Annual Average Daily Traffic Estimation Research Articles

Integrating traffic pollution dispersion into spatiotemporal NO2 prediction

Accurately predicting ambient NO2 concentrations has great public health importance, as traffic-related air pollution is of major concern in urban areas. In this study, we present a novel approach incorporating traffic contribution to NO2 prediction in a fine-scale spatiotemporal model. We used nationally available traffic estimate dataset in a scalable dispersion model, Research LINE source dispersion model (RLINE). RLINE estimates then served as an additional input for a validated spatiotemporal pollution modeling approach. Our analysis uses measurement data collected by the Multi-Ethnic Study of Atherosclerosis and Air Pollution in the greater Los Angeles area between 2006 and 2009. We predicted road-type-specific annual average daily traffic (AADT) on road segments via national-level spatial regression models with nearest-neighbor Gaussian processes (spNNGP); the spNNGP models were trained based on over half a million point-level traffic volume measurements nationwide. AADT estimates on all highways were combined with meteorological data in RLINE models. We evaluated two strategies to integrate RLINE estimates into spatiotemporal NO2 models: 1) incorporating RLINE estimates as a space-only covariate and, 2) as a spatiotemporal covariate. The results showed that integrating the RLINE estimates as a space-only covariate improved overall cross-validation R2 from 0.83 to 0.84, and root mean squared error (RMSE) from 3.58 to 3.48 ppb. Incorporating the estimates as a spatiotemporal covariate resulted in similar model improvement. The improvement of our spatiotemporal model was more profound in roadside monitors alongside highways, with R2 increasing from 0.56 to 0.66 and RMSE decreasing from 3.52 to 3.11 ppb. The observed improvement indicates that the RLINE estimates enhanced the model's predictive capabilities for roadside NO2 concentration gradients even after considering a comprehensive list of geographic covariates including the distance to roads. Our proposed modeling framework can be generalized to improve high-resolution prediction of NO2 exposure – especially near major roads in the U.S.

One-Dimensional Convolutional Neural Network Model for Local Road Annual Average Daily Traffic Estimation

One-Dimensional Convolutional Neural Network Model for Local Road Annual Average Daily Traffic Estimation

VEHICLE DENSITY ESTIMATION IN QUEZON CITY USING OBJECT-BASED FEATURE EXTRACTION ON SATELLITE IMAGES

Abstract. Manual vehicle counting is often tedious, expensive, and time-consuming. While automatic counting from CCTV allows for annual average daily traffic estimation, CCTV files in the Philippines are not available to the public and do not fully cover all road extents. In this study, Remote Sensing and Geographic Information Systems (GIS) techniques are employed to use readily available satellite images to obtain vehicle count in selected road segments in the Central Business Districts of Quezon City before and after the COVID-19 lockdown. Using the existing Google Earth Images, a segmentation algorithm using ENVI Feature Classification was developed to allow remote counting of vehicles from the earliest image in 2018. The devised algorithm was able to delineate, identify, and classify according to the types of vehicles that are visible on the image. An average error rate of 12.24% was found by comparison of automated counts and manual counts on the images, while a regression analysis yielded a value of R2 = 0.9227 that denoted a strong relationship between automated and manual counts. Vehicle density was calculated, and percent differences were obtained to determine the relative differences of the vehicle counts from the vehicle count of the earliest image taken in 2018. It was found that the vehicle density declined by at least 81% by March 25, 2020. The methodological framework presented in this study provides estimates of vehicle counts and vehicle density. It can be further improved if vehicle counts, on the same location and period, from field validation surveys are available.

