LTBP Program's Literature Review on Weigh-In-Motion Systems

Chapter 5. WIM Specifications and Accuracy

ASTM Specification

ASTM has established a standard specification for highway WIM systems. The latest revision of ASTM E1318 was published in February 2009, “Standard Specification for Highway WIM Systems with User Requirements and Test Methods.”⁽²⁾ The ASTM E1318-09 standard has specifications covering definitions, four various types of WIM systems, site specifications, testing and calibration requirements, data recording, and ESALs calculations. This standard is used as a guideline by most WIM users around the world.

The ASTM E1318-09 standard defines WIM as “the process of measuring the dynamic tire forces of a moving vehicle and estimating the corresponding tire loads of the static vehicle.”⁽²⁾(pg.1) In addition to tire-load information, a WIM system is capable of recording traffic data such as speed, lane of operation, date and time of passage, number and spacing of axles, and classification of each vehicle.

A WIM standard specification includes the following elements:

• Terminology.
  
  
• Type/classes: Can be classified according to application or accuracy class.
  
  
• Performance requirements: Include features/functions, applications, and tolerances (estimated tire loads, speed, and axle spacing).
  
  
• User requirements: Include site conditions (such as road geometry, surface smoothness, pavement structure, temperature ranges, etc.), recalibration procedure,
  and acceptance test.
  
  
• Test methods (to be specified by user): Include reference tire loads and axle spacing for static vehicles, calibration procedure, type-approval test, and onsite acceptance/ verification test.

Factors Affecting WIM Accuracy

Many factors can affect the accuracy of a WIM system, such as site condition, vehicle characteristics, and environment condition.

Temperature

Temperature and humidity can affect the accuracy of each sensor used in a WIM system and the accuracy of the overall WIM system. Temperature is a critical parameter because it can change the performance of many of the sensors and the pavement material properties. This can cause the contact force measured by the WIM sensor to vary at different temperatures. WIM sensors embedded directly into asphalt pavements have greater temperature variations than the sensors embedded in concrete pavements because asphalt material becomes soft in hot weather. The condition of the pavement located before and after the WIM site is also influenced by temperature, which can affect the dynamics of a vehicle as it crosses over the WIM sensors. The use of concrete pavement at the WIM site can help mitigate these effects, and WIM sensors installed in frames or housings that isolate the sensor from direct contact with the surrounding pavement are also less sensitive to temperature influences on pavement material properties. Since temperature influences on WIM system performance can never be completely eliminated, the operation of a WIM system should be validated over the full range of temperatures expected at a given WIM site. The specifications for a WIM site should include the full range of ambient temperatures that can be expected for a particular installation site, and sensors must be supplied for the WIM that are capable of providing reliable measurements while operating within the specified temperature limits.

Roughness

Conditions such as road geometry, slopes, and surface condition at the location where the WIM sensor is installed can affect the WIM measurement. Among these factors, road surface roughness has the most significant effect on the accuracy of a WIM sensor. Pavement roughness is defined as the deviation of a surface from a true planar surface with characteristic dimensions that affect vehicle dynamics and ride quality.⁽⁶²⁾ Short wavelength roughness affects the axle motion, and long wavelength roughness causes vehicle body motion.⁽³⁸⁾ This can cause variation of the dynamic axle force measured by the WIM sensor. The ASTM E1318-09 standard requires that the surface of the paved roadway 200 ft in advance and 100 ft beyond the WIM system sensors shall be smooth before sensor installation.⁽²⁾ The standard further stipulates that the surface smoothness shall be maintained such that a 6-inch diameter by 0.125-inch-thick circular plate cannot pass under a 16-ft-long straightedge that is swept across the lane at different distances from the WIM sensors.⁽²⁾ Recently, AASHTO published a provisional standard for pavement smoothness requirements in the approach to WIM systems.⁽⁶³⁾ The field operations guide developed for LTPP WIM sites specifies pavement smoothness criteria applicable for the 900-ft approach to the WIM sensors and for 100 ft after these sensors.⁽²⁹⁾ Longitudinal smoothness can be checked using a straightedge procedure in the 400-ft approach to the WIM sensors and for 100 ft beyond the sensors to determine if short wavelengths of dynamic vehicle motions are within acceptable limits.⁽²⁹⁾ The guide also has specifications for checking the transverse smoothness of the pavement at the WIM sections using the straightedge procedure.

Vehicle

Many vehicle characteristics, including speed, tire type and inflation pressure, suspension system, and axle configurations, affect the dynamic tire force, thus affecting WIM sensor measurement as well. The effects of vehicle characteristics on WIM sensor accuracy is interconnected with the effect of road surface roughness as the dynamic tire force is dependent on both factors.

