Theory and Practice of Metro Network Design

Most transit networks are designed empirically. For bus networks this process 1s often satisfactory because bus routes and networks are very dependent on local conditions, and they can be easily modified, allowing easy corrections of problems which line design may cause in operations. However, with transit systems which have extensive infrastructure, most typically metro lines and networks, corrections are extremely difficult to make. Development of an optimal network and avoidance of design features which result in operational problems are therefore of great importance. Yet, the experiences from the design and operation of such large older metro systems as London, Moscow, New York, Pan's and Tokyo, or from numerous recently built medium-size metro systems, such as Hong Kong, San Francisco, Sao Paulo and Washington, remain largely unknown to the designers of new metro networks. This paper presents a theoretical analysis of transit lines and networks and its applications. The focus is on network geometry and operational characteristics. The basic design and operational elements, such as geometric forms of lines, headways, schedules, etc., are discussed in general terms, valid for any mode; however, the main focus is on metro systems because of the particular importance of these analyses for fixed, permanent systems.