Improving Stratification Procedures and Accuracy of Annual Average Daily Traffic (AADT) Estimates for Non-Federal Aid-System (NFAS) Roads

The 2016 safety Final Rule requires states to have access to annual average daily traffic (AADT) for all public paved roads, including non-Federal aid-system (NFAS) roads. The latter account approximately for 75% of the total roadway mileage in the country making it difficult for agencies to collect traffic data on these roads. Many agencies use stratified sampling procedures to develop default AADT estimates for uncounted segments; however, there is limited guidance and information about how to stratify the network effectively. The goal of this paper is to enhance transportation agencies’ ability to improve existing stratification schemes, design new schemes, and ultimately develop more accurate AADT estimates for NFAS roads. The paper presents the results from five pilot studies that validated and compared the performance of current, updated, and new (traditional and decision-tree-based) schemes using readily available data. According to the results, the median absolute percent error of existing AADT estimates, developed by state agencies, ranged between 71% and 120%. Updating these schemes using recent counts resulted in an AADT accuracy improvement of 25%. The best-performing schemes were developed using DTs that improved the AADT accuracy of existing schemes by 41%. Overall, having more strata and very homogenous strata is better than having fewer strata and more samples within each stratum. The analysis revealed that a key to selecting an effective scheme is to determine a critical point, beyond which creating more strata improves the AADT accuracy marginally but increases the required sample size exponentially.

Modeling AADT on local functionally classified roads using land use, road density, and nearest nonlocal road data

The focus of this research is to model the influence of road, socioeconomic, and land-use characteristics on local road annual average daily traffic (AADT) and assess the model's predictability in non-covered location AADT estimation. Traditional ordinary least square (OLS) regression and geographically weighted regression (GWR) methods were explored to estimate AADT on local roads. Ten spatially distributed counties were considered for county-level analysis and modeling. The results indicate that road density, AADT at the nearest nonlocal road, and land use variables have a significant influence on local road AADT. The GWR model is found to be better at estimating the AADT than the OLS regression model. The developed county-level models were used for estimating AADT at non-covered locations in each county. The methodology, findings, and the AADT estimates at non-covered locations can be used to plan, design, build, and maintain the local roads in addition to meeting reporting requirements. The prediction error is found to be higher at urban areas and in counties with a smaller number of local road traffic count stations. Recommendations are made to account for influencing factors and enhance the local road count-based AADT sampling methodology.

Data-Driven Approach to Quantify and Reduce Error Associated with Assigning Short Duration Counts to Traffic Pattern Groups

Traffic monitoring agencies collect traffic data samples to estimate annual average daily traffic (AADT) at short duration count sites. The steps to estimate AADT from sample data introduce error that manifests as uncertainty in the AADT statistic and its applications. Past research suggests that the assignment of a short duration count site to a traffic pattern group (TPG), characterized by known traffic periodicities, represents a significant but poorly quantified source of error. This paper presents an approach to quantify the range of errors arising from such assignments and to mitigate these errors using a novel data-driven assignment method. The approach uses simulated 48-hour short duration counts sampled from continuous count sites with known AADT to develop a benchmark of the total error expected when AADT is estimated from such samples. Likewise, the analysis produces a set of AADT estimates using temporal factors from pre-defined TPGs to quantify the range of assignment errors. The data-driven assignment method aims to mitigate these errors by minimizing the absolute mean deviation in AADT estimates produced from multiple short duration counts in a single year. The approach is applied to traffic data collected in Manitoba, Canada, as a case study. The results indicate that the mean absolute error from 48-hour short duration counts is 6.40% of the true AADT and that improper assignment can lead to a range in mean absolute errors of 9%. When applied to previously unassigned sites, the data-driven assignment method reduced mean absolute errors from 10.32%, using a conventional assignment method, to 7.86%.

EXPLORATION OF MUNICIPAL MOBILITY USING SMARTPHONE GPS DATA

Abstract. Due to the advances in location-acquisition techniques, smartphone GPS location data has emerged with opportunity for research, development, innovation, and business. A variety of research has been developed to study human behaviour through exploring patterns from these data. In this paper, we use smartphone GPS location data to investigate municipal mobility. Kernel density estimation and emerging hot spot analysis are used to the GPS dataset to demonstrate GPS user (point) distribution across space and time. Flow maps are capable of tracking clustering behaviours and direction maps drawn upon the orientation of vectors can precisely identify location of the events. Case study with Indianapolis metropolitan area traffic rush hour verifies the effectiveness of these methods. Furthermore, we identify smartphone GPS data of vehicles and develop a concise and effective method for traffic volume estimation for county highway network. It is shown that the developed smartphone GPS data analytics is powerful for predicting reliable annual average daily traffic estimation. The study showcases the capability of GPS location data in identifying municipal mobility patterns for both citizens and vehicles.

