Integrated Data = Integrated Public Transport Networks

Need for integrated network

If you drive your car to the bus stop and then catch the express bus to the city, it is straightforward. But what if there was no parking at the bus stop?

If you had to travel into the city, walk four blocks and catch a different bus, operated by a different bus company, to your workplace, it is complicated but doable. However, if the service runs every 30 minutes and leaves 1 min after your incoming service arrives, the extra half an hour travel time might make this a very unattractive option.

If you caught the train using a weekly student ticket but needed a second ticket that could only be bought on the ferry to get to the university, you will pay two journey start fees, need additional change, and must follow different rules depending on the mode of transport. Even on a good day, this is not a great experience for the passenger.

These are all examples where transport networks are not integrated. In each instance, there is a strong disincentive to use public transport. For instance, if I need a car to get to the stop and there is no parking – I drive all the way. Easy.

What is an integrated network

An integrated network is one where a passenger can efficiently get to their destination without being affected by who runs the services even if the passenger must move between services. Passengers see a single transportation service that provides efficient options to get them where they want to go. Properly done the journey seems effortless and obvious.

PTA’s work in close alignment with government to set strategic direction for the city with the goal of reducing car use, achieving sustainability targets, and driving economic growth by reducing travel times. These all fuelled by providing an advanced, truly integrated and seamless network. This must be centrally planned and managed as no individual operator will incentivised to look at the picture in this way.

But it is only possible due to high levels of technology and planned integration. For instance, look at Journey Planning. A passenger wants to know the best overall way to get to their destination. They are seldom looking for the best ferry trip or the fastest bus trip. These can form part of the journey but are not the solution in themselves. Each mode could have their own journey planner but the real value to the passenger is being able to combine these modes to give a solution for the entire journey. Data must flow from the planning tools for each mode into a common journey planner with consistency in stop numbers, stop names and locations. The integrated journey planning tool can then give passengers options based around what they want (travel time, cost, accessibility etc). If you live next to the train line and work in the city near central station, then perhaps this integration is not so important. However, even for these passengers’ travel times and disruption information can make a difference to your journey and integrating these technologies will be of value.

So, what is needed to make this integration work?  First up, we need a vision of what the integrated network can look like. This rightly sits with the Transport Authorities. Many authorities have already recognised these issues, and some have made significant efforts to overcome them, integrating some or all, of the various aspects of Public Transport. Integrated ticketing is often the first visible sign of an integrated transport network. It is certainly necessary but is not always sufficient and needs support from other systems. The vision must identify that public transport services are needed to deliver services between the appropriate origin and destinations and, because cities grow and change, this should be monitored and managed.  Schedules need to be planned to allow the right frequency of service and to interconnect between modes. With timetables in place, services should be tracked in real time and information fed back to drivers to help them keep the services running to time. Setting these standards across all operators, is the governments job.

Visions should also recognise that things do not always go to plan, and when they don’t, disruptions need to be identified and managed, with passengers informed across a range of media including at the stop, on the vehicle and using the mobile phone. Even civil works play a part in the integrated network vision with park and ride car parking, BRT or light rail infrastructure and traffic signal priority for services that are “at grade”.

This is all a big ask. However, it is doable and indeed Transport Authorities have achieved this in several cities around the world like Singapore, London, and Zurich.

With the vision in place and political and financial commitments made, where do we start with delivering an integrated solution? Integrated models do not grow in isolation. Each part must support the other. Analogous to a truss where every part contributes so that the sum is far more than its parts. What underpins this integration and is the key thing we need to get right? The answer is data. Now, each subsystem in our future integrated transport network will have their own purpose-built proprietary data that enables them to do their job. The ticketing system needs to know about stop locations and who is driving (for auditing) but can operate without knowing about the timetable or the type of bus stop. However, this information is important to passengers, particularly if they are using a wheelchair. So, the key is not just having any data, but having integrated data. Data that can be shared, data that can flow from one system to another, data with a common base reference that that means the same in all the systems, data that can be used to deliver value to passengers.

Standards

To deliver this value and interconnect the various systems there needs to be a common way to refer to concepts and to share data between systems.

