Annual Average Daily Traffic Estimation Research Articles

Effect of missing values estimations on traffic parameters

Estimating missing values is known as data imputation. The proper imputation of missing values of permanent traffic counts (PTCs) could result in significant cost savings for highway agencies in their traffic data programs. However, little research has been done on missing values and only limited research has used factor or time series analysis models for predicting them. Moreover, studies of the effect of the imputations on traffic parameters estimations are not available. This study used factor models, genetically designed neural network and regression models, and autoregressive integrated moving average (ARIMA) models to update pseudo‐missing values of six PTCs from Alberta, Canada. The influences of these imputations on the estimations of annual average daily traffic (AADT) and design hourly volume (DHV) were studied. It was found that simple models usually resulted in large AADT and DHV estimation errors. As models were refined, resulting estimations for individual missing hourly volumes significantly improved. Usually these models provided highly accurate AADT and DHV estimations.

Using Geographically Weighted Regression Models to Estimate Annual Average Daily Traffic

Annual average daily traffic (AADT) is important information for many transportation planning, design, operation, and maintenance activities and for funding allocation. Many studies have attempted AADT estimation by using various methods, such as the factor approach, regression analysis, and artificial neural networks. An application of geographically weighted regression (GWR) methods for estimating AADT is presented. GWR models allow model parameters to be estimated locally instead of globally, as in the case of ordinary linear regression (OLR) analysis. The spatial variation of the parameter estimates and the local R2 from the GWR models were investigated, and the AADT estimation errors were analyzed. Compared with OLR models, the GWR models were more accurate and were useful for studying the effects of the regressors at different locations.

Wavelet Transform for Feature Extraction to Improve Volume Adjustment Factors for Rural Roads

Wavelet transform is a powerful technique for feature extraction from data that are characterized by frequent and large-scale fluctuations. This technique has been studied in recent years for incident detection and intelligent transportation systems data aggregation. Because of its attractive properties, such as time frequency localization, multirate filtering, and scale-space analysis as well as multiresolution analysis, this research explores the potential application of the technique to improve the accuracy of monthly and weekly adjustment factors in rural annual average daily traffic estimation. The technique was applied to data collected from automatic traffic recorder stations in northern Ohio, and the application showed that the technique could separate and filter out the high-frequency data oscillations and trace the changes in the fundamental pattern of traffic. The test results illustrated an impressive improvement, ranging from 10% to 20%, over conventional data estimation at those stations. This success shows that the proposed approach is feasible, and future improvements of the application of the technique can make it an effective tool for traffic volume estimation for rural highways.

Estimating Annual Average Daily Traffic from Satellite Imagery and Air Photos: Empirical Results

Vehicles can be identified in high-resolution satellite imagery that recently has become available to the civilian community. The vehicle information contained in this imagery, and in air-based imagery, could be used in annual average daily traffic (AADT) estimation, a task conducted by many transportation agencies around the world. However, because the imagery provides information equivalent to traffic counts of very short duration, it is possible that the information is too noisy to be of use. Empirical differences between AADT estimated from 14 satellite images and air photos of Interstate segments in Ohio and the corresponding AADT estimated from traditional, ground-based estimates are presented. The distribution in differences appears relatively unbiased, implying that averaging the estimates of several images of the same segment can decrease estimate errors. The empirical errors are small enough to indicate that AADT estimation errors and ground-based sampling efforts could both be reduced by combining satellite-based data with traditional ground-based data. The differences between the image-based and the ground-based estimates are smaller in the few cases in which ground-based estimates inspired greater confidence, implying that the image-based estimates may be better than what is indicated in the distribution of differences. Evidence also suggests that the differences tend to decrease when the image leads to longer equivalent traffic count duration, indicating the potential to condition the use of the image-based data on this readily available parameter.

Estimation of Annual Average Daily Traffic on Low-Volume Roads: Factor Approach Versus Neural Networks

Estimation of the annual average daily traffic (AADT) for low-volume roads is investigated. Artificial neural networks are compared with the traditional factor approach for estimating AADT from short-period traffic counts. Fifty-five automatic traffic recorder (ATR) sites located on low-volume rural roads in Alberta, Canada, are used as study samples. The results of this study indicate that, when a single 48-h count is used for AADT estimation, the factor approach can yield better results than the neural networks if the ATR sites are grouped appropriately and the sample sites are correctly assigned to various ATR groups. Unfortunately, the current recommended practice offers little guidance on how to achieve the assignment accuracy that may be necessary to obtain reliable AADT estimates from a single 48-h count. The neural network approach can be particularly suitable for estimating AADT from two 48-h counts taken at different times during the counting season. In fact, the 95th percentile error values of about 25 percent as obtained in this study for the neural network models compare favorably with the values reported in the literature for low-volume roads using the traditional factor approach. The advantage of the neural network approach is that classification of ATR sites and sample site assignments to ATR groups are not required. The analysis of various groups of low-volume roads presented also leads to a conclusion that, when defining low-volume roads from a traffic monitoring point of view, it is not likely to matter much whether the AADT on the facility is less than 500 vehicles, less than 750 vehicles, or less than 1,000 vehicles.

