Weigh-in-motion Research Articles

Train Classification Using a Weigh-in-Motion System and Associated Algorithms to Determine Fatigue Loads.

This paper presents a methodology for classifying train passages into different types with a weigh-in-motion (WIM) system to allow the calibration of railway fatigue load models and identify individual vehicles from the measurements for the continuous calibration of railway WIM stations from in-service trains. The quality assurance of the measured responses is demonstrated using statistical methods. This paper discusses the measurement station, the method used for processing the raw data, the algorithm used to identify the train types and vehicles automatically, and the limits of the obtained load spectra. The measurement errors are demonstrated to be satisfying for use in fatigue load model calibration. Furthermore, this paper proposes actions for accurately obtaining the actual traffic conditions and describes the future work required in this area.

A Non-parametric Bayesian Network for multivariate probabilistic modelling of Weigh-in-Motion System Data

Weigh-in-motion (WIM) systems help to collect data such as vehicular loads, individual axle loads, vehicle type, and number of axles. This is relevant in engineering because traffic load performs an essential function in the design of new bridges and in the reliability assessment of existing ones, in traffic analysis and other areas of engineering. Therefore, when WIM data is not available, computing synthetic WIM observations that adequately approximate the statistical dependence between variables is important. In this paper, WIM measurements from the Netherlands and Brazil were analysed, and a set of non-parametric Bayesian Networks (NPBNs) is presented. This paper significantly improves on previous results by allowing observations of inter-axial distance to be generated, by allowing several sources of data to be used in the modelling and by making software available to researchers and practitioners interested for generating synthetic observations based on the distribution of vehicle type. In particular, statistical models to describe the weight and length of different vehicle types are derived. Three NPBNs were quantified using data from: (i) six WIM locations of the motorway network of the Netherlands, (ii) one WIM location in one city route of Rotterdam, The Netherlands, and (iii) one WIM location of one highway in Araranguá city located in the south of Brazil. Additionally, a Graphical User Interface (GUI) for the six Dutch WIM motorways locations was developed. To illustrate a possible use of the model when WIM data is not available. The GUI was used to compute synthetic WIM observations using data collected through traffic counters gathered in Toluca city in central Mexico, as input. This paper shows that the methodology here presented is widely applicable and depends only on the assessment of vehicle type configuration.

Identification of Moving Load Characteristic on Pavement Using F-P Cavity Fiber Optical Technology

The weigh-in-motion (WIM) system is a necessary piece of equipment for an intelligent road. It can provide real-time vehicle weight and lateral distribution data on wheel load to effectively support pavement structure design and service life analysis for autonomous driving. This paper proposed an enhanced weigh-in-motion sensors system using Fabry–Pérot (F-P) cavity fiber optical technology. Laboratory testing was performed to evaluate the feasibility of the proposed system and field application was conducted as well. The laboratory results show that the traffic loads could be obtained by measuring the center wavelength changes in the embedded F-P Cavity tunable filter. The laboratory results also show that the vehicle load and the number of vehicle axles can be estimated based on the system transfer function between the dynamic loading and the wavelength variation. The field application indicates that the weighting accuracy of the proposed system could reach 94.46% for moving vehicles, and the vehicle passing speed is the potentially relevant factor. The proposed system also has the ability to estimate the number of vehicle axles and the loading position, and the precision could reach 97.1% and 300 mm, respectively.

Comparison analysis between plan and actual remaining service life of road pavement using Weight in Motion (WIM) data

Transportation mobility is increasing every year and road infrastructure experiences increased loads. If the traffic load exceeds the planned load, premature damage can occur immediately and need more resources to maintain the damage. Therefore, Weigh in Motion (WIM)is needed to analyze the actual vehicle load on road infrastructure. The purpose of this study is to identify the Remaining Service Life and direct its maintenance patterns. By using WIM, the value of Remaining Service Life can be estimated. This paper presents the analysis result of the Weight in Motion data at Serang Barat Toll Gate in Indonesia and the impact on the Remaining Service Life. The axle loads were converted to represent single-axle loads based on 4th power formula by MDP 2017. The analysis showed that the CESA value obtained by using WIM data increased by 60.91% compared to the planned load and would cause a decrease in Remaining Service Life from the 20-year plan period to 16.82 years.

