Principle of Demographic Gravitation to Estimate Annual Average Daily Traffic: Comparison of Statistical and Neural Network Models

Duddu, Venkata R.; Pulugurtha, Srinivas S.

doi:10.1061/(asce)te.1943-5436.0000537

Cited by 29 publications

(13 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several transportation modeling methods exist for estimating traffic patterns: generalized linear modeling (Seaver et al. ; Zhao and Chung ), neural network analysis (Duddu and Pulugurtha ), empirical Bayes estimation (Yang and Davis ), universal kriging (Eom et al. ; Selby and Kockelman ), and support vector machines (Castro‐Neto et al.…”

Section: Methodsmentioning

confidence: 99%

“…Information on species occurrence, traffic volume, and traffic speed is unlikely to be available for every place i and we propose that this can be modeled. Several transportation modeling methods exist for estimating traffic patterns: generalized linear modeling (Seaver et al 2000;Zhao and Chung 2001), neural network analysis (Duddu and Pulugurtha 2013), empirical Bayes estimation (Yang and Davis 2002), universal kriging (Eom et al 2006;Selby and Kockelman 2013), and support vector machines (Castro-Neto et al 2009). Modeling approaches to estimate species occurrence include correlative (Guisan and Thuiller 2005;Elith and Leathwick 2009) and mechanistic (Kearney and Porter 2009) species distribution modeling (SDM) and extensions of population viability analysis (PVA) (Gilpin and Soule 1986;Shaffer 1990).…”

Section: Conceptual Model Frameworkmentioning

confidence: 99%

See 1 more Smart Citation

A simple framework for a complex problem? Predicting wildlife–vehicle collisions

Visintin

Ree

McCarthy

2016

Ecology and Evolution

View full text Add to dashboard Cite

SummaryCollisions of vehicles with wildlife kill and injure animals and are also a risk to vehicle occupants, but preventing these collisions is challenging. Surveys to identify problem areas are expensive and logistically difficult. Computer modeling has identified correlates of collisions, yet these can be difficult for managers to interpret in a way that will help them reduce collision risk. We introduce a novel method to predict collision risk by modeling hazard (presence and movement of vehicles) and exposure (animal presence) across geographic space. To estimate the hazard, we predict relative traffic volume and speed along road segments across southeastern Australia using regression models based on human demographic variables. We model exposure by predicting suitable habitat for our case study species (Eastern Grey Kangaroo Macropus giganteus) based on existing fauna survey records and geographic and climatic variables. Records of reported kangaroo–vehicle collisions are used to investigate how these factors collectively contribute to collision risk. The species occurrence (exposure) model generated plausible predictions across the study area, reducing the null deviance by 30.4%. The vehicle (hazard) models explained 54.7% variance in the traffic volume data and 58.7% in the traffic speed data. Using these as predictors of collision risk explained 23.7% of the deviance in incidence of collisions. Discrimination ability of the model was good when predicting to an independent dataset. The research demonstrates that collision risks can be modeled across geographic space with a conceptual analytical framework using existing sources of data, reducing the need for expensive or time‐consuming field data collection. The framework is novel because it disentangles natural and anthropogenic effects on the likelihood of wildlife–vehicle collisions by representing hazard and exposure with separate, tunable submodels.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Conceptual Model Frameworkmentioning

confidence: 99%

A simple framework for a complex problem? Predicting wildlife–vehicle collisions

Visintin

Ree

McCarthy

2016

Ecology and Evolution

View full text Add to dashboard Cite

show abstract

“…Spatial interpolation algorithms like Kriging have been introduced to account for spatial autocorrelations ( 11 , 13 ). Other modeling techniques, such as generalized equation models and demographic gravitation methods have been employed by including additive parameters to capture spatial autocorrelations ( 14 , 15 ).…”

Section: Literature Reviewmentioning

confidence: 99%

Estimating Traffic Volume for Local Streets with Imbalanced Data

Chen

Shen

et al. 2019

Transportation Research Record

View full text Add to dashboard Cite

Annual average daily traffic (AADT) is an important measurement used in traffic engineering. Local streets are major components of a road network. However, automatic traffic recorders (ATRs) used to collect AADT are often limited to arterial roads, and such information is, therefore, often unavailable for local streets. Estimating AADT on local streets becomes a necessity as local street traffic continues to grow and the capacity of arterial roads becomes insufficient. A challenge is that an under-represented sample of local street AADT may result in biased estimation. A synthetic minority oversampling technique (SMOTE) is applied to oversample local streets to correct the imbalanced sampling among different road types. A generalized linear mixed model (GLMM) is employed to estimate AADT incorporating various independent variables, including factors of roadway design, socio-demographics, and land use. The model is examined with an AADT dataset from Seattle, WA. Results show that: (1) SMOTE helps to correct imbalanced sampling proportions and improve model performance significantly; (2) the number of lanes and the number of crosswalks are both positively associated with AADT; (3) road segments located in areas with a higher population density or more mixed land use have a higher AADT; (4) distance to the nearest arterial road is negatively correlated with AADT; and (5) AADT creates spatial spillover effects on neighboring road segments. The combination of SMOTE and GLMM improves the estimation accuracy on AADT, which contributes to better data for transportation planning and traffic monitoring, and to cost saving on data collection.

show abstract

“…The research has been completed using traffic data for the period 1994-1998 in Hong Kong [15]. Duddu and Pulugurtha (2013) have developed a model using statistical methods and ANN taking into account demographic principles in order to estimate link-level AADT based on characteristics of the land use, in the city of Charlotte, North Carolina [16].…”

Section: Introductionmentioning

confidence: 99%

Development of Traffic Volume Forecasting Using Multiple Regression Analysis and Artificial Neural Network

Дураку

Ramadani

2019

Civ Eng J

View full text Add to dashboard Cite

The purpose of this study is to develop a model for traffic volume forecasting of the road network in Anamorava Region. The description of the current traffic volumes is enabled using PTV Visum software, which is used as an input data gained through manual and automatic counting of vehicles and interviewing traffic participants. In order to develop the forecasting model, there has been the necessity to establish a data set relying on time series which enables interface between demographic, socio-economic variables and traffic volumes. At the beginning models have been developed by MLR and ANN methods using original data on variables. In order to eliminate high correlation between variables appeared by individual models, PCA method, which transforms variables to principal components (PCs), has been employed. These PCs are used as input in order to develop combined models PCA-MLR and PCA-RBF in which the minimization of errors in traffic volumes forecasting is significantly confirmed. The obtained results are compared to performance indicators such R2, MAE, MSE and MAPE and the outcome of this undertaking is that the model PCA-RBF provides minor errors in forecasting.

show abstract

Principle of Demographic Gravitation to Estimate Annual Average Daily Traffic: Comparison of Statistical and Neural Network Models

Cited by 29 publications

References 29 publications

A simple framework for a complex problem? Predicting wildlife–vehicle collisions

A simple framework for a complex problem? Predicting wildlife–vehicle collisions

Estimating Traffic Volume for Local Streets with Imbalanced Data

Development of Traffic Volume Forecasting Using Multiple Regression Analysis and Artificial Neural Network

Contact Info

Product

Resources

About