Installation and Calibration of Permanent WIM Systems

Best practices for WIM installations are available in the States’ Successful Practices Weigh-in-Motion Handbook.⁽⁶³⁾ The LTPP Field Operations Guide for SPS WIM Sites updated these recommendations based on the experience gained from the pooled fund study on traffic data collection.^(29,34) Some of these important recommendations are as follows:

• Installation must be done in good weather, not wet, freezing, or hot conditions.
  
  
• The sensors must be flush (within 0.04 inch) with the road surface.
  
  
• The top of the sensor must be separate from the road surface.
  
  
• The equipment must be protected from water and dust.
  
  
• The equipment cabinet must protect the system electronics from extreme temperatures, dust, humidity, and insect and rodent infestation.
  
  
• The equipment must be protected from lightning and power surges.
  
  
• The equipment must be installed so that routine maintenance can occur without disruption of data collection.

The States’ Successful Practices Weigh-in-Motion Handbook also summarizes the installation procedures for the WIM system for bending plates, load cells, and piezoelectric sensors.⁽⁶³⁾

The following is the installation process for a bending plate:

1. Initial test.
   
   
2. Prepare the road.
   
   
3. Install scale frames in the scale pit.
   
   
4. Install scale pads in the scale frame.
   
   
5. Final test.

The following is the installation process for a load cell:

1. Initial test.
   
   
2. Prepare the road.
   
   
3. Prepare the pit to receive the scale frames.
   
   
4. Prepare the scale frames.
   
   
5. Install scale frames in the scale pit.
   
   
6. Clean up the frame installation.
   
   
7. Prepare the single load cell scale pads.
   
   
8. Install the single load cell scale pads into the frame.
   
   
9. Test the operation of the load cell.
   
   
10. Final test.

The following is the installation process for a piezoelectric sensor:

1. Initial test.
   
   
2. Preparing the road.
   
   
3. Install sensor in the main slot.
   
   
4. Final test.

The measurements by WIM systems are made under dynamic conditions. This can result in difficulties associated with determining the reference value for the calibration procedure and developing a method for a WIM system’s accuracy assessment. A system calibration must be applied immediately and repeated periodically after the initial installation of a WIM system.

In order to ensure that WIM systems give estimated weights that are as close to the actual static weights as possible, a calibration procedure is required. Factors such as pavement temperature, vehicle speed, and pavement conditions affect the estimated weight. ASTM 1318 procedures recommend the following process to calibrate WIM systems:⁽²⁾

1. Adjust all WIM system settings to the vendor’s recommendations or to a best estimate of proper setting based on previous experience for the initial calibration.
   
   
2. Force vehicles that go through the system for calibration purposes to enter into the static scales at the site or a nearby facility to obtain static weight data. With a radar gun or other means, take speed data to measure the speed of the truck over the WIM sensors.
   
   
3. Record wheel loads and/or axle weights and axle spacing at the static scales.
   
   
4. Calculate the difference between the WIM system estimate and the reference value for speeds, wheel loads, axle loads, axle group loads, GVWs, and axle spacing measurements. Express the differences in percent, and obtain a mean value for each set of measurements.
   
   
5. Enter the calibration factors into the WIM system.
   
   
6. Determine whether the calibrated system can be expected to perform at the necessary tolerances. If the differences are greater than the ASTM specified tolerance values for a specified system, then the system is not expected to perform well.
   
   
7. Note precision and bias information, although no procedure has been developed to determine what effect this data has on WIM system performance at this time.

Calibration procedures for WIM systems are summarized in NCHRP Synthesis 359: High Speed Weigh-in-Motion System Calibration Practices.⁽⁶⁵⁾ The LTPP Field Operations Guide for SPS WIM Sites provides recommended field validation and calibration procedures and guidance for determining how often they should be performed to ensure that the WIM system data are reliable.⁽²⁹⁾

The calibration of WIM scales must be checked and, if necessary, revised to be in accordance with the LTPP procedures.⁽²⁹⁾ According to this guide, the LTPP Program anticipates that a maximum of three validation sessions (installation verification or calibration and with two additional validations) is needed for the first year of operation at most scale sites. The LTPP Program recommends that a minimum of two validations be performed each year for WIM sites where the environmental conditions did not change significantly during the year. The LTPP Program recommends that only one validation test would be needed per year if the testing proved that a given scale system (as installed) was operating accurately under the full range of environmental and highway operating conditions. A remote office monitoring process for the WIM systems is used to detect whether there is calibration drift for a given system, requiring additional system calibration tests.