Data Loss Reconstruction Method for a Bridge Weigh-in-Motion System Using Generative Adversarial Networks.

In the application of a bridge weigh-in-motion (WIM) system, the collected data may be temporarily or permanently lost due to sensor failure or system transmission failure. The high data loss rate weakens the distribution characteristics of the collected data and the ability of the monitoring system to conduct assessments on bridge condition. A deep learning-based model, or generative adversarial network (GAN), is proposed to reconstruct the missing data in the bridge WIM systems. The proposed GAN in this study can model the collected dataset and predict the missing data. Firstly, the data from stable measurements before the data loss are provided, and then the generator is trained to extract the retained features from the dataset and the data lost in the process are collected by using only the responses of the remaining functional sensors. The discriminator feeds back the recognition results to the generator in order to improve its reconstruction accuracy. In the model training, two loss functions, generation loss and confrontation loss, are used, and the general outline and potential distribution characteristics of the signal are well processed by the model. Finally, by applying the engineering data of the Hangzhou Jiangdong Bridge to the GAN model, this paper verifies the effectiveness of the proposed method. The results show that the final reconstructed dataset is in good agreement with the actual dataset in terms of total vehicle weight and axle weight. Furthermore, the approximate contour and potential distribution characteristics of the original dataset are reproduced. It is suggested that the proposed method can be used in real-life applications. This research can provide a promising method for the data reconstruction of bridge monitoring systems.

Fatigue reliability assessment and life-cycle cost analysis of roadway bridges equipped with weigh-in-motion systems

The fatigue safety assessment of critical details in steel bridges is dominated by several uncertainties mainly related to load modelling. The use of integrated weigh-in-motion (WIM) systems in roadway bridges allows to collect all the necessary information for tracking actual traffic conditions. Based on WIM data, the uncertainty associated to the evaluation of the probability of fatigue failure is reduced, allowing the optimal scheduling of repair and maintenance activities and thus improving safety and economic savings. The paper presents a methodology for estimating fatigue life cycle cost (LCC) of roadway bridges equipped with a WIM system that considers uncertainties in measurement and modelling errors. The procedure allows the quantification of the economic gain enabled by WIM systems, useful in assessing the cost/accuracy trade-off for a given WIM system. It is found that the continuous use of a WIM system may improve the rate of decrease of the structural reliability index, compared with temporally limited use of WIM data. The periodic update of the reliability index and of the probability of failure based on WIM measurements results in a rather accurate estimation of the fatigue life cycle cost, which is crucial in identifying the payback time of the WIM system.

High Accuracy Weigh-In-Motion Systems for Direct Enforcement.

In many countries, work is being conducted to introduce Weigh-In-Motion (WIM) systems intended for continuous and automatic control of gross vehicle weight. Such systems are also called WIM systems for direct enforcement (e-WIM). The achievement of introducing e-WIM systems is conditional on ensuring constant, known, and high-accuracy dynamic weighing of vehicles. WIM systems weigh moving vehicles, and on this basis, they estimate static parameters, i.e., static axle load and gross vehicle weight. The design and principle of operation of WIM systems result in their high sensitivity to many disturbing factors, including climatic factors. As a result, weighing accuracy fluctuates during system operation, even in the short term. The article presents practical aspects related to the identification of factors disturbing measurement in WIM systems as well as methods of controlling, improving and stabilizing the accuracy of weighing results. Achieving constant high accuracy in weighing vehicles in WIM systems is a prerequisite for their use in the direct enforcement mode. The research results presented in this paper are a step towards this goal.

