Nava-Fonseca, C. A.
Laboratorio de Transporte, Departamento de Sistemas, Facultad de Ingeniería, UNAM.
Received: September 1th, 2005.
Accepted: March 16th, 2006.
]]> Abstract
The methodology used for the prediction of trips through the allocation of these according to the origin-destination (OD) survey, constitutes a traditional methodology to cover the demand of existing transport between the zones of analysis. When the transportation system fulfills the characteristics of a network (topology, routed arcs, nodes with commodity supply and demand, etc.) in addition, fleet restrictions are modeled as constraints on this network (capacity, transfer, flow balance, etc.); therefore the transportation system can be worked like a Multicommodity Network Flow Problem (or MCNF). If these data are contained in georeferenced databases, it offers the possibility of spatial analysis through specialized software, allowing the manipulation of such so it can be formulated as a minimum cost multicommodity network flow. The results offer better policies for the analyzed network than the results of the traditional methodology. Finally, the variations applied in the sensitivity analysis, purpose options such as: design for the future expansion, recommendations to the Master Plan of Electric Transport Systems, as well as of operating options to the present topology of the network.
Keywords: Multicommodity Network Flow, Network Design, Transportation Network.
Resumen
La metodología usada para la predicción de viajes por medio de la asignación de viajes según la encuesta origen-destino (OD), constituye una metodología tradicional para cubrir la demanda de transporte existente entre las zonas de análisis. Cuando el sistema de transporte cumple con características de una red (topología, arcos diseccionados, nodos con entradas y salidas, etc.) y además se cuentan con las restricciones aplicadas a ésta red (restricciones de flotilla, capacidad, transbordo, balance de flujo, etc.) el sistema de transporte se puede trabajar como un modelo de red de flujo multiproducto (Multicommodity Network Flow o MCNF). Estos datos contenidos en bases de datos georreferenciadas ofrecen la capacidad de análisis a través de un software especializado en análisis espacial, permite la manipulación de tal forma que se pueda formular el problema de flujo en redes multiproducto de mínimo costo, y cuyos resultados ofrecen mejores datos de la operación de la red analizada. Finalmente, las variaciones a las restricciones en el análisis de sensibilidad, realizadas en el problema principal y el modelo planteado, resultan en opciones: tanto de diseño para la expansión futura, realizando recomendaciones al Plan Maestro, así como de opciones de operación a la topología actual de la red.
DESCARGAR ARTÍCULO EN FORMATO PDF
References
]]>[1] Kennington, J.L. and R.V. Helgason, "Algorithms for network programming", John Wiley & Sons, New York. 1980. [ Links ]
[2] I-Ling Wan, "Shortest Paths and Multicommodity Network Flows", Georgia Institute of Technology. 2003. [ Links ]
[3] Hartman, J. and Lasdon, L., "A generalized upper bounding algorithm for multicommodity network flow problems", Networks, vol. 1, no. 4, pp. 333-354, 1972. [ Links ]
[4] Kennington, J., "Solving multicommodity transportation problems using a primal partitioning simplex technique", Naval Research Logistics Quarterly, vol. 24, pp. 309-325, 1977. [ Links ]
[5] McBride, R., "Solving embedded generalized network problems" European Journal of Operational Research, vol. 21, no. 1, pp. 82-92, 1985. [ Links ]
]]>[6] McBride, R., "Advances in solving the multicommodity flow problem", Interfaces, vol. 28, no. 2, pp 32-41, 1998. [ Links ]
[7] McBride, R., "Progress made in solving the multicommodity flow problem", SIAM Journal on Optimization, vol. 8, pp. 947-955, 1988. [ Links ]
[8] Murtagh, B. and Saunders, M., "Large-scale linearly constrained optimization", Mathematical Programming, vol. 14, no. 1, pp 41-72, 1978. [ Links ]
[9] Castro, J. and Nabona, N., "An implementation of linear and nonlinear multicommodity network flows", European Journal of Operational Research, vol. 92, pp. 37-53, 1996. [ Links ]
]]>