Mike on Traffic

What is Trip Generation?

When traffic engineers do a study for a new development, one of the first steps is forecasting how much traffic will come in and out of the development (jargon alert:  this is termed trip generation).  This forecasting is typically done by using data from the Institute of Transportation Engineers' informational report, Trip Generation.  This is the largest database of measured trip generation by land use type, with the current edition (7th) containing more than 4,250 trip generation studies.  The data has been collected for several decades throughout the United States.  Some land uses have a very robust data set, such as Single Family Detached Housing which has data from 350 studies.  Other land uses have a minimal data set and should be used with extreme caution, such as a Synagogue which has data from 2 studies.

Although Trip Generation is the best database available, it is a tool that requires interpretation.  A good example is the land use Pharmacy/Drugstore with Drive-Through Window.  It contains data largely from stores with one drive through lane.  There is data in the set from only one pharmacy that had two drive through lanes, which is a concern since nearly all pharmacies built today have two drive-through lanes.  I performed a trip generation study for pharmacies with two drive through lanes in 2004 and found the trips generated by a pharmacy with two drive through lanes were nearly twice the trips calculated by using the average trip rate in Trip Generation (which was largely based on pharmacies with the single drive through lane).  I submitted this data to the Institute of Transportation Engineers for their use in the upcoming 8th Edition of the Trip Generation report.   

Traffic engineers must spend time reviewing the trip generation data sets at the beginning of any traffic study to make sure their forecasting is technically sound.   As with any engineering study, "engineering judgment" is involved when interpreting the data set.

Telecommuter vs. Nomad: The New Office

The current issue of The Economist has a great special report on Nomads - people who can work anywhere (park, coffee shop, cubicle, etc.) because of wi-fi, blackberries, and laptops.   

One of the articles in the report is titled The New Oases (Nomadism changes buildings, cities, and traffic).  I previously blogged about Travel Demand Management, with "telecommuting" being one of the Travel Demand Management strategies.  This is an outdated concept.  It basically has people working 9-5 out of a home office with a pc, fax, and land line.  Being a Nomad has supplanted this concept.  Now - most of us can jump on any computer, log into our work network, and dive into our files.  Blackberries, iPhones, etc. have made communication amazingly easy (so easy we have addicts to their "Crackberries"). 

I can see several phd theses coming out of this trend:

• Urban Planning - Mixed use, flexible space makes more sense than segregated land zoning.  Also, how many more giant office complexes are going to be built when most people don't need a cube anymore?
• Traffic Engineering - Diving into the travel data to identify shifting travel patterns.  Are the rush hours becoming less of a peak?  Are we driving more during the day?  I try to schedule my meetings in the middle of the day so I don't hit rush hour (I am the only person I know who walks to work, but that is another post).  Maybe we move away form the giant roads/intersections in the suburbs and go back to the smaller roadways of the inner cities.  We need more complicated signal systems because traffic flow isn't just driving in during the morning and out in the evening.
• Management - Shifting managers to think like college professors instead of factory line bosses.  In college we were taught an hour or two a week on a subject, we did some self study, and then wrote a paper/did a design/took a test.  Our professors had office hours in case we needed extra help.  They didn't tell us we had to write our paper from 8 am to 10 am at cubicle B.  They told us we had to turn in a paper at 4 pm next Thursday. 

This is definitely a trend to think about.

Miovision Video Camera Unit Pictures

We put out our two video camera units yesterday and programmed them to start recording this morning at 6:00 a.m.  They are at two busy intersections with no intersections or driveways between them.  We'll be able to make sure the counted cars going between the two intersections add up over the twelve hour study.  We are also counting cars the old fashioned way (manually) at the intersections from 7:00 to 9:00 a.m. and 4:00 to 6:00 p.m. to give us a baseline.  Bonus - the weather is bad (low of about 30 degrees Fahrenheit, High of about 40 degrees Fahrenheit, it started raining at 2:00 p.m., wind of about 10 mph with gusts up to 25 mph, rain likely transitioning from rain to sleet to snow over the course of the afternoon). 

As promised, here are a few pictures from today.  I was happy to find out the units are a lot more inconspicuous at an intersection than in my driveway. 

Miovision is working on a bracketing system so we could remove the tripod legs and strap the unit to a pole.  That will be a much better system for urban areas. 

Counting Cars with Video Cameras (Miovision)

Doing turning movement counts automatically with computers is the holy grail of the traffic counting industry.  Right now we deploy people to sit at the intersection and count cars manually using computerized tally machines.  Miovision appears to have found the holy grail using machine vision.  And this isn't the machine vision your grandaddy used (i.e. Autoscope and Vantage).  Older systems detect cars as they go over specific points on the computer screen.  Miovision locks onto anything that moves on the screen and tracks its path.   I  believe this is going to allow us to collect 12 hours of turning movement data cheaper than what some firms charge for 2 hours.  Aside from slashing costs, two major bonuses:  (1) it can do turning movements at roundabouts (which is nearly impossible to do in the field) and (2) we'll be able to collect long term turning movement counts (it appears we can do 48 hours now and soon we'll probably be able to do a week).

Corry from Miovision came out to my office today and helped me get up to speed with the video system.  I am testing out two units to make sure they count as accurately as the company claims.  So far, they have followed through on all of their promises.  The unit only takes about 15 minutes.  We have started an extensive training regimen to reduce that time (Corry promises a lumberjack style competition to crown the setup champion).  I'll be reporting back in a few weeks on my testing results.  I'll also post some pictures of the system out in the field.  I forgot my camera today.  Stay tuned....

Dots on Roads - MnDOT's Stop Tailgating Program

My friend Jenni asked me about the dots painted on Highway 55 outside of Buffalo, MN.  They were painted as part of a pilot program in 2006 in an effort to reduce tailgating.   The 2.5 feet x 7.5 feet ovals were painted 225 feet apart on Highway 55 between Rockford and Buffalo in Minnesota.  Signs along the corridor encourage motorists to leave 2 dots between them and the vehicle in front of them.  Jenni said she definitely left more distance between her and the next car while she was in the zone with the dots, but she want back to her normal driving pattern outside of the zone. 

MnDOT's report on the program states the average following distance increased by 17.9 feet when the dots were present vs. before the dots were painted.  The report doesn't give any before/after crash statistic comparisons, but the program was justified on the basis that it may help reduce rear end crashes (26.36% of all crashes in MN between 2002 and 2006 were rear end crashes).   To my knowledge, no other roads in Minnesota have the dots. 

Two closing shots -

1.  Pennsylvania tested a similar program on one of their rural highways.  I vaguely remember reading an article that attributed a traffic jam in bad weather to the dots.  Motorists stayed 200+ feet apart even though they were traveling much below the posted speed limit.  Any links on the article would be appreciated.

2.  Although the engineers with the project didn't appreciate this artistic impression, I am a Pacman fan...