I hope everyone had a Merry Christmas and a Happy New Year. Mine was quite enjoyable.
Prior to the holidays, I was checking e-mail and on the front page was a headline about congestion reduction. When I clicked on it, I initially dismissed the report, but have thought about it since so what happens when I over-think an urban issue? I blog it o' course.
The basis of the article is that if select neighborhoods decided against making car trips, then overall congestion would be reduced dramatically. From the lead:
Canceling some car trips from just a few strategically located neighborhoods could drastically reduce gridlock and traffic jams in cities, a new study suggests.
They go on to list individual cities' neighborhoods and their effects on overall city congestion:
"In the Boston area, we found that canceling 1 percent of trips by
select drivers in the Massachusetts municipalities of Everett,
Marlborough, Lawrence, Lowell and Waltham would cut all drivers’
additional commuting time caused by traffic congestion by 18 percent,"
said researcher Marta González, a complex-systems scientist at the
Massachusetts Institute of Technology. "In the San Francisco
area, canceling trips by drivers from Dublin, Hayward, San Jose, San
Rafael and parts of San Ramon would cut 14 percent from the travel time
of other drivers."
The problem with this, as is any hard science when they delve into the soft or social science, is that everything is a mathematical formula, a linear cause and effect. Water boils at a certain temperature, oxygen is composed of two oxygen atoms, the day is 23 hours, 56 minutes long, etc. are all measurable, repeatable things. One person can do it, report it and pass it on to someone who can independently do the same thing.
In social sciences, that is impossible. Human behavior, like nature itself, is chaotic, erratic and unmeasurable. That is where statistics come in. Human behavior can be measured in a bell curve, but it is not precise. Traffic modeling is an example of this measuring of human behavior. There are several different types of mathematical models out there for MPO's to use, but most follow a basic formula, trip generation, trip distribution, mode split and trip assignment.
However, there is a step called calibration. Ever seen a meter strip in the street? That helps calibrate the model. Using the formula, modelers see where the model says trips are coming from and where they are going. They then compare that to real world measurements and repeatedly calibrate the model until it resembles the observed numbers. This model is then forecast out for however long the modeler chooses too extend it, usually 30 years.
The reason calibration is needed is human behavior. New York is different than Dallas for example. The exact same mathematical formula would be highly erroneous.
It, like the study linked above, ignores the human behavior and attempts to quantify it scientifically. I am an alternative transportation advocate. There have been times where I decided to use different modes at different times for the exact same trip. Add in things like Induced Traffic and the flaws become more noticeable.
Individually, there is no way to model behavior. But, using the bell curve, there are ways to measure, but it isn't exact, and that's okay. What isn't okay when folks portray human behavior as a linear function. I don't for a minute, believe that study. Procedurally, it may be sound, the math exact and correct. But it ignores the fact that social sciences aren't neat, compact and unerringly precise. Spouting off precise numbers such as 18 or 14 percent reductions ignores human behavior, and that may cause more problems in a different way.