured operational strain time histories. The predicted remaining service life derived from the model can contribute to effective management of steel bridges. The proposed method assists bridge engineers, bridge owners, and state officials in objective assessment of deteriorated bridges for retrofit or replacement of deteriorated bridges. Timely repair and retrofit increase the safety levels in bridges and decrease costs.


Ahmad Soltanzadeh, Iraj Mohammadfam, Abbas Moghim Beygi, Reza Ghiasvand,
Volume 15, Issue 7 (10-2017)
Abstract

Construction industries are the most dangerous worksites with high risk of occupational accident and bodily injuries, which ranges from mild to very severe cases. The aim of this study was to explore the causal factors of accident severity rate (ASR), in 13 of the biggest Iranian construction industries. In this analytical cross-sectional study, the data of registered accidents from 2009 until 2013 were obtained from an official database. Data of HSE risk management systems and HSE training were also gathered from comprehensive accident investigation reports. Data analysis and regression modeling were done using SPSS statistical software (version 22). The mean and SD of ASR of studied construction worksites was 257.52±1098.95. The results show that the system associated with HSE and HSE risk management established only 41.8 and 18.4%, respectively. The results of multiple linear regression indicated that some individual and organizational factors (IOFs), HSE training factors (HTFs), and Risk Management System factors (RMSFs) were significantly associated with ASR (p<0.05). The study revealed the causal factors of ASR. Hence, these findings can be applied in the design and implementation of a comprehensive HSE risk management system to reduce ASR.



Page 1 from 1     

© 2019 All Rights Reserved | International Journal of Civil Engineering

Designed & Developed by : Yektaweb