Written by: Michel Mondor, Stinson ITS’s Eastern Canada Account Manager
Travel time sensors have been an essential part of smart work areas for many years, but their use has often been very narrowly focused on displaying passenger information and not on the large amounts of data that sensors can collect. Especially in the past year, when COVID-19 has had a direct effect on traffic flows in cities and on construction sites. Data from travel time systems can provide key insights. By measuring traffic volume trends, traffic managers in the construction sector can instantly make smarter decisions when road closures and other activities should be allowed to reduce their impact on congestion. With this data-driven approach, construction companies can justify prolonged road and lane closures when traffic volumes are low, speeding up their construction schedule.
During the pandemic, Stinson ITS received multiple requests from several of our clients for help understanding how traffic volumes were changing in order to provide accurate reports to project managers. This led to conversations about the strategic use of data and revealed an interesting business case where traffic data is used to speed up the work schedule and complete work faster.
As we begin to move away from the continuous lockdowns and crazy traffic patterns created, we are starting to consider using this same model more generally in order to make smart decisions about traffic management in large urban work areas.
How does it work now?
I If you’re not familiar with travel time sensors, they are radio transceivers that “listen” to wireless devices such as Bluetooth and WiFi devices (think smartphones, vehicle audio systems) as they pass by. Traffic departments use them on roads and highways as a way to track unique devices as they drive on the road. Each device radio emits a unique identifier called a Media Access Control (MAC) address. This unique address is how your devices remember each other, so when you turn on your car, your phone automatically connects and starts playing music, how your headphones connect automatically when you turn them on, and more. Now, your privacy antennas might ring right now and rightly so, this information is personal and so it is important that this information is protected and anonymized. Typically, the way this is done is through hashing, which is a process of scrambling a set of numbers that you can’t untangle. However, the scrambling process is consistent so that when the same “hash key” is used on an identical MAC address, it will generate the same scrambled set of characters (allowing the device to still be “paired” from A to B, but still not being able to see the underlying MAC address itself). Hash keys are usually changed every day so that devices cannot be tracked over several days.
An important note here is that these sensors don’t track every device that passes on the road, there are simply too many things going on, and these devices only broadcast their MAC address from time to time. These sensors typically capture 15-25% of vehicles on the road. As such, these sensors can’t measure specific traffic volumes, but they can measure traffic patterns accurately (i.e. I can tell you there are 40% fewer cars today than yesterday with confidence, but I can’t say exactly how many total cars there are.
What kind of data does this allow you to get?
Now you understand how data is collected, what we’re going to look at now is what we can do with that data. The two main data sets that can be calculated from the data are:
Travel time and average speed data – that is, the travel time from sensor A to sensor B, which can also be expressed as average speed. When we place multiple sensors in a corridor or city, we can combine the data to determine how long it takes to travel long stretches of road (I wonder how they can tell you that it takes “36 minutes to reach the DVP” when you’re on the Gardiner Highway and still 50 km away. Tip: It’s not using Google Maps)
Original destination data – When devices pass through multiple sensors, it is possible to determine their approximate route through the road network. This allows you to see where drivers are coming from and congestion and where they are trying to get. (If you want to know where commuter traffic in York Region is coming from, you can place sensors around the city and ramps and see that: 30% go home on 401 East each night, while 40% head north on the 400, and another 20% leave the city on 401 West. With enough sensors, you can see exactly which main roads to take it to get there, and often the shortcuts they take to try to beat the traffic)
How does this help my work area?
This data is still limited by the number of sensors deployed, but in large municipal work areas, typically all major entry and exit roads to the construction area have sensors deployed on them. This allows a rich analysis of data on congestion conditions (travel time / average speed) and sources of congestion (original destination). Travel time and average speed are by far the most popular method of analysis for a work area, we calculate the data every minute and, as such, we can look very carefully at traffic fluctuations throughout a day. This data can be correlated with other information such as:
A. A major lane closure that occurred recently
Question: What impact has this had on traffic?
B. A reconfiguration of an intersection layout
Question: With this new layout, do cars still move smoothly through the intersection?
C. Signal retiming
Question: Does this improve congestion and traffic? At any time or only at certain times of the day?
D. Use of strategic data – When does traffic resume in the morning in this area of the work area?
Question: If I had to extend my night closure to finish at 6:30 a.m. instead of 5 a.m., what impact would I have? What happened yesterday when I tested this?
For some, this analysis may seem too complicated, while others might assume that, of course, this analysis is already carried out in all work areas! However, Stinson ITS has been in the smart work zone industry for over 10 years and I can confidently tell you that this analysis is not done at all on a large scale.
Using traffic data to make informed decisions is a concept that has been talked about for a long time, but is generally not followed. Work area policies are quite rigid and usually based on what the project owner wrote in their specifications when the project was first designed. With Ontario launching several jobs in the LLP construction sector, lasting 5 to 10 years over the next few years, figing out how to inject innovation and technology into it will be a significant challenge. To learn more about how we help our customers use data in new ways, contact us using the links below.
You can also join us for our webinar next month which will focus on the Stinson AIS Bluetooth and WiFi sensing sensor. A travel time sensor designed and built by Stinson ITS for over 6 years now. We will highlight the different ways in which the device is used and focus on a recent major project we deployed in Montreal. Register Here