A PSO based transportation network design optimization of the mega city Dhaka

摘要

Transportation network is a key issue for urban areas and is require to update to fulfill the growing demand modifying existing road(s) and/or constructing new road(s) considering various constraints. In general, transportation network design problem (TNDP) concerns optimal selection of several projects from various alternative proposed projects (with estimated costs) maintaining associated constraints to ensure benefit versus cost. Various approaches have been investigated to solve TNDP in last several decades. Recently, particle swarm optimization (PSO) based method is shown to outperform other methods. However, the most of the existing methods, including PSO, tested on the small scale TNDP. Therefore, the aim of this study is to identify the effectiveness of the PSO based method for a mega city heaving large number nodes and arcs. The mega city considered in this study is Dhaka, the capital of Bangladesh. The experiment has been conducted on the roughly estimated data for exiting network as well as proposed projects. Experimental results for revealed that the method able to select projects for optimal (or near optimal) utilization of a given budget amount. The selected projects are seem to be effective to increase traffic flow while observe on the schematic diagram of the road network.