عنوان مقاله [English]
The planning of bus transportation system is one of the most important parts in planning of public transportation systems. This planning type is difficult because of some inconsistent goals. For example we cannot expect paths that are planned for minimum travel time, maximum covering and access. At this field, we must consider many criterions for reaching to best answer. In this paper, a new model has been represented for planning bus transportation system that used GIS capabilities. This model is based on travel demand information between urban blocks. To prediction of travel demand between urban blocks, a method has represented in section 4. This prediction is divided into two parts: trip production and attraction. Prediction at each part is done based on different travel purposes such as working, study and etc. For each purpose, a separate model is developed to forecast the production and attraction trips of urban blocks. These models have been represented in section 4.1.
After prediction of travel demand for each urban block, it is time to compute the distribution of travel demands between blocks. This work is done by a gravity model. Various gravity models have ever introduced and used. In this research we use a gravity model that is based on the distance between urban blocks. For each travel purposes, a distinct gravity model must be developed and used. These models are explained in section 4.2.
In this article, we have represented a model for planning of bus transportation paths which has some major goals, such as maximum people transferring and minimum travel time. Reaching to these goals needed a model that uses the predicted travel demand, travel distribution between blocks and travel time between them. For using of all information in model simultaneously, we have used weighted graph theory. Weighted graph is a graph that its all edges have a distinct weight. Weights of edges in GIS can be based on various parameters, such as time, distance and etc. In this research we give a weight to each graph edge, based on its travel time and the travel demands of its surrounding blocks. The way for forecasting travel demand is explained in previous paragraph, and travel time for each edge is computed based on its traffic rate, length and etc. The way to produce this weighted graph has been explained in section 5.
After producing the weighted graph, it is time to extract the final paths for bus. For this purpose, the produced weighted graph has a major role. Also distributed travel demands between blocks are used in this step. The best paths are extracted by path finding between blocks. Path finding is done based on produced weighted graph, using of network capabilities in ArcGIS. This step is repeated for all blocks that have a large number of demands between themselves. Finally, a bus network is extracted, which covers some other goals that have been explained in section 3.
In section 6, theses models are applied with a small part of region 10 of Tehran city as a case study. At this case study, all steps explained in this paper have been represented as a diagram. All stages of this diagram are implemented for this case study.
Finally some conclusions and recommendations have been stated in section 7.