Loading...

Uncertainty Analysis in Urban Travel Demand Estimation Using Fuzzy Theory

Seyedabrishami, Ehsan | 2011

625 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 41576 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Shafahi, Yousef
  7. Abstract:
  8. Travel demand estimation is a complicated process as travel demand is highly influenced by uncertainties. These uncertainties can be analyzed in two levels: aggregate and disaggregate. Models in aggregate level are usually influenced by uncertainties in data insufficiency in some regions of data space because of non uniform distribution of input data. Uncertainties embedded in travelers’ perceptions of some influential variables on demand estimation (such as travel time) moreover affect these models. An expert-guided algorithm to incorporate expert knowledge into adaptive network-based fuzzy inference system for compensating data insufficiency and considering uncertainties inherent in influential variables has been offered. The proposed model is applied to a real-world problem for estimating trip production, attraction, distribution and modal split in Shiraz, a large city in Iran. The results comparison with traditional models shows that proposed model increases performance of estimation models in terms of learning and generalization. Uncertainties in disaggregate models rooted in travelers’ perceptions of some influential variables and random behavior of travelers. A joint model for traveler destination and mode choice has been offered as a disaggregate model. The model uses fuzzy set and probability theory for considering uncertainties embedded in travelers’ perceptions and travelers’ random behaviors. The model is structured as a decision tree in which fuzzy and non-fuzzy classification of influential variables regarding destination selection and mode choice expand the tree. Probability theory is utilized to extract choice probabilities from the decision tree. The most influential explanatory variables among all of the variables are selected based on minimizing the fuzzy entropy value. An aggregation method is designed to provide aggregate estimates for transportation planning based on the proposed disaggregate choice model. A data set containing travelers’ information from 9177 trips in Shiraz, is used for fuzzy decision tree expansion and evaluation. When compared with actual travel demand and estimation results of traditional disaggregate models, the proposed model’s aggregate and disaggregate estimates of trip generation, distribution, and modal split indicate acceptable accuracy in terms of learning and generalization.


  9. Keywords:
  10. Travel Demand Estimation ; Uncertainty ; Disaggregate Model ; Probability Theory ; Fuzzy Theory ; Aggregate Model ; Decision Making Tree

 Digital Object List

 Bookmark

No TOC