A common time reference seems obvious nowadays, but in 1840 every town in the UK had their own local time and the Great Western Railway in England was looking for a way to effectively schedule their trains and tell passengers when they would arrive.  They were the first to use Railway time [1] as a standard across the UK. This transport driven innovation worked well and quickly spread across the country bringing the whole population to the same time base. The USA was a bit slower to catch up taking another 43 years. This was despite a major accident in 1853 when two trains heading towards each other on the same track, collided as the train guards had different times set on their watches, resulting in the death of 14 passengers.

The use of railway time meant that schedules could be more simply planned, passengers would know when the train was leaving and accidents like that in the USA avoided. The electric telegraph was quickly used to provide synchronisation of the various railway clocks. Nowadays this is almost universally done by GPS signals.

The need for integrated data extends right across the transport spectrum. Do we all know what a trip is? What is a journey, how is a stop number described? Integration requires us to use the same words for the same concepts. This is where standards come to play.

The European CEN Transmodel was designed to provide a reference for common PT data elements and data structures. This is supplemented by NeTEx, which defines the interchange of this static data, and SIRI which defines the interchange of real time data.

Adopting these standards gives us the language to be able to interchange data.

But, whilst data schemas allow us to reference the same data elements, it does not mean that we have an integrated data set. The content must be right too.

Data needs to be aligned. For instance, stops used to be numbered based on the line. In early trams systems in Brisbane and Adelaide this was straightforward enough. The lines were radial, and the tram would start in the city and stop 1 would be the first stop outbound, stop 2 the next outbound and so on. This flowed on to the bus routes and resulted in a lot of duplicate stop numbers and where routes shared common sections meant that the stop could have multiple stop numbers.  Ie. it could be route 5 stop 3 and route 17 stop 4. Rationalising this to provide unique stop numbers was the first stage to integrating the data.

Stop names are not much better. Often picked by the nearest landmark there might be 2 or three stops named “Blue Lake”. You don’t have to look hard to see that it’s not just stop data but also the road and track network, restrictions, and special public transport road/tracks – the base layer for planning that can have these types of issues. In the extreme, common data includes the names of walkways to be taken to move from one service to another, say from the bus stop to the railway station.

Standards helped again with the UK implementing the NAPTAN schema to define all the stops in the UK in a consistent way. The hierarchical structure provided a mechanism for providing national uniqueness and still allowed local flexibility.

Integrating the technology

With a schema sorted out and the data rationalised, we are now able to make use of this integrated data. The planning tools can derive timetables that connect the various modes, prioritise connections and minimise waiting times. These are sent to a journey planning system for passengers. Some of this data is sent to the ticketing system and AVLC systems.

The data allows the AVLC system to implement connection protection between different services including trains, ferries and buses, or to manage vehicle headways. These features rely not only on integrated planning data but real time monitoring and feedback on the service being delivered and they further improve the passenger experience.

Emerging standards like ITxPT provide a data centric way to access on-vehicle information. Location data, passenger counts, passenger information, even CanBus data are all available without the need for the requesting system to know how this data was generated or where it is stored. They only need to know what the data is called. This enables operators to purchase new technology that with minimal integration, enhances the quality of the passenger experience and the management information available.

SIRI – frees the data from its source. For example, in Melbourne the DoT take data feeds from multiple AVLC sources, aggregate these and then make them available to web sites, and information management systems. Systems like the Trapeze PSP send these integrated real time arrival predictions to a range of at-stop passenger information displays.

Mobility as a Service (MaaS) apps are also reflecting the integrated views of the network. First/last mile options including taxi’s, bikes and scooters, can be planned into the trip with passengers booking and paying for these via a single integrated MaaS app.

Ticketing is evolving with Mobile fare systems rapidly making the mobile phone not just the payment device but also the ticket itself and requiring little to no on-board hardware.

Barriers to integration

As any PTA who has gone down this path knows, there are many barriers to an integrated PT network. Including:

Political

In many cities the rail networks are run by a totally separate organisation to the bus operators and authority. Buses are in competition with rail or tram rather than complementing them.

