Evaluation of Speed Display Boards in School Zones in London Ontario

The functioning of speed display boards (SDBs) has been documented at two school zone sites in east London, Ontario.On September 3 and 6, 2019 observations were made on Tweedsmuir Ave in front of St Bernadette’s Catholic Elementary School. On May 17 and 19, 2011 and on April 30, 2020 similar observations were made on Brydges Street near Prince Charles Public Elementary School. Both sites had their posted maximum speeds reduced from 50 to 40 km/h in recent years. The figure below shows the locations of the two sites on a map of east London, Ontario.

The methodology involved the set-up of multiple, synchronized, video cameras at equidistant intervals along the road on the approach to the SDB.  As vehicles passed these cameras the time interval of their passing through the known distance allowed for the calculation of average speeds within each road segment. Meanwhile additional cameras directed along the road also allowed for observations of the front and rear of each vehicle as it approached the SDB. A camera was also dedicated to a view of the display from the SDB. By this methodology the positions and speed of vehicles approaching the SDB and the display from the SDB could also be accurately documented.

The reason why SDBs are installed within the City of London can be obtained from the City’s explanation on the official City of London website. This discussion is reproduced below.

Speed and Display Boards

To remind drivers of their speed in residential areas, primarily around elementary schools and parks, the City of London continues to install speed and display boards on streets across the city.

Speed display boards are pole-mounted devices equipped with radar speed detectors and an LED display. The boards are capable of detecting the approaching speed of a vehicle and displaying it back to the driver.

When combined with a regulatory speed limit sign, a clear message is sent to the driver that they may be travelling too fast. The objective of the program is to improve road safety by making drivers aware of their speed.

As of April 1, 2019, four speed and display boards have been installed and will be rotated throughout London as part of the Public Education & Empathy Program (PEEP) program. Locations have been identified through the Active and Safe Routes to School program as well as feedback received from local residents.

Part of this Gorski Consulting analysis is to determine whether the noted objectives were being achieved.

The following sections of this article will provide descriptions of each site followed by the results of the analysis of the functioning of the SDB.

1. Tweedsmuir Ave at St. Bernadette’s Catholic Elementary School

The figure below shows an overall view of the Tweedsmuir Ave site. The green zone is a park located just to the west of St Bernadette’s Catholic Elementary School. The red circle shows the location of the SDB on the south side of Tweedsmuir Ave just east of the school.

The figure below is a closer view of the site and the red circle which shows the location of the SDB.

The close-up view below shows the location of the SDB depicted by the red circle. A distance of 125 metres is shown measured westward from the SDB. This is the zone within which video cameras were placed, spaced out at 25 metre intervals.

The figure below shows an example of the SDB registering a speed of 54 km/h as a blue eastbound vehicle is approaching.

The figure below shows a view looking east along the south side of Tweedsmuir. In this view at least two tripods and cameras can be seen along the edge of the park vegetation. The SDB is not visible but it is located on the south side of the road in the  background.

The figure below shows a camera in the foreground, positioned on the lawn of St Bernadette’s School and pointing at the “Zero” marker which is at the position of the SDB. Another camera in the background is positioned in the driveway of the school and it is pointing at the “25 Metre” marker. Additional cameras are located in the background but are not visible.

The figure below shows a view looking west toward the camera in the school driveway at the “25 Metre” marker. Other cameras are located in the background along the brush of the park.

The figure below is a view looking east, showing another camera that is partially screened by a small sapling in the foreground. This camera is pointing toward the SDB in the background and it was able to document whether driver’s applied their brakes as they approached the SDB. Another camera positioned near the SDB was pointing westward and this allowed a view of eastbound vehicles as they approached the SDB. Such views are important to observe if vehicles leave the roadway by making turns into a private residence or road; in that case they need to be removed from the analysis.

The views shown above provide an example of the typical set-up that is used when documenting vehicle motions and speeds.

2. Brydges St Near Prince Charles Public Elementary School

The videotaping at this site on April 30, 2020 was previously discussed in an article posted to the Gorski Consulting website on May 9, 2020 entitled ” Latest Site Where Covid-19 Effect Being Evaluated: Brydges Street in London, Ontario“.

The Brydges Street site is a busy collector road located on the east side of the City. The City lists its traffic volume (AADT) as 9,500 vehicles west of Kiwanis Park and 10,500 vehicles east of the Park. Just east of the testing area Brydges Street takes on a new name, Wavell Street.

The image below is a view of the Bridges Street site with the green zone of Kiwanis Park. Although the roadway runs from northeast to southwest, for ease of discussion we will refer to it running east to west. The curves in the road are located to the east and west of Kiwanis Park. The two elementary schools are located along the south side of Brydges Street east of Kiwanis Park.

The image below is from Googlemaps and shows the same view as the image above. There are bike trails running through Kiwanis Park and riders must cross Brydges Street in the vicinity where the videotaping was conducted.