Improved multi-lane traffic flow simulation based on weigh-in-motion data

Vehicle load is the main variable load of highway bridges, which significantly affects the security and duration of bridges. Highly precise vehicle load models are the basis for studying vehicular load responses and providing reliability assessments of highway bridges, whose core is the spatio-temporal distribution of vehicular load on a bridge deck. Vehicle type, axle weight, axle distance and position are the main parameters of spatio-temporal simulation of traffic flow. Weigh-in-motion (WIM) systems can obtain these original vehicle parameters efficiently. However, there are two main problems about how to accurately simulate the above traffic flow parameters based on WIM data: (1) A large number of WIM systems have low accuracy for vehicle classification and an efficient classification method for vehicles from massive WIM data has not been proposed; (2) The most popular MC simulation method ignored the statistical dependence of parameters (vehicle sequence, axle weight, axle distance, etc.), which results in the deviation between traffic flow simulation samples and measured data. In this study, to obtain traffic flow simulation results closer to measured data, a multi-lane traffic flow simulation method via fusion of efficient vehicle classification and the statistical dependence of parameter was proposed. First, the overall framework of the traffic simulation was introduced. Subsequently, the vehicle classification process and traffic flow simulation were introduced respectively. For the vehicle classification, the abnormal vehicles were removed based on the box diagram method, and then the vehicle classification was completed based on the improved K-means++ clustering algorithm. The vehicle sequence and the axle weight/distance samples were obtained by MCMC simulation method and Copula function theory respectively. Finally, a case study was presented using WIM measured data of an 8-lane expressways. The results showed that the optimal number of clusters determined based on the K-means++ clustering algorithm was reasonable. The vehicle simulation samples considering the statistical dependence of parameter were closer to the measured data than previous methods and the normal copula function was slightly better than the t-copula function.

Designing the Calibration Process of Weigh-In-Motion Systems

Weigh-In-Motion (WIM) systems provide information on the state of road traffic and are used in activities undertaken as part of traffic supervision and management, enforcement of applicable regulations, and in the design of road infrastructure. The further development of such systems is aimed at increasing their measurement accuracy, operational reliability, and their resistance to disturbing environmental factors. Increasing the accuracy of measurement can be achieved both through actions taken in the hardware layer (technology of load sensors, the number of sensors and their arrangement, technology used in the construction of the pavement, selection of the system location), as well as by implementing better system calibration algorithms and algorithms for pre-processing measurement data. In this paper, we focus on the issue of WIM system calibration. We believe that through the correct selection of the calibration algorithm, it is possible to significantly increase the accuracy of vehicle weighing in WIM systems, from a practical point of view. The simulation and experimental studies we conducted confirmed this hypothesis.

Analytical evaluation of the rutting critical response function under the heavy vehicle’s moving in flexible pavement

Accurate and realistic identification of critical response of pavement structure leads to sustainable pavement structure design. This study aimed to obtain and analyse the rutting critical response function under heavy trucks using Abaqus three-dimensional finite element and particle swarm optimization (PSO) methods. For modeling, asphalt viscoelastic properties, data of the Weigh-In-Motion (WIM) device, and the Moving Load method were used to simulate trucks passing over the pavement. The thickness of pavement layers and their elastic and viscoelastic properties were calculated by field study and theoretical relationships, respectively. The truck specifications including loads, axles and wheels configurations, and dimensions were collected by WIM data and field survey. The results showed that the best statistical equation that could be fitted to critical responses was the modified Gaussian function. Also, according to analysis results, the effect of truck speed on the rutting critical responses was greater than its weight.

