Optimal Planning of Transit Routes for Large Cities

The developed system involves: selection of optimal hubs and delineation of their influence areas; estimation of demand; and generation of optimal paths and service frequencies. Fuzzy clustering and Self-organizing-map methods aggregate the stops into a number of clusters. A hub is located with the objective to minimize the `Total Travel Distance' within a cluster. Optimal set of clusters is one that provides the minimum `Total Passenger Time' for the network. Routing model generates reasonably direct routes with high frequency. Optimal path is selected on the basis of maximum `Desire passenger-km per km'. Number of trips to be operated for a route depends upon the demand served along the route and the passenger flow on various links of the route. To estimate the optimal trips for a route, an iterative heuristic algorithm is adopted in the scheduling model. The devised methodology is successfully implemented for New Delhi, the capital city of India. The total daily transit demand generated at 1542 stops is 7.67 million passengers. A set of fifty optimal hubs is identified and 181 inter hub routes are generated. Eighty-eight of these inter-hub routes, which have very high demand, need to be operated with automated transit vehicles (APM). A total of 305 secondary routes (intra-hub) are also generated. Optimal scheduling is done for all the 486 transit routes. For peak period, a total fleet of 11497 standard size buses is required at the desired level of service.