A closer view of the west end of the site is shown in the image below. During the April 30, 2020 session a video camera was placed at the bus stop (depicted by the blue square) on the south side of Brydges just east of Cornish Street. This camera pointed eastward and documented details such as the activation of brake lights on eastbound vehicles. A second camera was positioned near the same location on the north side of Brydges Street, pointing southward at a traffic cone which was the “100 metre marker” for our testing. This location can be seen near the lower centre of the image below.

Additional cameras were placed at 50 and 75-metre intervals along the south side of the road, eastward, up to a location at the very top of the image above. This monitoring was over a total distance of 250 metres. The SDB was positioned 100 metres east of the cameras at Cornish Street and 150 metres west of the last camera to the east.

The image below shows a view looking west at the video camera placed at the bus stop near Cornish Street. Eastbound vehicles would be travelling toward the camera. The testing area is located behind this camera.

The image below is a view looking east toward the Speed Display Board (SDB) which is mounted on the closest utility pole on the south side of Brydges. A video camera was placed at the utility pole facing westward so that eastbound vehicles could be observed approaching the SDB. Parked vehicles on the right are at the Kiwanis Park and further in the background is a small bridge over Pottersburg Creek.

The two images below show the location of one of the video cameras positioned at 150 metres east of the SDB.

In the above image a cone can be seen near the road edge. This was the “150-metre-east” marker.

Results at Tweedsmuir Site

The table below shows the results from the videotaping session on September 3, 2019. Cameras were set up at 50, 75 and 100 metres west of the SDB. Thus average speeds were calculated within two distance segments. Only 18 minutes of videotape was analysed and 32 observations of eastbound vehicles were documented. The table below shows all the eastbound vehicles including those that made turns off the roadway within the testing area.

The word “Conflict” along the right edge of the table refers to those instances where another vehicle was present within the detection zone of the SDB and thus the potential existed that the SDB might display the speed of that other vehicle rather than the one of interest.

As can be seen in the above table only one vehicle (Silver car in Observation #9) actually travelled consistently below the speed limit of 40 km/h.

There appeared to be a discrepancy in Observation #13 (Dark Minivan) where the average speeds from the video indicated 61.22 and 58.82 yet the SDB displayed speeds of 52, 52 and 50 km/h. Review of the video confirmed that the average speeds were correct. However it was discovered that the SDB kept reporting the speed of the previous vehicle (Observation #12, Dark Blue Pkp) even though the Dark Minivan was passing through the SDB detection zone. The correct speed of the Dark Minivan did not begin to be reported until just after the Dark Minivan passed the 50-metre-marker. So this is an example of a “Conflict” that was not detected in the original analysis.

As a curiosity, we filled in the SDB row of speeds for Observation 13 will a speed of 58 km/h in all three columns and then observed that the Average for the SDB values at the bottom of the table rose to “47.19, 46.53 47.14”. So the change made a difference in the Average by about 0.5 km/h. This brings the values more in line with the Average at the video analysis.

However, overall, the speeds calculated from the distance intervals were quite close to the instantaneous speeds displayed by the SDB. Certainly we should not expect them to be identical.

The obvious problem in the SDB display is that, even with relatively low traffic volumes, the SDB could display confused speeds whenever two or more vehicles were within its detection range. This is not helpful if the intention is to provide a driver with a true indication of their speed.

The investigation at the Tweedsmuir site continued on September 6, 2019 where more cameras were installed as shown in the previous figures. Videotaping occurred over 40 minutes and 68 observations of eastbound vehicles were documented. Observations were removed from  the analysis if the vehicles did not travel straight along the road but made turns or came to stop. Further observations were removed if there was interference to their motion from vehicles ahead.

In one instance a vehicle came to a stop in the “No Parking” zone in front of St Bernadette’s elementary school and the left rear door (the door facing the road) was left open for several minutes as passengers were shuttled in/out of the vehicle. In this instance 6 observations of eastbound vehicles became spoiled because their motion was interfered with by the stopped vehicle.

In the end this left 30 valid observations. The average speed of these vehicles, as they travelled through the four distance segments between 25 and 125 metres west of the SDB, are shown in the table below.

Although the average speeds seemed to be comparable to the September 3, 2019 data, 6 of the 30 vehicles were observed to be travelling above 60 km/h.

The average speeds were then compared to the instantaneous speeds that were displayed on by the SDB. For these SDB values the speed was documented when the front end of each vehicle reached the roadside marker at each camera. This is different than the previous speed calculation which was developed from an average over a full 25-metre distance. The SDB speed data is shown in the table below.

There is a discrepancy in the speed data for Observation #49 (White Car). The average speed calculated between 75 and 50 metres west of the SDB was 60.40 km/h. Yet the speeds displayed on the SDB were 54 km/h at the 75-metre marker and 55 km/h at the 50-metre marker. To be certain there was no calculation error the video was re-examined and the position of the vehicle was observed as it passed through each marker. This  confirmed that there was no calculation error and that the speed of 60.40 km/h was accurate.

