NCDOT Quality Control Methods for Weigh-in-Motion Data

Self Cite

A multidimensional clustering approach to generate regional average truck axle load distribution factors (ALDFs) for North Carolina is presented. The results support the Mechanistic–Empirical Pavement Design Guide (MEPDG). Weigh-in-motion data collected on North Carolina roadways are used in the analysis. A multidimensional clustering analysis based on ALDF data develops representative clusters for different highway functional classifications. Findings show that ALDF clusters have distinct characteristics for primary roads, secondary roads, collectors, and local roads. An easy-to-use decision tree based on available traffic parameters and local knowledge helps the pavement designer select the proper ALDF input. Specific contributions include a multidimensional clustering analysis that is guided by MEPDG damage-based analysis, well-defined ALDF clusters that represent specific traffic patterns in North Carolina, and a decision tree characterized by its simplicity to help pavement designers select ALDF inputs.

“…North Carolina DOT provided 12 consecutive months of calibrated volume and weight data for each WIM station. A quality control procedure confirmed that the data were reliable (16).…”

Section: Traffic Datamentioning

confidence: 79%

Axle Load Distribution for Mechanistic–Empirical Pavement Design in North Carolina

Sayyady

Stone

List

et al. 2011

Self Cite

“…The quality control procedure confirmed that the data were reliable. For the worst cases only 5% of the data were excluded (12).…”

Section: Traffic Datamentioning

confidence: 99%

Clustering Analysis to Characterize Mechanistic–Empirical Pavement Design Guide Traffic Data in North Carolina

Sayyady

Stone

Taylor³

et al. 2010

Self Cite

This paper presents attempts to generate regional average truck axle load distribution factors (ALFs), monthly adjustment factors (MAFs), hourly distribution factors (HDFs), and vehicle class distributions (VCDs) for North Carolina. The results support Mechanistic–Empirical Pavement Design Guide (MEPDG) procedures. Weigh-in-motion data support the analysis and generate seasonal factors. MEPDG damage-based sensitivity analysis shows that pavement performance is sensitive to North Carolina site-specific ALFs, MAFs, and VCDs. Similar results occur for national default values of ALF, MAF, and VCD. Hierarchical clustering analysis based on North Carolina ALFs and MAFs develops representative seasonal traffic patterns for different regions of the state. Findings show that seasonal truck traffic has distinct characteristics for the eastern coastal plain, the central Piedmont, and the western mountains. A simplified decision tree and a related table help the pavement designer select the proper representative patterns of ALF and MAF. To develop VCD factors, the approach uses 48–h classification counts and a seasonal factoring procedure to account for day-of-week and seasonal variations. The approach incorporates site-specific truck traffic to improve the accuracy of pavement design. On the basis of sensitivity analysis results, pavement performance is found to be insensitive to North Carolina site-specific and national default values of HDF; thus, the average statewide HDF values may be used as input to MEPDG. Specific contributions of this research are the relative insensitivity of pavement performance to HDF, the use of 48-h classification counts to estimate VCD inputs, and a decision tree and table to help pavement designers select the proper ALF and MAF inputs.

“…Accurate records in WIM data are essential to the success of pavement design related to long-term pavement performance and load-pavement impact (2,3). Much research has been conducted with respect to quality assurance (4) and quality control procedures (5)(6)(7) addressing the accuracy performance of various traffic patterns recorded from WIM equipment. However, little work has been done in the field of classifying vehicles that WIM systems failed to classify.…”

mentioning

confidence: 99%

Neural Network Optimal Model for Classification of Unclassified Vehicles in Weigh-in-Motion Traffic Data

Peng

Jiang

et al. 2021

A weigh-in-motion (WIM) system has the capability to perform on-site vehicle classifications based on the FHWA schema. However, WIM datasets often contain a significant portion of vehicles that could not be classified into any of the 13 vehicle classes by WIM devices. Possible reasons for the WIM classifier failing to classify these vehicles are tailgating, lane changing, traffic congestion, and equipment malfunction. Analysis of unclassified vehicles was performed with WIM-recorded data. A neural network model was established to determine the appropriate allocations of unclassified vehicles to vehicle classes. Since the number of unclassified vehicles is often fairly high, the allocations will help to improve the accuracy of truck traffic data and thus improve pavement design. Video records of traffic streams on an interstate section and traffic data from a nearby WIM station were used to identify causes for vehicle misclassifications. The optimal model was developed through model algorithm design, data processing, model training, validation, robustness analysis, and verification of video records. It was found that the optimal model was effective in allocating unclassified vehicles to appropriate vehicle classes. The optimal model was able to reclassify the unclassified vehicles that had non-zero attributes with high accuracy. The optimal model provides a useful tool for properly allocating the unclassified vehicles to the FHWA specified vehicle classes. The developed allocations can be applied to allocate unclassified vehicles appropriately to vehicle classes for pavement design and would potentially increase benefit and reduce cost with reliable and realistic pavement designs.