Written by: Sheri Rowell
Traffic data collection is an integral tool used by cities and towns to help make decisions related to public roadways, and it’s nothing new. Data collection has been used on roadway systems since as early as the 1930’s and it’s continued to evolve with technology. Back in the 30’s the collection method was typically manual counts; however, in the 1940’s there was a transition into mechanical measurement which changed the approach to data integrity. Computerized pneumatic traffic counters revolutionized traffic data collection in the 1970’s. Now, forty years later, radars, video processing, and artificial intelligence are again revolutionizing traffic data collection, increasing reliability and cost-effectiveness. Yet, we still find many traffic departments using costly, unreliable older generation technologies such as manual data collection, pneumatic tubes etc.
The purpose of this blog is to give a simplified understanding of the different types of traffic data, traffic data collection devices, and explain how radars are currently the most cost-effective solutions for accurate, reliable traffic data collection and analysis. We will also review how different radars can be used as permanent and temporary traffic data collection devices on highways and other roads.
Why Traffic Data is Needed
Traffic volume data and statistics are foundational to the transportation industry. Transportation departments, local governments, and transportation providers rely on travel forecasts to support transportation plans and to help prioritize investment policies. Traffic impact studies have been used for decades to answer questions about current and future transportation system performance.
Traffic summary statistics are used for a wide range of purposes:
Pavement structure and condition
Roadway classification and design speed
Right of way, road centerlines and property lines
Motorized vehicles, pedestrian, bicycles that use the space (public r.o.w)
Traffic volumes and composition
Intersection turning movements
Vehicle operating speeds
Level of service, delays, and queues
Speed limits (statutory, regulations/by-laws)
Traffic signs (regulatory, warning, information, temporary conditions)
Intersection Yield/Stop/Traffic Control Signals/Crossovers
Traffic Data and Devices
As technology has evolved so has the types of traffic counting devices evolved. Traffic data collection can be captured in a variety of ways and by a variety of means.
Types of Data
Vehicle Volumes Vehicle Speed
Mid Block Counts Vehicle classification
Turning Movement Counts Lane occupancy %
Headway / Gap
Traffic data is not only useful to calculate some of the standard traffic statistics—AADT, AAWT, SADT, WADT etc.—but these statistics in turn are used for traffic control signal optimization, level of service determination, and traffic calming. Local transportation departments and federal ministry’s use these statistics to help support planning, design, construction, and operation of road networks. Traffic data collection and statistic are a worldwide initiative used to determine current traffic patterns and help predict future traffic trends.
The technologies that are used to capture traffic data can be classified into two main categories – Intrusive and non-Intrusive.
· Intrusive devices
Any traffic data collection device which needs installation on or under the surface of the roadway are classified as intrusive devices. These devices need closure of traffic lanes for installation and maintenance. Inductive loops, magnetometers and road tubes are examples of such devices.
· Non-Intrusive devices
Traffic data collection devices that do not require installation on or under the surface of the roadway are classified as non-intrusive devices. These devices are often mounted on poles installed near the shoulders or on medians and often do not need road closures for installation or maintenance. Cameras, manual counting devices, and radars are examples of such devices.
A Look at the Available Technologies
Radars are an excellent, if not the best alternative to competing technologies. However, traffic data managers are sometimes misled by old technologies or by biased information presented by marketing campaigns. Below we look at the technologies available and the pros and cons of each.
As you can see from the table above radars provide the highest accuracy for vehicle counts, speed, and accuracy for both temporary and permanent counting stations. Video-analytics is the nearest competitor and works very well for turning movement counts. However, radars prove superior when it comes to all-weather operations and affordability.
Although manual counts and road tubes are low-cost solutions, their count, speed, and classification accuracy are inferior, proving that they are outdated technologies. All weather operation is also a major issue. Inductive loops on the other hand usually give very accurate traffic counts but fails to deliver speed and classification of vehicles.
· Manual Counting
Typically used for temporary counts, this collection method has been around the longest, counting only, no speed data collected, can have a wide range for the accuracy level due to human error, challenging to get multiple lane accuracy.
· Road tubes
Used for many years, provides counts and speeds, typically limited to the summer season, cannot be used year-round in the Canadian climate, accuracy starts to degrade due to wear and tear on the tubes as they are exposed to increased traffic, operators need to enter lanes of traffic to secure road tubes – potential safety issues.
Permanent count station, very accurate for vehicle counts, does not offer speed or vehicle classification, installation initially impacts traffic flow, loops can fail due to pavement heaving through annual seasons.
Can be used for both permanent and temporary count stations, provides counts and speed analysis, count accuracy is in the 80% range and weather does affect overall accuracy; up and coming technology that bares watching.
Can be used for both permanent or temporary stations, more accurate in terms of overall count accuracy, speed accuracy and even vehicle size/classification accuracy, can be operated in all weather conditions year-round, lo
w maintenance, low capital and operational costs, variety of connectivity options to download data.
