Hi Alan,

I worked in transportation and these are some of the serious
methodological drawbacks. The generation-attraction models always
predict more traffic (cities usually tend to grow), giving the
engineering a pseudo-scientific justification for this sort of
intervention in urban space (wider and faster roads, less pedestrian
friendly and so on).

Of course Transportation is not a science, but a field of enquiry - or
an art as some classic authors, such as Potts and Oliver (1972)
acknowledge. (I am not sure how authors see these models in Geography
- they are pretty much the same).

However, this is all very old-fashioned as you say. I am not sure how
Transportation engineers work in these times of more pedestrian and
environmentally friendly policies. Traffic reduction (congestion
charges, for instance) are pretty much a negation of the old models.
Pedestrianised areas are more down to urban design. What are they
doing? What kind of models they use now? Do this people innovate or
will they stick to these old models forever?

I guess they are using the same models to plan public modes of
transportation, which are less destructive than cars (even though
areas of a city can also be destroyed by the excess of buses).

Regards,
Lucas

2008/9/25 Alan Penn <[log in to unmask]>:
> I have often noticed how in London the effects of traffic management are
> often to cut the connections between major primary routes and the rest of
> the street grid. An obvious case is the raised sections of the Marylebone
> Road, however even on the ground level sections the cross routes are
> generally disconnected, either physically or by one way and turning
> restrictions. The consequence is that these routes are optimised for fast
> through traffic, but when an accident or a breakdown does happen you are
> completely stuck with no way of getting out of the resulting jam.  As a
> cynic I have always suspected that this is because when these routes were
> 'traffic planned' in the 60's and 70's the engineer's models could only cope
> with a fairly limited number of nodes and links. It was only really feasible
> to run the model if you eliminated all the routes through the surrounding
> urban area. Then of course you have to build the model... the result, a less
> resilient system.
>
> Alan Penn
>
>
>
>
> On 25 Sep 2008, at 03:58, Hoon Park wrote:
>
>> Dear all Some of you may be interested in the following paper: 'Price of
>> anarchy in transportation network: efficiency and optimality control'
>> (http://link.aps.org/abstract/PRL/v101/e128701) by H Youn, M T Gastner, H
>> Jeong, in Physical Review Letters. The paper was also introduced in the
>> recent issue of Economist (13 Sep). The authors demonstrate, based on an
>> analysis of travel times or delay in road networks of Boston, London,
>> Newyork, that the rational decisions made by uncoordinated individuals to
>> choose the quickest paths do not always achieve the social optimum. A proper
>> method to reduce such social inefficiency, they suggest, is to modify the
>> underlying road network structure. Interestingly, according to their
>> simulation, it is often the most carefully installed streets to reduce
>> travel times that are responsible for the worsening of traffic conditions;
>> so it may be better to close them and encourage detours for the benefit of
>> the society as a whole! Any comments? Best wishes Hoon Park
>



-- 
Lucas Figueiredo