Malaysia’s LRT franchise was a good example of politics being a barrier to integration. After the concessions were let for the 3 LRT lines, the government found it was unable to get concessionaires to accept needed route alignments or physical changes to stations needed for effective system integration. This was compounded by the inability to streamline bus operations or prevent private bus operators from plying LRT routes in competition to the LRT. Politics meant the government failed to restrict parking in the city and regulate parking charges and failed to restrict private cars in the city centre.[2] The result was no integration between the lines and no integration to the rest of Kuala Lumpur’s transport network. There was little incentive to use the systems which eventually went broke and were brought under government control.

Economic.

If operators are dependent on the fare box to deliver services, then there is no incentive for them to make it easier to work with a competitor’s service. Similarly, if the government only runs the unprofitable services, then there is no incentive to move passengers to these services. Private operation does not drive integration. For instance, in Bangkok the Skytrain (BTS) and the MRT rail lines are run by separate organisations and use different ticketing systems. There is no desire to integrate with each other or the bus services that are seen as competitors.

Many governments such as Victoria’s DoT, TfNSW, Dubai RTA, Singapore’s LTA or London’s TfL are bringing all modes under a single transport entity and these Authorities have come to the view that they need to own all the infrastructure, take the fare box risk, and tender out the operations. They can then define the service levels they want to be delivered to address the needs of the passengers, leaving the day-to-day operations with the operators to deliver this at the lowest overall cost. In Singapore, this meant a major change from private to public ownership with a strong focus on operator performance against a service quality and ongoing reporting against that standard. An advanced claims processing was introduced for events that were outside the operators control.

Technical.

Whilst technology is in a large part the solution to an integrated network, legacy systems are often well entrenched in an operator’s business and are both difficult and expensive to replace. This is one reason to move to open standards that allow for an incremental move from a legacy system to a modern ITS solution.

How can it work?

Integration can take several forms. In Bogota, civil works meant the dedicated BRT lines were built with bike lanes alongside enabling safe travel. They went even further with the popular Ciclovía turning many key roads into “exercise lanes” on a Sunday morning for bikes, rollerblades, walking and all sorts of human powered transport [3].

On a larger scale is the Swiss public transport network (Zurich and surrounds). This is a multimodal system that integrates buses, trams, trolley cars and the national rail system into a single view for the passenger. There are hundreds of connections enabling passengers to travel throughout the country across multiple modes. This information is easily shared across six different transport agencies, 40 transport companies, 46 depots and their associated computer systems using agreed standards, with mutually agreed codes.

In Zurich, government vision and data integration has allowed the transport authority to increase network capacity and maintain punctuality by applying network-wide planning [4]. This plan allows for more frequent services on fewer lines and provides flexibility.

Passengers can book and pay for services, travel across the city, region or country by bus, train and ferry and have confidence in their journey times and connections. Often a car is not required at all. When it comes to Public Transport, Switzerland just works and it is all due to using the right technology driven by integrated data.

Conclusion

Integrated Public transport networks make life better for passengers and encourage the use of public transport. Done well, there is often no need for a car to be used. However, private operators in competition with each other have no incentive to integrate the network or the supporting data. Deliver an integrated network requires a strong vision by the public transport authority, a common data schema and rationalisation of the data itself. This needs to be supported by a mechanism where the government takes holistic responsibility for the operations and operators’ focus on delivering to a defined service standard.

With the vision in place, the use of standards will support integrated data, and ITS technology can deliver valuable services and information to passengers about the network – across the network. These services help make passengers use of public transport frictionless and make the integrated public transport network far more than the sum of its parts.

This really is a case where integrated data means integrated public transport.

Reference:

[1] https://en.wikipedia.org/wiki/Railway_time
[2] Tan, Jeff, Privatization in Malaysia : regulation, rent-seeking, and policy failure, Routledge, 2008
[3] Moro, Aris, Andrew Eil, and Prajwal Baral (2018). “Cycling Infrastructure in Cities: Bogotá’s Quinto Centenario Cycle Avenue – Creating the Enabling Environment
[4] https://trapezegroup.com.au/ee_assets/downloads/Zurich_trapeze_feature_rail_express.pdf