03107nam|a2200265 i 450000500170000000800410001703500150005810000270007324500550010026000520015530000110020749000270021850001450024550200590039052019510044965000570240065000390245770000320249670000470252870000450257570000480262071000490266881000590271785600650277620260817162354.0010299s1988    th           0000|0 eng d  a.b100813181 aItorralba, Emmanuel C.10aAnalysis of individual travel behaviour in Bangkok  aBangkok :bAsian Institute of Technology,c1988  a127 p.1 aThesis ;vno. GT-87-36  a A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Engineering and Technology  aThesis (M. Eng.) - Asian Institute of Technology, 1988  aIn modelling the choice of mode or destination, it . is  very important to understand the characteristics of the trip  makers as well as the levels of service of various competing  alternatives. Accordingly, the importance of estimating a  generalized travel distance function cannot be ignored.  Various estimates have been proposed but certain flaws in  the model structure have been found. The analysis of modal  choice has therefore given considerable problems in  modelling.  This study attempts to estimate a generalized distance  function using the Method of Maximum Likelihood. The work  and shopping trip models are of the Multinomial Legit type .  A descriptive analysis of trip maker characteristics is  also considered as an integral part of the study. The sample  data of 560 observations was taken from a survey conducted  in five destination points within the Greater Bangkok Area.  Estimation results for both work and shopping trip models  yielded rather satisfactory results with about five level of  service and socio-economic variables considered. The results  give an insight on how individual trip makers weigh the  different variables, which agree with a lot of other  existing studies on mode choice . Comfort, safety, and  reliability may indeed be important factors affecting the  use of taxi, but they cannot easily be quantified and are  more open to argument about differential perception than are  the other factors. Prediction results were quite good with  some models correctly predicting up to 92% of the actual  choice. In addition to this, bus fare elasticities were also  analyzed.  On the whole, the Logit structure satisfied its  structural diagnosis conditions. The results can therefore  be used to represent generalized distance between any two  points. The log-sum function can also be used as an  indicator of spatial separation between two points, in a  higher level of choice, e.g. destination choice. 0aTravel time (Traffic engineering)zThailandzBangkok 0aTraffic surveyszThailandzBangkok1 aKuroda, Koji, eChairperson1 aMiyamoto, Kazuaki, eExamination Committee1 aInamura, Hajime, eExamination Committee0 aPrapon Vongvichien, eExamination Committee2 aRoyal Danish Government, eScholarship Donor2 aAsian Institute of Technology.tThesis ;vno. GT-87-36  3Full-Textuhttp://203.159.5.9/ait-thesis/detail.php?q=B18691