This study evaluates BRT system in Surabaya, Indonesia considering various mode choice contexts based on choice context SP design. Mode choice models of multinomial logit type have widely been developed based on stated preference survey considering some future hypothetical scenario where access mode, main mode and egress mode availability differs from one trip makers to another. The model used in this study analyzes commuters' main mode and access mode choice behavior for work trips in respect to individual characteristics, socio economic variables, mode attributes, mode availability, vehicle ownership etc using data from the Surabaya City. The analysis implies that important deterministic variables for the individuals travel decision include household monthly income, age, job status, travel time and access mode choice was found conditional to main mode preference. This study reveals commuters' hidden preferences for modes that are not in existence, in particular the bus rapid transit System in the Surabaya City due to be fully implemented in 2012. The developed models were then used to express complex travel behavior and forecasting travel demand for new transport services by simulation analysis. Changes in mode choice probabilities in response to changes in travel fare under various mode choice contexts were determined in proposed bus rapid transit system. Finally, policy recommendations are made for effective use of existing transportation modes in Surabaya city for successful implementation of BRT.