Machine learning of Truck Traffic Classification groups from Weigh-in-Motion data

The pavement Mechanistic-Empirical (ME) design requires high-dimensional traffic feature inputs by categories, including Vehicle Class Distributions (VCD), Monthly Distribution Factors (MDF), Hourly Distribution Factors (HDF), and Normalized Axles Load Spectra (NALS). In simplifying the Pavement ME design practice, Truck Traffic Classification (TTC) groups are commonly used for characterizing traffic inputs. Thus, properly defining TTC groups is critical for state-specific pavement ME design practice. In this study, the truck traffic data from existing Weight-in-Motion (WIM) stations were mined to develop specific TTC groups to assist with pavement ME design practice in Georgia. An effective data analytics procedure was developed by leveraging unsupervised machine learning techniques to reduce the high-dimensional traffic features by stratified Principal Component Analysis (PCA), followed by K-means clustering to establish appropriate TTC groups. For a case study, the performance of two typical designs was evaluated using the AASHTOWare pavement mechanistic-empirical (ME) design software with respect to two scenarios of traffic inputs: (1) the derived cluster-based groups, and (2) the national default TTC groups. The results indicated that direct application of the national default TTC groups resulted in over-design of pavement structure in Georgia. Therefore, it is highly recommended that customized TTC groups should be developed using state-specific WIM data.

Development and verification of winter traffic hazard models for the Alberta highway network

This paper develops a weather traffic model and verifies its validity through temporal transferability within the context of the Alberta highway network. This research used traffic data and weather data collected at four weigh-in-motion (WIM) sites and weather stations for five years to develop models for two-vehicle classes. We collected an additional year of traffic data from the exact four WIM locations to conduct a temporal transferability test. We evaluated the estimation accuracy resulting from the temporal transferability by measuring the value of R2and the magnitude of five error measures. All in all, the developed models were transferred successfully to a different year. The study results reveal that different structural types of models could be better suitable for each vehicle type at other road functions.

A novel traffic conflict risk measure considering the effect of vehicle weight

Introduction: Vehicle weight is deterministic to the impact force in collision, and thus the injury risk of vehicle occupants. In China, involvement of heavy vehicles in overall and fatal crashes are prevalent, even though heavy vehicles only constitute a small proportion of overall registered motor vehicles. However, vehicle weight is rarely considered in the existing traffic conflict risk prediction and assessment models because of the unavailability of required data. Method: Novel risk indicators for the diagnosis of traffic conflict risk map, considering the effect of vehicle weight, are proposed, with the advantage of comprehensive traffic flow characteristics and vehicle weight data using Weigh-in-Motion (WIM) technique. Weight-incorporated risk level (WRL) and weight integrated risk level (WIRL) are established to quantify the traffic conflict risk, at an instant and over a specified time period, respectively, by extending the conventional traffic conflict risk measures including time-to-collision (TTC) and modified potential collision energy (PCE). Then, a microscopic traffic simulation model is adopted to estimate the traffic conflict risk map along a highway segment that has partial lane closure. The traffic conflict risk performances, between the risk indicators with and without considering the vehicle weight, are compared. Results: The traffic conflict risks estimated using conventional risk indicators without considering the vehicle weight are generally lower than that based on WRL and WIRL. The difference is more profound when the proportion of heavy vehicles in the traffic stream increases. Conclusions: The finding is indicative to remedial engineering measures including variable message sign, speed limit, and ramp metering that can mitigate the real-time crash risks on highways, especially in adverse environmental and weather conditions, with due consideration of vehicle composition and crash worthiness of vehicles.

Incorporating the unevenness of lane truck loading into fatigue load modeling of multi-lane bridges