Enhancing Statewide Annual Average Daily Traffic Estimation with Ubiquitous Probe Vehicle Data

Annual average daily traffic (AADT) is a critical input into many transportation applications, particularly safety reporting. For example, the Highway Safety Improvement Program in the U.S. requires states to make AADT data for all public paved roadways accessible by 2026. Because collecting traffic counts on every network segment is prohibitively expensive, a method capable of accurately estimating AADT on unmonitored segments is of great value to state DOTs. The ubiquitous probe vehicle data present a great opportunity to this end. This paper presents an enhanced method for statewide AADT estimation by leveraging such data in Kentucky. The use of the probe data is explored in two ways. First, an annual average daily probes (AADP) variable is derived from hourly probe counts; second, a betweenness centrality (BC) variable is calculated using probe speeds. Including both variables and using the random forest model results in model performance that exceeds those previously reported for statewide applications. Incorporating AADP and BC improves the accuracy of AADT estimates by 30%–37% for all roads and 23%–43% for highways in functional classes 5–7, compared with only using sociodemographic and roadway characteristics. These results demonstrate the value of the probe data for enhancing AADT estimation. The analysis further shows that on roadways having more than 53 AADP or an average of 2.2 probe counts per hour, the median and the mean absolute percent errors are below 20% and 25%, respectively. These findings have practical implications for state DOTs wanting to maximize the utility of probe vehicle data.

Annual average daily traffic estimation in England and Wales: An application of clustering and regression modelling

Collection of Annual Average Daily Traffic (AADT) is of major importance for a number of applications in road transport urban and environmental studies. However, traffic measurements are undertaken only for a part of the road network with minor roads usually excluded. This paper suggests a methodology to estimate AADT in England and Wales applicable across the full road network, so that traffic for both major and minor roads can be approximated. This is achieved by consolidating clustering and regression modelling and using a comprehensive set of variables related to roadway, socioeconomic and land use characteristics. The methodological output reveals traffic patterns across urban and rural areas as well as produces accurate results for all road classes. Support Vector Regression (SVR) and Random Forest (RF) are found to outperform the traditional Linear Regression, although the findings suggest that data clustering is key for significant reduction in prediction errors.

Interpretable machine learning approach in estimating traffic volume on low-volume roadways

Abstract Many state and local agencies are currently facing challenges concerning the collection and estimation of traffic volumes, particularly regarding the collection of annual average daily traffic (AADT) on low-volume roads. To overcome these challenges, there is a need to develop new affordable methods to collect data and estimate traffic volume on low-volume roadways. In this study, the research team developed an innovative interpretable machine learning framework and applied it to low-volume roads in Vermont to estimate traffic volumes. This study used several databases (e.g., U.S. Census, the American community survey) to prepare the final dataset for the model development. The findings show that population density and work area characteristic (WAC) density are the best predictors in estimating AADT. The model outcomes show that the machine learning models yield better estimates than the conventional parametric statistical methods. By improving the accuracy of AADT estimations, this study contributed to traffic monitoring and safety improvement, and it can help reduce costs of data collection. This study developed the top five decision rules for three types of low-volume roadways. Stakeholders can use the findings of this study to meet the new requirements pertaining to availability of AADT estimates for low-volume roads. Additionally, the best fit estimates and the developed rules from the current study could enhance the predictive power of the SPF development for the low-volume roadways in Vermont and therefore improve the decision-making process.

