Your Cart

Your cart is empty

Home  >  Volume 28. No.2 (Nov. 2014)

33. Stop Dwell Time Modeling for Unscheduled Urban Bus Public Transport Operation by Douglas O. A. Osula and Osagie J. Akumabor - Volume28, No. 2, (November, 2014), pp223 – 234
Sale price: $5.00


Regression model for stop dwell time regressed on number of passengers alighting and boarding was developed for bus transit operation in developing countries using Benin City in Nigeria as a case study.This has been to investigate the hypothesis that the bus service as it is operated in this setting studied is mathematically tractable for the purpose of public  transport infrastructure planning and design. The regression model was achieved through Box-Cox transformation approach in order to allow the data evolve the mostappropriate functional form. Further exercise was carried out to test the stability of the model that evolved against an earlier one developed in the same setting for minibus paratransit service.The result obtained reveals that stop dwell time-number of passengers for unscheduled, flag stop bus transit operation in a developing country setting can be adequately represented by a semi-log model, which is transformable toanexponential model.The parameters of the model are interpreted thus: 3.089 is the theoretical minimum dwell time (in seconds) of a bus at a stop for the simultaneous activities of passenger alighting and boarding, while 0.253 is the rate at which dwell time increases with the number of alighting and boarding passengers processed at a stop. This theoretical minimum stop dwell time compares well with the actual value of 3seconds measured in the field. Though this exponential functional form is similar  to an earlier one developed for minibus paratransit service in the same city, analytical stability test using the parameters of the models (i.e. the bus transit and  minibus paratransit models) revealed that the models are not stable across modes.This may be due to the fact that while the minibus processes both boarding and alighting passengers through one door, the bus processes passengers through two doors. In addition, boarding passengers pay the fare as they board the vehicle in the bus service, while for the minibus, passengers pay while already seated in thevehicle.The implications of the results of this work are that: (1) the bus transit  operation is tractable and the model can be used for transit planning and design purposesfor this setting and similar settings, and(2) for this setting and similar settings, stop dwell time characteristics for one mode are not usable for planning and design purposes for another mode. Keywords:Bus transit, stop dwell time, flag stop, regressionmodel, system stability.