The current codes in practice define fatigue load models (FLMs) to characterize the bridge damage that is induced by traffic loads. However, most FLMs are specified to load on any single traffic lane, and the characteristics of the uneven traffic load distribution over multiple lanes under real-world conditions are not revealed. This study analyzed the uneven distribution characteristic of truck loading over multiple lanes based on multiple sets of weigh-in-motion (WIM) data and investigated its effect on the fatigue performance of bridges. A novel multi-lane FLM is proposed, comprising a standard fatigue truck model and multi-lane fatigue coefficients. The standard fatigue truck model is the same as FLM II of the Chinese design specification; however, the load value is adjusted using a calibrated coefficient correlated with the bridge length. The multi-lane fatigue coefficients help define the reduction coefficients for the load value and number of fatigue cycles when applying the standard fatigue truck model to a certain traffic lane. Finally, the proposed model and the code-based model are applied to evaluate the fatigue performance of typical steel–concrete composite girder bridges, and the results are compared with the estimations using realistic multi-lane WIM data. The proposed model yields similar estimations of the stress range and fatigue life as using WIM data, whereas the code-based model shows large deviations. The comparisons validate the feasibility and superiority of the proposed multi-lane FLM for the fatigue evaluation of bridges under uneven multi-lane truck loadings.

Testing the accuracy of WIM systems: Application to a B-WIM case

The assessment of the accuracy of a weigh-in-motion (WIM) system is described in this paper. It is based on a statistical composite hypothesis test for which the Type I error is the supplier risk for the WIM system. This statistical test is built in the asymptotical setting, namely for large samples of measurements, with an implicit estimator of the tolerance. Considering an observation sample of independent and identically distributed relative errors, this procedure is used in the classical WIM assessment method described in the COST 323 action and compared to the classical estimator of the tolerance. The test also gives formulæ to practitioners implying the observation sample size comparing to the usual estimator for which the observation sample size is tabulated. An application on a bridge-WIM system is also done.

Incorporating weather variables with probabilistic approach for trip planning

Adverse weather conditions are known to negatively impact traffic operations by creating unsafe road conditions, slowing vehicles, and causing congestion – events that subsequently lead to extra costs and travel time delay if they are not considered explicitly in trip planning. This study proposes an approach that considers incorporates adverse weather conditions into trip planning modeling. As such, speed data from Weigh-in-Motion (WIM) sensors along with weather information from multi-region within the State of Ohio were used. To assess departure time and route choice under varying levels of snowstorm events, a probabilistic model was developed. Travel time probability is used to quantify the impact of weather events on travelers' decisions regarding both route choice and departure time. As expected, the results showed that heavy snow associated with low visibility has the most impact on driver speed behavior. A measure of driver's speed variability associated with different weather conditions is provided for the collected data from the studied regions in Ohio. Moreover, some drivers showed hesitation during the storm and dropped their speed to as low as 20 mph. In addition, the vehicle speed performance during the events was used to calculate the expected time throughout each event. Simulations on a road network were used to compare the expected total travel time and the variance along with the probability of paths under a heavy snowstorm. Results indicated that, under adverse weather conditions, drivers' route choice is not based on the shortest travel time but on the decision of alternate routes being attractive (safe, snow cleared, etc.). Giving the additional trip information of these events, decision-makers should pay more attention before deciding the departure and route choice on heavy snow events.

A deep neural network-based vehicle re-identification method for bridge load monitoring

The accurate tracking of vehicle loads is essential for the condition assessment of bridge structures. In recent years, a computer vision method that is based on multiple sources of data from monitoring cameras and weight-in-motion (WIM) systems has become a promising strategy in bridge vehicle load identification for structural health monitoring (SHM) and has attracted increasing attention. The implementation of vehicle re-identification, namely, the identification of the same vehicle from images that were captured at different locations or time instants, is the key topic of this study. In this study, a vehicle re-identification method that is based on HardNet, a deep convolutional neural network (CNN) specialized in picking up local image features, is proposed. First, we obtain the vehicle point feature positions in the image through feature detection. Then, the HardNet is employed to encode the point feature image patches into deep learning feature descriptors. Re-identification of the target vehicle is achieved by matching the encoded descriptors between two images, which are robust toward scaling, rotation, and other types of noises. A comparison study of the proposed method with three published vehicle re-identification methods is performed using vehicle image data from a real bridge, and the superior performance of our proposed method is demonstrated.