Neural Networks as Alternative to Traditional Factor Approach of Annual Average Daily Traffic Estimation from Traffic Counts

Presented in this paper is a comparison of the neural network approach and the traditional factor approach for estimating annual average daily traffic (AADT) from 48-h sample traffic counts. Minnesota’s automatic traffic recorder (ATR) sites are investigated. The traditional AADT estimation approach involves application of volume adjustment factors to sample counts. The neural network model used in this study is based on a multilayered, feed-forward, and back-propagation design for supervised learning. The results of AADT estimation from a single shortperiod traffic count indicate that as compared with the neural network approach, the estimation errors for the factor approach can be lower under a scenario in which ATR sites are grouped appropriately and the sample sites are correctly assigned to various ATR groups. Unfortunately, the current recommended practice offers little guidance on how to achieve the assignment accuracy that may be necessary for obtaining reliable AADT estimates from sample counts. The advantage of the neural network approach is that classification of ATR sites and sample site assignments are not required. The neural network approach can be particularly suitable for estimating AADT from two or more shortperiod traffic counts taken at different times during the counting season.

Sources of error in estimating truck traffic from automatic vehicle classification data

Truck annual average daily traffic estimation errors resulting from sample classification counts are computed in this paper under two scenarios. One scenario investigates an improper factoring procedure that may be used by highway agencies. The study results show consistent and substantial overestimates of truck traffic when truck counts are estimated using adjustment factors obtained from total traffic volume. In the second scenario, better estimates result when the required factors are obtained from a permanent automatic vehicle classifier exhibiting a pattern of truck traffic that is similar to the pattern at the sample count site. A limited research analysis of truck type estimation from sample counts is also presented in this paper.

Traffic Count Estimates for Short-Term Traffic Monitoring Sites: Simulation Study

Traffic characteristics, such as the annual average daily traffic (AADT) and the AADT for each vehicle class, are essential for highway maintenance and planning. In practice, selected road segments are monitored continuously every day of the year to identify their traffic characteristics. A sample of the remaining road segments is monitored for 1 or 2 d each year, and the resulting data are adjusted (by using factors based on data collected from the continuously monitored road segments) to produce estimates of annual average daily traffic characteristics. A simulation study empirically considered how the precision of an estimate from a continuously monitored site compares with the precision of an estimate from a short-term monitored site. The original estimates of traffic characteristics (i.e., AADT and AADT by vehicle class) treating the site as a continuously monitored site are on average quite close to, but smaller than, the simulated estimates treating the site as a short-term monitored site. The original estimates (continuous monitoring) appear to be more precise, on average, than the simulated estimates (short-term monitoring). This decrease in precision typically occurs for vehicle classes that account for less than 1 percent of the daily traffic volume, suggesting that these less-common vehicle classes could be combined to achieve reliable AADT estimates.

Statewide Traffic Volume Studies and Precision of AADT Estimates

Investigators examined the statistical precision of annual average daily traffic (AADT) estimates derived from short period traffic counts (SPTC). Many automatic traffic recorder (ATR) sites located on Minnesota's highway system are studied. Effects of various factors on estimation errors are explored. The appropriateness of volume adjustment factors is expressed in terms of assignment effectiveness, which mirrors the degree of correctness with which a sample site is assigned to an ATR group. Investigators found that AADT estimation errors are very sensitive to the assignment effectiveness. Study findings indicate that highway agencies should put more emphasis on sample site assignments to correct ATR groups than on the duration of count. By analyzing ATR data from two Canadian provinces--Alberta and Saskatchewan--the findings are confirmed. Highway agencies may find the material presented in this paper beneficial in the rationalization of their traffic volume counting programs.

Using Seasonal and Day-of-Week Factoring to Improve Estimates of Truck Vehicle Miles Traveled

Procedures developed by FHWA for “factoring” short-duration traffic counts for seasonal and day-of-week variations in traffic volumes are capable of producing estimates of annual average daily traffic (AADT) that are quite accurate. Moreover, there is virtually no bias in these estimates, so AADT estimates for a set of road sections can be used to produce unbiased estimates of total vehicle miles traveled (VMT) for systems of roads. Unfortunately, corresponding procedures are not generally used for estimating AADT by vehicle class, and the less sophisticated procedures that are commonly used contribute to substantial overestimates of truck AADT and VMT. Current procedures apparently overestimate VMT by 25 to 40 percent for combination trucks and possibly more for single-unit trucks. Modified versions of the FHWA factoring procedure that are capable of producing substantially improved estimates of truck VMT and of AADT of combination trucks are presented. These procedures use seasonal and day-of-week factoring to reduce the errors in truck AADT estimates and to eliminate the upward bias in truck VMT estimates that result from the use of unfactored weekday classification counts.

Using Seasonal and Day-of-Week Factoring to Improve Estimates of Truck Vehicle Miles Traveled

Procedures developed by FHWA for short-duration traffic counts for seasonal and day-of-week variations in traffic volumes are capable of producing estimates of annual average daily traffic (AADT) that are quite accurate. Moreover, there is virtually no bias in these estimates, so AADT estimates for a set of road sections can be used to produce unbiased estimates of total vehicle miles traveled (VMT) for systems of roads. Unfortunately, corresponding procedures are not generally used for estimating AADT by vehicle class, and the less sophisticated procedures that are commonly used contribute to substantial overestimates of truck AADT and VMT. Current procedures apparently overestimate VMT by 25 to 40 percent for combination trucks and possibly more for single-unit trucks. Modified versions of the FHWA factoring procedure that are capable of producing substantially improved estimates of truck VMT and of AADT of combination trucks are presented. These procedures use seasonal and day-of-week facto...