S. L. Dhingra

Micro-simulation of residential location choice and its variation

Traditionally land use models are designed to represent aggregate behaviour even though land use development patterns are the manifestation of micro-level individual location choices. This paper develops a micro-simulation model for residential location choice in a medium-sized city in India. The model consists of two major parts: (a) the simulation of the characteristics of the individual decision units, the households of the city, and (b) the simulation of their locational choice for residences. The first submodel simulates the characteristics of households using the Monte Carlo method; the second is developed using Artificial Neural Network (ANN) theory. The integrated model simulates variations in the locational choices of the population and the residential land use distribution by aggregating the decisions of all the individual decision units. This model is developed and tested using data from the Indian city of Guwahati.

Heterogeneous Traffic Flow Modeling and Simulation Using Cellular Automata

This study presents an attempt to simulate heterogeneous traffic using the principles of cellular automata. A conventional cellular automata mode for the homogeneous traffic is modified to handle heterogeneity. The principle adopted in this study is to retain the passenger car equivalent concept of HCM and the basic structure of the CA model. The modifications are done at three levels: First, the cell size is reduced to take the concept of passenger car equivalents. Second, the randomizing rule is modified to suite heterogeneous traffic conditions. Finally, the lane changing rules are modified to handle over taking maneuvers and conditions different types of vehicles characterized by their dynamic characteristics. The models were calibrated and validated with limited data sets and the performance of the model is presented.