Innovative Carbon-Doped Composite Pavements with Sensing Capability and Low Environmental Impact for Multifunctional Infrastructures

Recently, smart composites that serve as multi-functional materials have gained popularity for structural and infrastructural applications yielding condition assessment capabilities. An emerging application is the monitoring and prediction of the fatigue of road infrastructure, where these systems may benefit from the ability to detect and estimate vehicle loads via weigh-in-motion (WIM) sensing without interrupting the traffic flow. However, off-the-shelf applications of WIM can be improved in terms of cost and durability, both on the hardware and software sides. This study proposes a novel multi-functional pavement material that can be utilized as a pavement embedded weigh-in-motion system. The material consists of a composite fabricated using an eco-friendly synthetic binder material called EVIzero, doped with carbon microfiber inclusions. The composite material is piezoresistive and, therefore, has strain-sensing capabilities. Compared to other existing strain-sensing structural materials, it is not affected by polarization and exhibits a more rapid response time. The study evaluates the monitoring capabilities of the novel composite according to the needs of a WIM system. A tailored data acquisition setup with distributed line electrodes is developed for the detection of moving loads. The aim of the paper is to demonstrate the sensing capabilities of the newly proposed composite pavement material and the suitability of the proposed monitoring system for traffic detection and WIM. Results demonstrate that the material is promising in terms of sensing and ready to be implemented in the field for further validation in the real world.

Data-driven analytical load rating method of bridges using integrated bridge structural response and weigh-in-motion truck data

Load rating is a widely used approach for evaluating the load-carrying capacity of bridges in an effort to ensure safe bridge operation under expected traffic loads. Load rating often relies on simplified analytical models including empirically derived model parameters that do not reflect bridge-specific information resulting in conservative ratings. To reduce this conservatism, this study proposes a novel data-driven framework that utilizes long-term bridge response data to extract bridge-specific model parameters that can be used within in the Load and Resistance Factor Rating (LRFR) process. The data-driven LRFR (DD-LRFR) framework is empowered by a cyber-physical system (CPS) architecture that uses Internet connectivity to integrate measured bridge responses with truck weights measured by a weigh-in-motion (WIM) station. The CPS architecture uses computer vision of camera images to confirm trucks observed at a WIM station are identical to those observed at a bridge. Bridge response and axle weight data are then used to extract probabilistic models of dynamic load allowances and unit influence lines. The DD-LRFR method is validated using a 20-mile (32.2-km) segment of the I-275 northbound highway in Michigan that is monitored continuously by the CPS architecture. Six girders associated with two bridges along I-275 are rated using the proposed DD-LRFR methodology with rating factors compared to those obtained using conventional and refined load rating methods. The DD-LRFR method yields inventory- and operational-level rating factors that are less conservative than those from the approximate LRFR method and comparable to those using finite element modeling of the bridge.

Simple Span Bridge Loading Based on Weight in Motion Data

Over Dimension Over Load (ODOL) vehicle is claimed to be the cause of a number of fatal accidents and damage to road infrastructure in Indonesia and has become a serious issue for roads and bridges maintenance. Over loaded vehicles can result in damage to the bridge, reduced bridge capacity and can even result in collapse of the bridge. Therefore the characteristics of the actual vehicle load need to be known and compared to the bridge design load according to the bridge design code and standard. In this paper, an analysis is carried out on the actual traffic load data obtained from the measurement results using Weight in Motion (WIM) in Kaligawe, Semarang, Indonesia. Monte Carlo simulation is used to simulate the arrival of the vehicle. The results of vehicle load simulation are then applied to a simple span bridge with a span of 30, 40, 50 m. The bending moment and shear force obtained due to actual load compared to the magnitude of the bending moment and shear by applying Indonesia Bridge Loading Standard. The results show that 43 percent of vehicles are overloaded. The bending moment and shear force caused by actual load is less than the design bending moment and shear force for 30 m, 40 m and 50 m bridge span.