In evaluating where the problem might lie, it was observed that as the SDB was transiting from displaying 54 to 55 km/h it stopped its display for a brief period and a set of lines began to flash at the bottom of the screen while no speed was displayed. After this short interval the next speed displayed was 56 km/h and then it reverted back to displaying 55 km/h. Such functioning has been observed in other instances when vehicle speeds are approximately 55 km/h. While it cannot be known for certain, it is possible that the 55 km/h value is a pre-set threshold such that when it is reached the SDB enters into a different mode, perhaps to store the value in some sort of storage media. During this delay the SDB fails to display any speed readings. Thus it may have failed to display the full extent of the short and quick acceleration of the subject vehicle. The average speed was reduced to 51.43 km/h in the distance between the 50 and 25-metre markers.

Results at Brydges Site

On April 30, 2020 132 eastbound vehicles were documented between 16040 and 1740 hours on Brydges Street between Cornish Street and Kiwanis Park.

The location of the SDB was selected as “Zero” and cameras and markers were placed at the following locations with respect to the SDB”

• At 100 metres west
• At 50 metres west
• At Zero
• At 75 metres east
• At 150 metres east

From this arrangement four distance intervals could be created allowing for average speeds to be calculated within each interval. The average speeds obtained within these intervals are shown in the table below.

We wanted to know what the effect was of eastbound drivers being prevented from travelling at their desired speed because they were forced to travel behind slower vehicles. Thus we reviewed the 132 observations and selected 34 instances (“Single Vehicles Only”) where it was obvious that they were the only eastbound vehicles within the detection range of the SDB. This usually involved gaps of at least 10 seconds between vehicles ahead and behind the subject vehicle. Thus the average speed of those “Single Vehicles” is shown in the above table along with the full 132 vehicles

The 34 observations of Single Vehicles were reduced to a smaller set when it was required that a full display of correct speeds had to be available from the SDB. In 13 of the 34 observations such data was not available from the SDB. In a majority of instances this was because the SDB continued to display the speed from a vehicle ahead, even though that vehicle had travelled past the SDB. Thus the speed readings being displayed were not related to the subject vehicle of interest.

In one instance (Observation #77: Silver Chevrolet Car) the SDB never activated through the whole distance that the vehicle travelled through the detection zone. The average speeds calculated within the four road segments indicated the following average speeds for this vehicle: 100m to 50m West = 59.41 km/h, 50m West to Zero = 60.61 km/h, Zero to 75m East = 57.82 km/h, 75m to 150m East = 51.92 km/h.

In Observation #17 a Silver Honda Civic entered the detection zone but no speed was displayed until the vehicle had travelled almost 4 seconds past the 100-Metre West marker. Then as the vehicle was well past the 50-Metre marker the SDB displayed its first speed of 55 km/h. It then shut off 2.3 seconds later as it started to reach the SDB’s anchorage location.

In Observation #108 the SDB flashed a speed of 54 km/h about 1.5 seconds before a Grey Hyundai SUV reached the 100-Metre West marker. It then flashed a speed of 55 km/h just ast the Hyundai reached the marker. After flashing the speed of 55 km/h for another half second the SDB suddenly stopped its display until 1.5 seconds before reaching the location of the SDB at which time it began to display the speed of 55 km/h, reducing to 53 km/h on its final display when it approached the SDB location.

The table below shows the remaining 21 observations where the SDB provided sufficient data to allow a comparison with the average speeds calculated from the four distance segments.

The above table shows that, on average, the SDB started to display speeds about 0.45 seconds before the vehicle reached the 100-Metre West marker. But there are some large discrepancies. For example in Observation #14 the Dodge Ram Pick-up was not detected until it had travelled 3.67 seconds past the 100-Metre marker. And in Observation #17 the Honda SUV travelled almost four seconds past the 100-Metre marker before the SDB began to display speed values.

The table below shows the average speeds calculated over the four distance segments for the same 21 Single Vehicles shown above. Although the number of observations is small it shows a slight and gradual reduction in speed, not only during the approach to the SDB but also after the vehicles travelled past it.

There is some agreement between the speeds displayed by the SDB and the calculated average speeds. However one cannot expect that these values should be identical. The instantaneous speed displayed by the SDB cannot be exactly the same as the average speed over a distance of 50 or 75 metres unless in an isolated instance by pure coincidence.

Summary

Overall, the functioning of the SDBs at the two sites has indicated that some problems exist. At times the SDB would continue displaying speeds related to a vehicle that had already vacated the detection zone, even when another vehicle was already passing through the detection zone. In other instances the SDB would fail to display a speed even through a vehicle was travelling for several seconds within the detection zone. In one instance the SDB did not display a speed for the full distance that a vehicle travelled through the detection zone. And finally, the SDB sometimes became confused when there were two or more vehicles in the detection zone; sometimes displaying the speed of the vehicle ahead, sometimes displaying the speed of the vehicle behind.