The main function of WIM scale systems is to estimate the static weight of different vehicles within specific tolerances (see table 3).^(2,66) In order to determine these tolerances, the percentage of weight difference between static and dynamic loads must be calculated. The errors must then be converted into percentage form for evaluation. The standard deviation of that error can then be used to determine the 95-percent confidence limit.

Minimum requirements are important for these calibration steps and must follow the manufacturers’ calibration instructions. Different vehicle speed ranges, temperatures, and/or GVWs may require separate calibration factors for some systems but not for others. The system must work correctly at all times under different traffic and climatic conditions.⁽⁶⁷⁾

Examining the influence of vehicle classification is the second stage of WIM equipment calibration. Field checks are necessary for all WIM system parts, such as AVC and automatic vehicle identification systems, to check the status of the algorithm status. This calibration review involves extensive testing of the algorithm itself. In addition, it is important to test the classification results of the WIM system against those produced by an agency’s AVC equipment to ensure that the results from these alternative devices are compatible.

LTPP Program’s WIM Experience

A significant objective of the FHWA’s LTPP Program is to develop a comprehensive understanding of the relationship between pavement performance, truck volumes, and axle loadings. The LTPP Program implemented a traffic pooled fund study, TPF-5 (004), Long-Term SPS Traffic Data Collection, to fill in gaps and improve the quality and quantity of monitored traffic data.⁽³²⁾ The objective of the pooled fund study was to improve the quality and increase the quantity of monitored traffic data.⁽³⁴⁾ The knowledge gained from a series of pilot studies conducted prior to implementing the LTPP Program pooled fund study is summarized in a report, SPS Traffic Site Evaluation—Pilots Summary and Lessons Learned.⁽⁶⁵⁾ The following significant conclusions were made:

• The ASTM E-1318 Type I WIM system performance specification is achievable with current technology and practices.
  
  
• The recommendation of bending plate sensors in smooth portland cement concrete (PCC) will meet the performance specification.
  
  
• The minimum two trucks with a 3-by-3 matrix of temperatures and speed is sufficient to determine whether the weight data is of research quality.
  
  
• Smoothness is an important factor in determining variability of weights.
  
  
• The current acceptance criteria for straight edge and profiler measurements is inappropriate.
  
  
• The straight edge and the profiler smoothness methodologies are not equivalent.
  
  
• A better definition of “smooth enough” must be developed for both the straight edge and profiler evaluations.
  
  
• Attention needs to be paid to PCC/asphaltic concrete interfaces where they exist on the site.
  
  
• A go/no-go standard needs to be established to determine which sites are suitable for evaluation and which will not produce research quality data without prior remedial work.
  
  
• A greater emphasis on vehicle classification and a measure of the quality of that aspect of the data is needed.
  
  
• The ability to do the analysis of data on site is needed.

The objective of the pooled fund study was to improve the quality and increase the quantity of monitored traffic data (i.e., volumes, classifications, and weights) at the LTPP Program SPS test sites. The study was divided into two concurrent phases due to the scale of the work performed. The scope for phase I of the study included assessing existing WIM equipment for its potential to meet the LTPP Program’s precision requirements and performing annual field validations of new and existing WIM equipment. The scope of phase II included evaluating the suitability of sites for installing new WIM systems; installing, calibrating, and maintaining new WIM systems; collecting and validating WIM data; and maintaining WIM systems in accordance with a 5-year warranty period.

SPS Traffic Site Evaluation—Pilots Summary and Lessons Learned also indicates that the 500-ft length of slab at the PCC sites was found to be a reasonable minimum, but this should be considered a minimum criteria rather than a fixed length specification.⁽⁶⁸⁾

The knowledge gained related to collecting research quality traffic data from the LTPP Program WIM sites in phase 2 of the study is reflected in the LTPP Traffic Data Collection and Processing Guide, Version 1.3 and the LTPP Field Operations Guide for SPS WIM Sites.^(69,29) The former document presents the processes and procedures used by the LTPP Program to collect and store the traffic data that is used to estimate pavement loadings, while the latter document contains the guidelines for traffic data collection at SPS sites. The second report covers all aspects of the process in the field and is divided into six major operational sections: (1) Site Assessment, (2) Site Validation—Weight, (3) Site Validation—Classification, (4) Pavement Smoothness, (5) WIM Equipment Installation & Calibration Auditing, and (6) Data for Use in LTPP Activities. The LTPP Program pooled fund research project demonstrated that research-quality traffic load and characterization data can be collected from WIM systems for an extended period of time.^(34,36) These guide documents serve as important references for guiding the selection, installation, operation, validation, and maintenance of a WIM system expected to provide reliable traffic data.