Updating Annual Average Daily Traffic Estimates at Highway-Rail Grade Crossings with Geographically Weighted Poisson Regression

Highway-rail grade crossings (HRGCs) are unique nodes in the transportation system that facilitate the movement of rail and highway traffic. Various mathematical models are available that provide safety assessments of HRGCs. A chief ingredient of these models is the annual average daily traffic (AADT). One of the main sources of data for such models is the Federal Railroad Administration (FRA)’s Grade Crossing Inventory. A substantial portion of the AADT data in the inventory is outdated. This paper investigates the effects of using out-of-date rather than up-to-date AADT values, using two safety assessment models to isolate the differences. Results show that the use of out-of-date AADT data generates biased rankings of HRGCs based on safety considerations. Since collection of AADT data is resource-intense, a methodology based on a geographic information system for estimating updated AADT is presented. This methodology utilizes limited traffic counts that are supplemented with additional publicly available data. An application using a geographically weighted Poisson regression model for 14 HRGCs gave results that closely matched AADT values based on 2018 field traffic counts at those HRGCs. This method provides an alternative to costly field-data-based updating of AADT in the relatively extensive Grade Crossing Inventory database. Limitations of the research and suggestions for future research complete this paper.

Estimating Traffic Volume on Minor Roads at Rural Stop-Controlled Intersections using Deep Learning

Safety Performance Functions (SPFs) are regression models used to predict the expected number of collisions as a function of various traffic and geometric characteristics. One of the integral components in developing SPFs is the availability of accurate exposure factors, that is, annual average daily traffic (AADT). However, AADTs are not often available for minor roads at rural intersections. This study aims to develop a robust AADT estimation model using a deep neural network. A total of 1,350 rural four-legged, stop-controlled intersections from the Province of Alberta, Canada, were used to train the neural network. The results of the deep neural network model were compared with the traditional estimation method, which uses linear regression. The results indicated that the deep neural network model improved the estimation of minor roads’ AADT by 35% when compared with the traditional method. Furthermore, SPFs developed using linear regression resulted in models with statistically insignificant AADTs on minor roads. Conversely, the SPF developed using the neural network provided a better fit to the data with both AADTs on minor and major roads being statistically significant variables. The findings indicated that the proposed model could enhance the predictive power of the SPF and therefore improve the decision-making process since SPFs are used in all parts of the safety management process.

Spatial Interpolation of Missing Annual Average Daily Traffic Data Using Copula-Based Model

Accurate estimation of traffic counts [(i.e., annual average daily traffic (AADT)] is essential to transportation agencies for traffic demand forecasting, emission evaluation, pavement design, and project prioritization. Traditional AADT estimation methods rely on either temporal data imputation techniques based on historical records or kriging-based spatial interpolation approaches. However, Kriging method utilizes the correlation function as the sole descriptor of spatial dependency, posing limitations to yield accurate interpolation results for unstable AADTs under complex traffic patterns due to diverse road functions or land uses. This study proposed a copula-based model that combines spatial dependency and marginal distribution for missing AADT interpolation to weaken the limitation of Kriging method. Thus, the proposed model not only can describe the spatial dependency but also is robust to outliers. AADT data collected from the California state highway network were used to evaluate the effectiveness of spatial copula models with varying missing data rates. Four road segments with regular and recreational traffic patterns were selected to compare with existing kriging-based approaches. Results suggested that the spatial copulas yielded significantly higher accuracy rates than kriging did for irregular travel patterns with high missing data rates. Spatial copula models hold a great potential to improve the performance of large-scale transportation network-wide data imputation for planning and operational usages.

Development of Statewide Annual Average Daily Traffic Estimation Model from Short-Term Counts: A Comparative Study for South Carolina

Annual Average Daily Traffic (AADT) is an important parameter for traffic engineering analysis. Departments of Transportation continually collect traffic count using both permanent count stations (i.e., Automatic Traffic Recorders or ATRs) and temporary short-term count stations. In South Carolina, 87% of the ATRs are located on interstates and arterial highways. For most secondary highways (i.e., collectors and local roads), AADT is estimated based on short-term counts. This paper develops AADT estimation models for different roadway functional classes with two machine learning techniques: Support Vector Regression (SVR) and Artificial Neural Network (ANN). The models predict AADT from short-term counts. The results are first compared against each other, using the 2011 ATR data, to identify the best models. Then, the results of the best models are compared against both the regression-based model and factor-based model. The comparison reveals the superiority of the SVR model for AADT estimation for different roadway functional classes over all other methods. Among models for different roadway functional classes, developed with the 2011 ATR data, the SVR-based models show minimum errors in estimating AADT compared to the ANN-based, regression-based, and factor-based models, depicting the superiority of the SVR-based model for all roadway functional classes over other models in terms of AADT estimation accuracy. SVR models are validated for each roadway functional class using the 2016 ATR data and short-term count data collected by the South Carolina Department of Transportation (SCDOT). The validation results show that the SVR-based AADT estimation models can be used by the SCDOT as a reliable option to predict AADT from the short-term counts.

