Guest Post by Bryant Ficek, PE, PTOE, Vice-President at Spack Consulting
Earlier this year, I detailed how our standard process for a Traffic Impact Study has several points of assumptions at best or guesses at worst. This post continues that discussion. The previous post, “Top 6 Ways to Pick Apart a Traffic Study” is available here. Other posts will follow on this subject.
Existing traffic data is the basis of our modern Traffic Impact Study. Almost all studies will start with collecting data and then building upon it. At the start of my career, we had to sit at an intersection using a count board or a simple pad of paper to record turning movements at an intersection. It was boring, expensive, and hard to get much more than peak period counts.
With video technology, we have moved from those peak period counts to collecting 13-hour or 24-hour turning movement counts, at least here in Minnesota. This change has allowed to us to better determine the true time of peak hour traffic, whether traffic around lunch is another peak, and accurately examine traffic signal warrants, to name a few benefits (shameless plug for our CountingCars.com systems here).
However, our Traffic Impact Studies are still built on essentially a snapshot in time. Our counts are from one day of traffic data, which could have been impacted by any number of items: an office takes an afternoon off, a restaurant is closed for renovations, the local chapter of Hell’s Angels decides to ride by the park, etc. With our studies guiding roadway improvements, a five or ten percent adjustment in the base traffic data, either high or low, can greatly influence thousands or millions of dollars in infrastructure decisions.
Having discussed and debated this point of weakness in our studies over the past couple months, we at Spack Consulting are moving toward a new system. We will collect video for a 48-hour period, counting either the 13-hour or 24-hour periods for each day (depending upon the project needs). The two counts will be added together for a raw two-day count. Then a seasonal factor, based upon the month of the count and general site characteristics, will be applied. The resultant two-day adjusted count will then be divided by two to determine the adjusted turning movement count for use in the study. Minnesota readers may be familiar with this process as it pertains to roadway tube counts as part of the MnDOT State Aid process.
If you complete the form below you can download a copy of the process used on a recent project. A view of the final adjusted counts is shown at left.
Since we already needed to collect data for one day, the increase in cost to our standard Traffic Impact Study is expected to be minimal. Leaving the cameras out on site for one more day will be a scheduling impact rather than an increase in cost. The expense will be the extra counting for one more 13-hour or 24-hour period as well as a short amount of time to add, apply the seasonal factor, and divide to obtain our adjusted turning movement counts.
Through this process, we expect to provide a Traffic Impact Study better targeted to determining needed improvements. The adjusted turning movement counts will remove the potential risk of recommending, or not recommending, mitigation based on initial base counts that happened to be too high or too low. We also get to move one part of our study from an assumption to a methodology that is harder to challenge.
I am making a copy of the Turning Movement Count Process spreadsheet available for free. Just complete the form below to download the file.
Did you miss the other installments of the Traffic Impact Study Improvements series? Here are the links to the other articles:
I agreed that traffic count is one of the weaknesses of most traffic studies. However, conducting count for more days and longer period may not necessarily improve data qualities. We have conducted peak hour turning count for three days (Tuesday thru Friday) and results are quite consistent with just one day count.
I think the important thing here is to conduct counts on days that are “normal”, and back it up with additional daily volume counts for a couple of days. This will give you a good level of confidence about data qualities.
One additional step I normally take to ensure data quality is to review counts against historic counts, if available, and check counts against the functional characteristic of the the street and land uses in the area. This will always give you clues whether or not your counts are good.
Thanks for the comment Pang. Our main goal here was to develop a process that will hopefully provide better base data for all situations, which will include some intersections that are consistent day to day, as well as having a process that is more defendable in discussions with agencies, city councils, county boards, etc.
I like the idea of checking historic counts too – just be careful of recent development changes that can make historic counts unusable.
– Bryant J. Ficek
Mike, I am not an engineer, but just a citizen trying to understand and perhaps dispute a tube study done by the city (Lakewood, WA) and the 85th percentile. I live on a residential 2 lane 25mph zone. At the end of the road is the VA Hospital and within a mile, a City park at American Lake. Here is what I was told : The 85th percentile speed was 30.4 mph. The Average Daily Traffic (ADT) was 4,097. Unfortunately, the 85th percentile speed would have to 10 mph over the posted speed limit (35 mph in this case) to enter phase one the City’s Neighborhood Traffic Control Program (NTCP). This tells me about 80 cars per day are going over 35 mph. To fast for this street and conditions. Do I have any leg to stand on and fight for speed bumps or some kind of traffic calming measures, as enforcement is non existant. They say its not in the budget.