RETT2 - Field trial for improving regularity on trunk buslines in Stockholm
RETT2 is a joint project between SL, Keolis and KTH for testing a new strategy to improve bus regularity. A field experiment during October 2011 demonstrated the benefits from applying a new control strategy. The real-time headway-based holding strategy was designed to improve service regularity by seeking uniform headways along the line while considering each stop as a potential time point. The field experiment realized the expected benefits from introducing the proposed strategy that were first analyzed in a traffic and transit operations simulation model, BusMezzo, which was developed in the VINNOVA Dymobus I and II projects
Main results
The evaluation of the project consisted of detailed before-after data analysis and feedback from various parties. The main findings are:• Headway variation decreased for the peak period as well as for the entire day along the entire line for both directions by 11-26%. In addition, the share of bunched buses decreased by 13-24% .
• This led to a decrease of 38% in the excess waiting time and an extra budgeted waiting time that is shorter by 10% compared with the before period.
• The more regular service also resulted in passenger loads being spread 50% more evenly over the buses.
• The total dwell time along the line decreased by 10% and a dwell times were much more evenly spread along the line as each trip involved holding at different stops.
• However, the overall average bus running time remained unchanged. The variability of total running time reduced during the trial, mostly due to smaller tails of the running time distributions (very short of very long runs).
• The on-time performance improved by 9% on the eastbound and remained at the same level on the westbound.
• The share of buses that departed on-time from the origin stop, driver relief point and key stops remained at the same level or increased. It is worth noting that if the normal penalties mechanism was applied during the experiment, Keolis would have had to pay a smaller penalty than if the schedule-based control was had been place.
• The share of buses that arrived on-time to the last stop increased on both directions. Hence, driver scheduling was affected positively by the new strategy as drivers have the same or higher probability to make it on-time for their break or next trip. This was also reflected in the feedback received from drivers as they found the new working routine to be less stressful.
It is important to stress that the project benefits were achieved with existing technology (with updated software), thus without an investment in infrastructure or technology . The performance and level-of-service was improved by introducing a new operations strategy.
The promising results of the RETT2 project yielded a consensus among all involved parties that the new strategy should become the common practice for high-frequency lines. An extended demonstration in the fall of 2012 will test the applicability of such operations on more lines, together with additional measures to improve service regularity.