| Author | Phan Trung Lam |
| Call Number | AIT Thesis no. IE-91-30 |
| Subject(s) | Transportation--Thailand--Bangkok--Mathematical models
|
| Note | A thesis submitted in partial fulfillment of the requirement
for the degree of Master of Science, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Abstract | The conventional Urban Transportation Planning (UTP) model is an essential and integrated
model and has been useq in many transportation studies to estimate the origin - destination (0-D)
trip matrix. The gravity distribution model is easy to apply and petforms well in many cases.
Unfortunately, the UTP process with the gravity model is found to be insensitive to the changes of
the transportation system. On the other hand, the gravity model's theoretical background is not very
clear. This study tries to introduce some improvements upon this model.
The approach includes utilization of mathematical optimization techniques to formulate and
solve the trip distribution model. Enhancement to the model's behavioral richness has been made
by introducing the activity and system accessibility measures into both the trip generation and the
trip distribution model. The mod l implementation can be divided into following steps:
Firstly, the system accessibUity measures are computed based on the transportation system's
current state. These measures are then used along with the other socioeconomic characteristics to
calibrate the trip generation model.
Second step consists of calculating the combined accessibility measures and classifying the
inter-zonal trip distributions by trip lengths and trip accessibility indices. The trip distribution is
solved as the entropy maximization problem with these constrai~ts, using the simple and efficient
Bregman's balancing algorithm. The model's error is estimated by some well-tested goodness of
fit measures.
The proposed methodology has been applied to the case study: to estimate the 0 -D trip matrix
for Greater Bangkok Area's transportation network. The model has been found to petform well for
this large size real-world problem. The results show some improvements to the previous studies
using conventional models. Although not all variance of the model could be tested due to short time
and limitation of available data, but it is found the newly involved accessibility measures can act as
the key to improve the model's system-sensitivity. |
| Year | 1991 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Other Field of Studies (No Department) |
| Academic Program/FoS | Industrial Engineering (IE) |
| Chairperson(s) | Oudheusden, Dirk L. van
|
| Examination Committee(s) | Fujiwara, Okitsugu ;Huynh, Ngoc Phien
|
| Scholarship Donor(s) | The Australian Government |
| Degree | Thesis (M.Sc.) - Asian Institute of Technology, 1991 |