· Video analytics
Tend to be used for permanent locations, good count accuracy, speed accuracy is impacted, weather can impact accuracy levels, video analytics have and continue to improve through machine learning and AI.
Why radars as a technology are so reliable.
Radar technology has become more commonly used in the last 15 years. Companies like Houston Radar have taken the technology and developed it to target specific needs in the traffic industry. The technology has matured and been proven as a reliable, accurate source of traffic data.
Radar has a number of inherent advantages over competing technologies which we will detail below; however, there are two primary types of radar which we should review first:
· Doppler radar
A specialized radar that uses the Doppler effect to produce velocity data about objects at a distance. It does this by bouncing a microwave signal off a desired target and analyzing how the object’s motion has altered the frequency of the returned signal. This variation gives direct and highly accurate measurements of the radial component of a target’s velocity relative to the radar.
· FMCW radars
Radars that measure the differences in phase or frequency between the transmitted and the received signal, ability to measure very small ranges to the target, ability to measure simultaneously the target range and its relative velocity, very high accuracy of range measurement.
The application for data collection would determine whether Doppler or FMCW radar would be more effective. Due to the lower cost and ease of deployment we see temporary count devices typically utilizing doppler radar, while more permanent counting stations will utilize FMCW radar due to its higher accuracy, range and ability to measure vehicle length in more detail.
· Accuracy – Count, Speed & Classification
As indicated in our comparison chart the overall accuracy ranges for radars is 90-99% and have the ability to collect and classify all three types of data.
· Power consumption
Modern radar sensors operate on extremely low power, as such they consume a lot less power than camera & Lidars sensors. This enables them to be used with solar powered and battery-only options even in places like Canada with extreme climate and latitude.
· All weather operation
Radar units are NEMA rated and optimized for extended temperature ranging from -40 to 85oC (depending on the product) as there is no need for any equipment in or on the roadway radar units can be deployed at any time of the year allowing for data collection in all weather conditions.
Due to radar units smaller size, 1.5 – 3kg, and ease of deployment, with simple quick connect steel mounting, units can be deployed for short studies 1-2 weeks or by adding a small solar panel longer studies or permanent count stations.
Feature sets and configuration play a part in the overall cost for radar units, most units have a low total cost of ownership ranging from $2500 – 10,000 per unit. Annual access to cloud-based software can factor into operating costs.
Regardless of which kind of radar you use, there are a number of inherent advantages to the technology:
Biggest Drawback of Radar
While radar has several features which set it apart as the best technology for data collection, it’s not perfect. Radar technology is built on the concept/principle of motion – therefore traffic needs to be moving in order to be measured. If deployed at a location where traffic is stopped or slowed below 8km, the sensor cannot “see” the vehicles. For data collection purposes this does not impact overall accuracy by very much (since in stopped traffic, cars are not driving past the sensor) but it certainly is worth mentioning. Houston Radar has tried to mitigate this issue in its more advanced SpeedLane sensor by embedding algorithms which are used to detect when traffic is stopped in front of the sensor. While it cannot count the slow-moving vehicles, it can detect that it’s experience that stopped traffic condition and can report/alert that information.
LIDAR – A rising star, but still needs development
LIDAR, more than any other competing technology, seems to be the up-and-coming technology that may surpass radar as a better solution for traffic data collection. However, it is still a very new technology which has not yet been heavily developed, as such it is still quite expensive and likely needs 5+ years to mature and become more cost-effective.
The main advantage of LIDAR is its ability to see stopped objects as well as small objects such as bicycles and pedestrians.
Stinson ITS has tested a number of LIDAR solutions but found them inaccurate and not well matured. We’re excited to see how this technology develops but currently it is just not there yet.
How radars are being used as permanent and temporary count stations.
In a survey conducted by Stinson ITS in May 2020 we found that traffic departments in Canada are now starting to require more accurate traffic data collection and prefer non-intrusive devices over intrusive devices. However, there is a clear difference between cities and smaller towns on whether they wish to install these devices as permanent or temporary data collection devices.
Cities with higher budgets tend to prefer permanent count stations on all important and strategic roads. They also have specialized teams with analytical and software skills to analyze and present the traffic data to planning departments and other stakeholders in the city.
On the other hand, Counties and towns prefer to use temporary count stations to maximize the returns from their limited budgets. They often conduct speeding studies based on resident complaints and simultaneously record traffic data. Many road supervisors are now utilizing Excel features to generate graphs out of collected traffic data.
There are several needs to address with the collection of accurate traffic data and several methods to meet those needs. Considering each method has its advantages and disadvantages, responsible agencies must determine which combination of reliability, accuracy, ease-of-use, and cost-effectiveness best suits their purposes.
Non-intrusive radar sensors reduce workers’ exposure to traffic by taking less time for setup and maintenance and have less associated costs than intrusive sensors. Radar’s have superior accuracy, increased ease-of-use, faster installation with more verifiable alignment, and lower performance costs.