This article proposes an optimization model to set frequencies, vehicle capacities, required fleet and the stops serving each route along a transit corridor which minimize the total user and operating costs. The optimization problem is solved by applying the 'Black Hole' algorithm, which imitates the movement of stars (solutions), towards a black hole (Best solution). The main contributions of the model are based on incorporating variable dwell times depending on bus stop demand not only to the passenger perceived journey times but also to the bus cycle times and on considering capacity constraints in both vehicles and bus tops. This led to a more accurate and realistic operating times and user perceived journey times. The application of the model to two case studies and the sensitivity analysis carried out demonstrate that for low levels of demand, constant dwell times can be assumed but being these times different between the different stops of the corridor, considering their demand. However, with high level of demand the difference found in operating costs and travel times strongly recommend incorporating variable dwell times in the model in order to achieve a more realistic design of transit corridor strategies.