AIDS TO NAVIGATION SYSTEMS ON INLAND WATERWAYS AS AN ELEMENT OF COMPETITIVENESS IN ULCV TRAFFIC

Estimation of Annual Average Daily Traffic (AADT) is vital for departments of transportation (DOTs) work because AADT provides the basic information for planning new road construction, determination of roadway geometry, traffic control needs, congestion management strategies and safety considerations. AADT is used to determine state wide vehicle miles travelled on all roads and are used by transportation agencies to determine compliance with federal and state rules and regulations. DOTs spend heavily to collect traffic counts on state roads, but mostly traffic counts are not available for off-system, or low volume roads. Often estimates rely on a comparison with roads that are subjectively considered to be similar. Such comparisons are inherently subject to large errors, and also may not be repeated often enough to remain current. Therefore, a better method is needed for estimating AADT for off-system roads. This research developed a technique to estimate AADT for local roads in Alabama incorporating various facets from previous studies. A model has been developed using linear regression using known AADTs and collection of socio-economic and location variables as a means to estimate the AADT. The model relied upon five independent variables: nearby population, number of households in the area, employment in the area, population to job ratio and access to major roads. The model was used to generate AADT estimates on low-volume rural, local roads for 12 counties in Alabama. The model was developed using 70 percent of the collected data and validate to the remaining 30 percent of the data. Consistent with the recent literature on AADT estimation, a log transformation was attempted to determine if any improvements were determined. The paper concludes that a straight linear regression model can be used to predict the AADT for low-volumes roadways in Alabama for future applications.

Estimating annual average daily traffic and transport emissions for a national road network: A bottom-up methodology for both nationally-aggregated and spatially-disaggregated results

The regular and robust collection of traffic data for the entire road network in a given country will usually require high-cost investment in traffic surveys and automated traffic counters. This paper provides an alternative and low-cost approach for estimating annual average daily traffic values (AADTs) and the associated transport emissions for all road segments in a country. This is achieved by parsing and processing commonly available information from existing geographical data, census data, traffic data and vehicle fleet data. Ceteris paribus, we find that our annual average daily traffic estimation based on a neural network performs better than traditional regression models, and that the outcomes of our aggregated bottom-up road segment emission estimations are close to the outcomes from top-down models based on total energy consumption in transport. The developed approach can serve as a means of reliably estimating and verifying national road transport emissions, as well as offering a robust means of spatially analysing road transport activity and emissions, so as to support spatial emission inventory compilations, compliance with international environmental agreements, transport simulation modelling and transport planning.

Road safety performance measures and AADT uncertainty from short-term counts

ObjectiveThe objective of this paper is to enable better risk analysis of road safety performance measures by creating the first knowledge base on uncertainty surrounding annual average daily traffic (AADT) estimates when the estimates are derived by expanding short-term counts with the individual permanent counter method. BackgroundMany road safety performance measures and performance models use AADT as an input. While there is an awareness that the input suffers from uncertainty, the uncertainty is not well known or accounted for. MethodThe paper samples data from a set of 69 permanent automatic traffic recorders in Manitoba, Canada, to simulate almost 2 million short-term counts over a five year period. These short-term counts are expanded to AADT estimates by transferring temporal information from a directly linked nearby permanent count control station, and the resulting AADT values are compared to a known reference AADT to compute errors. The impacts of five factors on AADT error are considered: length of short-term count, number of short-term counts, use of weekday versus weekend counts, distance from a count to its expansion control station, and the AADT at the count site. ResultsThe mean absolute transfer error for expanded AADT estimates is 6.7%, and this value varied by traffic pattern group from 5% to 10.5%. Reference percentiles of the error distribution show that almost all errors are between −20% and +30%. Error decreases substantially by using a 48-h count instead of a 24-h count, and only slightly by using two counts instead of one. Weekday counts are superior to weekend counts, especially if the count is only 24h. Mean absolute transfer error increases with distance to control station (elasticity 0.121, p=0.001), and increases with AADT (elasticity 0.857, p<0.001). ImplicationsThese results can support evidence-based risk analysis of road safety performance measures that use AADT as inputs. Analytical frameworks for such analysis exist but are infrequently used in road safety because the evidence base on AADT uncertainty is not well developed.

