|
The urban population has been increasing at a rapid pace
for various reasons. Efficient management of the transportation
system is an important aspect of urban development.A neural
network approach of the Orientation Destination (OD) distribution
is importantdata for the design and reconstruction of the
city. Modern cities are now adopting this technique for smooth
and sound urban life. Hay (1977) has shown that the traditional methods
of collecting data for transportation from all parts of
the city for OD matrix needa great financial and technological
support. They may not be precise and further affectthe growth
of the urban development.
It is easy to get the urban link volumes of the transportation
network through trafficcounts, and it is feasible to build
mathematical models based on link volumes throughwhich we
establish the OD matrix. The optimization mathematical model
reduces thisproblem more concisely. As the city is fast
expanding, and various new link zones areintroduced, a lot
of computer resources and time are required to solve the
problem. Thisproblem has been studied by various authors
including Lida and Nguyen (1978),Takayama (1986) ; Vanzuylen
and Willumsen (1980), and Zhejun (1997). Though thepioneering
work of MC by Culloch and Pitt (1943), laid a foundation
stone in the fieldof neural networks, the work of Hopfield
and Tank (1985) extended this approach tosolve many complex
problems not only in the fields of science, engineering
and medicine but also social problems.
Traffic counts have been used for the formulation of the
mathematical model becauseof their availability, low cost
and non-disruptive character. So the estimation of tripmatrices
is obtained from link volume counts and the OD matrix thus
obtained helps tosolve the mathematical optimization model.The transportation network model is based on parameters
such as bus stops,intersection of routes, etc., which are
considered as the nodes. The city is divided intomany small
zones and the OD activities of the zone residents are considered
to takeplace at the nearest node to that zone.
|
