Improving the Effectiveness of Smart Work Zone Technologies, Part 1

illinoisstudyA brilliant study was published in November by the Illinois Center for Transportation Studies entitled “Improving the Effectiveness of Smart Work Zone Technologies.” The principal researchers were Yanning Li, Juan Carlos Martinez Mori and Daniel Work. Download it here.

It is brilliant for a couple of reasons. It moves past the studies we conduct over and over again that look at the effectiveness of smart work zone systems. Those have been done, and there really isn’t much more to be learned from additional studies. We know they work. It has been proven. Enough said.

Instead, this study looks at ways of making something good, even better. There is so much new information is this study that we will discuss it in two separate blog posts. This first one will look at their conclusions regarding sensor types, sensor spacing, and missing data. The second will consider travel time estimation and, in particular, their recommendation of the Kalmar filter algorithm.

Let’s begin with sensor types. The study focuses primarily on doplar and side-fire (RTMS) sensors.  They compared the advantages and disadvantages of each and concluded that for most work zones systems there is no advantage to RTMS type sensors. The RTMS provide more accurate flow measurements than radar due to fewer occlusion issues. But they do cost more and are more difficult and time-consuming to set-up and configure. They wrote the, “analysis indicates the types of sensors do not have significant influence of the performance of existing smart work zone systems. It is suggested the choice of sensor types should depend on the specific requirements and constraints in each work zone.”

They went on to say that given the extra cost of RTMS, you are normally better off spending that money on additional doplar sensors, as more sensors improve the data quality far more than the sensor type. Of course, if lane by lane counts and classifications are needed to meet the deployment goals, RTMS is the only practical solution.

Next they discussed sensor spacing. Most of us in the work zone ITS industry suggest sensor spacing of between 1 mile and one-half mile apart. This study confirms what we learned through experience. They said the closer you place your sensors to each other, the better the data quality. “When more sensors are deployed, the systems provide faster notification of changes of the traffic conditions and increase the estimation accuracy of the traffic conditions.”

However, anything closer than a half mile apart provides negligible additional benefits. “When the sensor spacing is smaller than 0.5 mile, the benefit of additional sensors … is marginal.”

A third topic they covered was missing data records. This has not been covered in any detailed way in previous studies. But it is important. They studied only two projects but for those projects found they were missing as much as 10% of the data records. They did point out that the system messages continued to be sent to the message signs. But what if the interruption was sufficient to delay those messages?

One vendor suggested the data records were missed due to the cellular carrier dropping them in favor of voice transmissions. If true, this underlines the need for redundant communications. Satellite backup or better cell service is a must. If data is dropped for more than a few seconds it could affect the timeliness of warnings to traffic upstream. And if that data will also be used for work zone performance measurement, it causes additional problems. A measure of system data transmission performance should be included in the evaluation of every work zone ITS deployment.

So, in short, simple systems with more sensors are better than more complicated ones.

Spacing of a half to one mile is best. Anything greater quickly loses accuracy and anything less is not cost-effective.

And agencies should require a data transmission reliability report to be sure that most of it is getting through, even during major incidents.

In a future post we will return to this study to examine algorithms, especially for travel time systems. In the meantime, download this important study!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s