RE: [CTPP] statewide TAZs - ctpp-news

30 Apr 1999

First, census tracts (and block groups) are not designed with traffic
forecasting in mind.  Tract boundaries do not reflect the access to the
transportation network that you are trying to model.  Depending on the level
of detail that you need, they may "straddle" several transportation
facilities.  If you want to save a lot of front-end time by using a unit of
geography already defined by others for other purposes, that is your choice,
but don't be surprised later on if you get bad answers at the other end of
the process.

Second, I'm curious to know on what basis you consider QRSII a "simplistic
model."  If a simplistic approach to forecasting is all that your project
requires, that's fine, but that's a self-imposed user limitation rather than
a software limitation.  (I do recognize that this model does allow for some
very elementary setups that others won't, making it the model of choice of
many agencies who put forth minimal efforts in forecasting, but it is not
limited to only these setups.)

...
  -----Original Message-----
 From:	Ed Christopher [SMTP:berwyned@mcs.com]
 Sent:	Friday, April 30, 1999 7:53 AM
 To:	Jack Pascoli; ctpp maillist
 Subject:	Re: [CTPP] statewide TAZs

 Jack, first as you can see i posted your note to the listserve. this is
 exactly the type of information we need to hear and i think other would
 benefit. as to the specific of the zone sizes if it were a perfect world
 and
 we were doing typical travel demand forcasting we would want to look at
 trips procuded and trips attracted and use those numbers to ease into
 zones
 that are all realatively ballanced on a travel basis.  i suppose this is
 one
 reason why folks gravitate to census tracts. in theory they are balanced
 atleast on the residents side. other issues play into the zone creation as
 well like geography and network configuations.

 it sounds like counties are certainly a geography that is too rough for
 you
 but something smaller would work. given the context of what the ctpp is
 (resident location data, work location data, and data on the flows between
 zones) and where you might speculate the state is heading, could tracts be
 a
 suitable geography or would you prefer something smaller?  in one hand
 sounds like you have the population base and travel demands to develop you
 own TAZs in the non-MPO counties.  on the other hand tracts are more off
 the
 shelf and a decent geography to use as a starting point. ultimately, you
 will be making the call.

 Jack Pascoli wrote:

  Ed,
 In West Virginia, we have several county studies that are not in MPO
 areas.  We use essentially the same procedures that an MPO would use.  If
  we had the information from the Census that we
use as our independent
 variables on a TAZ basis it would save us a lot of work in trying to
 manipulate the block group data into TAZs.  In addition to
 that, we have many projects that will have a profound effect on our
 smaller communities that we would like to use  simplistic models (QRS  II)
  to determine the traffic impacts of project
alternatives.  
  While we do not have a statewide model at this
time, I suspect that in  the
  next ten years we will.  Depending on the
requirements for the work to
 develop TAZs for CTPP, as a minimum we would expect to coordinate the  TAZs
  in the five (5) separate active county studies we
currently have.  As a
 maximum we would do the whole state outside of the
 MPO areas.  
 > I would appreciate your comments.
 > Jack Pascoli P.E.
 > 