This paper presents and tests a method to design high-performance transit networks. The method produces conceptual plans for geometric idealizations of a particular city that are later adapted to the real conditions. These conceptual plans are generalizations of the hybrid network concept proposed in Daganzo (2010). The best plan for a specific application is chosen via optimization. The objective function is composed of analytic formulae for a concept‘s agency cost and user level of service. These formulae include as parameters key demand-side attributes of the city, assumed to be rectangular, and supply-side attributes of the transit technology. They also include as decision variables the system‘s line and stop spacings, the degree to which it focuses passenger trips on the city center, and the service headway. These decision variables are sufficient to define an idealized geometric layout of the system and an operating plan. This layout-operating plan is then used as a design target when developing the real, detailed master plan. Ultimately, the latter is simulated to obtain more accurate cost and level of service estimates.
This process has been applied to design a high performance bus (HPB) network for Barcelona (Spain). The idealized solution for Barcelona includes 182 km of one-way infrastructure, uses 250 vehicles and costs 42,489 €/h to build and run. These figures only amount to about one third of the agency resources and cost currently used to provide bus service. A detailed design that resembles this target and conforms to the peculiarities of the city is also presented and simulated. The agency cost and user level of service metrics of the simulated system differ from those of the idealized model by less than 10%. Although the designed and simulated HPB systems provide sub-optimal spatial coverage because Barcelona lacks suitable streets, the level of service is good. Simulations suggest that if the proposed system was implemented side-by-side with the current one, it would capture most of the demand.
[1]
Francesc Robusté,et al.
Improving Bus Travel Times with Passive Traffic Signal Coordination
,
2009
.
[2]
Gordon F. Newell,et al.
Some Issues Relating to the Optimal Design of Bus Routes
,
1979
.
[3]
Hani S. Mahmassani,et al.
Hybrid route generation heuristic algorithm for the design of transit networks
,
1995
.
[4]
Carlos F. Daganzo,et al.
Structure of Competitive Transit Networks
,
2009
.
[5]
Michael Eichler,et al.
Bus lanes with intermittent priority: Strategy formulae and an evaluation
,
2006
.
[6]
Nigel H. M. Wilson,et al.
Bus network design
,
1986
.
[7]
Carlos F. Daganzo,et al.
A headway-based approach to eliminate bus bunching: Systematic analysis and comparisons
,
2009
.
[8]
Christoph E. Mandl,et al.
Evaluation and optimization of urban public transportation networks
,
1980
.
[9]
S. B. Pattnaik,et al.
Urban Bus Transit Route Network Design Using Genetic Algorithm
,
1998
.