Improved Annual Average Daily Traffic Estimation Processes

Annual average daily traffic (AADT) volume is estimated for federal, state, and local uses for roadway segments throughout the United States. The AADT estimates derived from permanent or portable counts are critical to roadway planning, system operations, and distribution of funding. The AASHTO AADT estimation formula is the most commonly used method, in part because it can be used under many circumstances when hourly traffic volume observations are missing, which is a common measurement issue with permanent traffic-counting sites. Despite its utility, the AASHTO AADT method has limitations in that it requires complete hourly data for any day that is included, and it does not account for variations in the numbers of each day of the week within a month or variations in the number of days within a month. This research evaluated new AADT estimation methods that incorporate days in which some, but not all, hourly observations are available and adjusts volume for the number of times each day of the week occurs in a month and the number of days per month. The bias and precision of four methods were examined for AADT estimation from hundreds of permanent traffic-counting sites. The analysis used more than 48 million records from the FHWA Travel Monitoring Analysis System, covering 14 years (2000 through 2013). This research quantified the improvement in accuracy and precision of the new methods, which effectively allow for a greater proportion of data to be retained in the estimation formula.

Spatial Proximity and Dependency to Model Urban Travel Demand

AbstractLink level annual average daily traffic (AADT) or travel demand is used in several urban planning, roadway design, operational, and safety analyses by transportation planners and engineers. Existing AADT estimation methods do not adequately account for spatial proximity, variations, and dependency to address modeling needs. The primary focus of this paper, therefore, is to incorporate these aspects and develop a method to estimate link level AADT by the urban road functional class. Geospatial analytical techniques were explored to capture spatial data within proximal areas of selected roadway links and develop statistical models to estimate link level AADT. Polygon-based network buffers were generated within the proximal roadway distance of each study link to account for actual connectivity and capture off-network data instead of Euclidean distance-based buffers. On-network characteristics of the study links and upstream, downstream, and cross-street network links were considered to account for th...

Use of Support Vector Machines to Assign Short-Term Counts to Seasonal Adjustment Factor Groups

The traditional method of estimating annual average daily traffic (AADT) involves several steps, with the most critical being the assignment process that involves allocating short-term counts to groups of seasonal adjustment factors (SAFs). The accuracy of AADT estimates highly depends on the assignment step, which is subject to errors stemming from human judgment. Support vector machines (SVMs), a supervised learning and statistical method, are employed to construct a series of assignment models that are compared with the traditional method and discriminant analysis (DA) models. Traffic volume data obtained from permanent traffic recorders are used to train and validate the models. The analysis is conducted with SAFs calculated for each direction of travel and separately for two-way traffic. The results reveal that the Gaussian kernel-based SVM model yields the lowest errors, improving AADT accuracy by 65% and decreasing the standard deviation of absolute percentage error by 73.7% over the traditional method. Another finding is that the assignment errors of the directional volume-based analysis are lower than those of the total volume-based analysis by 41.8%. The comparison between the two model parameters examined—the average daily traffic and the hourly factors—indicates that the combined use of both parameters in SVMs is more effective than when hourly factors are used alone. However, in the case of DA, the opposite results are obtained. A possible explanation is that the SVM kernels transfer data from the input space to a feature space and thus provide the ability to assign counts effectively with different types of data within the same model.