The central business district is not the only, or even the main destination of people living in big cities. Residents need to travel to numerous places across a city, which is why the modern approach to public transport planning is to build multi-destinational networks providing mobility from and to almost anywhere 1, 2. Ideally, this is achieved by arranging lines in a grid pattern with high frequency services, which is how metro systems operate in London, New York, Paris and Tokyo. The inconvenience of transferring from one line to another is kept to a minimum because transfer waits are short.
But what about a network where demand does not just justify services operating more than twice an hour? Doesn’t that mean it is impossible to have convenient transfer waits at interchanges?
Many European countries, best exemplified by Switzerland, have overcome this problem by adopting integrated pulse timetables. Where intercity trains run at 30 minute intervals, for example, they are scheduled to arrive at interchanges a few minutes before the hour and half hour and depart a few minutes after the hour and half hour. This allows passengers to effortlessly transfer from one line to another, just as they would in a high frequency grid network. Schedules are also organised so that trains always cross at the same intermediate stops. This optimises bus connections, providing quality public transport even for rural communities.
It would not be difficult for Sydney Ferries to use the same approach. Almost all lines already operate at 30 minute intervals in the off peak. With minor adjustments, lines converging on the main terminal at Circular Quay could operate to a pulse timetable, with transfer waits of between 5 and 11 minutes. The ferry travelling from Darling Harbour, for example, would have favourable connections at Circular Quay with ferries to Manly, Taronga Zoo, Watsons Bay, Double Bay, Mosman and Neutral Bay. Using this method it can be shown that the number of origin-destination pairs in the Sydney Ferry network with regular, convenient connections can be increased from the current number of 96 to more than 400, with a rise of only 11% in revenue hours 3.
Ferry to bus connections are also improved because ferry cross overs occur at the same intermediate stops all day. Convenient transfers can be made from a bus to ferries travelling in both the inbound and outbound directions.
As well as benefiting passengers, there are unexpected operational efficiencies from an integrated pulse timetable. The regularity of the timetable allows crew rosters to be modular, with minimal wastage beyond mandated crib breaks. This reduces the government subsidy per revenue hour. Investment in infrastructure, especially wharf upgrades, can be better targeted as operational requirements are anticipated more accurately than where the timetable is aperiodic (follows a different pattern at different times of the day). The network itself is modular, which would allow Transport for NSW to contract additional lines or increase service frequency without interfering with existing services.
There are constraints or “rules” to be observed in implementing an integrated pulse timetable. The cycle time for each route (time it takes for the vessel to travel out and back from its destination, including layovers) must be a whole integer multiple of the service interval. And at an interchange like Circular Quay, there needs to be a separate berth for each route to avoid congestion at the pulse point. This second constraint can be overcome by terminating routes originating west of the city at the new Barangaroo terminal.
The science of public transport timetabling needs to be recognised as important in New South Wales as it is in continental Europe, where quality scheduling is considered the most important element of the customer offer.
Robin Sandell is Principal, Sandell Consulting, Member, Ferry Transportation Committee, US Transportation Research Board, Associate Member, Royal Institution of Naval Architects.