Each trip planner has its own data integration strategy

There is no single model for trip planners! It is possible to classify them based on the diversity of their features, their target audience, the slickness of their user interface, etc. And why not based on their data integration strategy?

The main categories of trip planners…

Today, the market of journey planning solutions can be divided into three main categories:

• Proprietary mainstream solutions designed for a general audience (e.g., Google Maps, Transit App, Citymapper…),
• White-label applications provided as commercial solutions for public transport authorities (PTAs), whether they are based or not on open-source trip planning engines,
• Solutions that specialise in a particular mode of transport (cycling, walking, etc.), whether open-source or not.

Very often, the latter two categories rely on open-source route calculation engines. These engines can provide either end-to-end journey or a specific leg of a multimodal journey. In that case, they are very often combined to provide the final journey calculation to the user, in accordance with specific interfacing and interoperability rules.

… have completely different data integration strategies

Each category of journey planner has its own specific characteristics when it comes to data integration.

Mainstream solutions rely on mass data integration

Very often, mainstream solutions rely on essential information about transport services. This information is integrated primarily using GTFS Schedule and GTFS Realtime.

In this case, open data is used as:

• The primary means of analysing the completeness of the available data,
• A tool for assessing the integration effort required in relation to the volume of passengers covered.

As these solutions are designed to cover as much territory as possible, they only occasionally opt for direct integration of datasets from transport authorities or operators. And, in all cases, they ensure a certain level of completeness in the datasets before they move into the preparation phase. This preparation phase involves correcting or supplementing the collected data, or transcribing it into an internal format that enables the algorithm’s performance to be optimised. The data that has been reworked or enriched is generally not republished. As it is a fundamental differenciating factor for the trip planning solution, it will most certainly not share it for potential competitors to reuse.

White-label solutions prioritise data collection from source systems

White-label solution providers tend to collect data directly from the source systems. Given the commercial relationship between the service provider developing the application and the commissioning PTA, access to scheduling and/or CAD/AVL systems is facilitated.

Different approaches can be identified depending on the level of responsibility the PTA give the service provider.

Some PTAs largely handle data management in-house, taking responsibility for preparing the data prior to its integration into the trip planner’s database. In this situation, the white-label vendor largely withdraws from the process and does not assist the PTA in data preparation.

Other PTAs, by contrast, prefer to delegate data preparation to the vendors responsible for developing the trip planning application. Then, the vendors take a more proactive role in data preparation and integration. They are required to set up specific connectors (e.g., for on-demand transport or shared mobility services) and to ensure the quality of the data collected. Consequently, they turn these data management services into a genuine selling point.

In terms of data formats, a more diverse range of approaches can be observed: for example, relying simultaneously on the GTFS file produced by the scheduling tool and on the SIRI feed from the CAD/AVL system.

White-label solution providers will also tend to collect more specific data. This often involves integrating more modes of transport than mainstream platforms do, or creating links with other services, particularly ticketing systems. Consequently, the data is available in the internal format used by the journey planner, making it more complex to extract in standardised data formats for future use by the PTA.

Specialised solutions favour direct data collection

Specialised journey planners rely on highly detailed and non-standardised data (or not yet standardised). The processes involved in collecting, assessing quality, and integrating this specific data are an integral part of their differentiation strategy. Thus, these vendors become experts in a particular mode of transport (e.g., walking, cycling), enabling them to interface their solutions more easily with white-label solutions.

Where possible, these solutions will prioritise standardised data formats, so that they can make the collected data available to the transport authority. We can highlight the use of NeTEx for pedestrian and accessible route planning solutions, which enables these solutions to open up further market opportunities whilst benefiting from enhanced integration with the topology of the public transport network.

Open-source engines primarily use open data

As for open-source calculation engines, they tend to favour the use of open datasets made available in GTFS Schedule and GTFS Realtime formats. This provides them with numerous datasets forming a common basis for refining their calculation algorithms. It is on this last point that their value proposition is strongest, rather than in the area of integrating or enhancing multimodal mobility data.

They can serve as building blocks for a white-label solutions, providing end-to-end journey planning for the user, as well as a user interface specific to the PTA.

Hybrid approaches

It is important to underline that the data integration strategies implemented by the various stakeholders can be combined. For example, a mainstream solution could use data integration as a selling point and negotiate a partnership with the transport authority to guarantee an access to richer data, or even other data types that sets it apart from the competition.

Insofar as public transport authories and operators want to be visible to all journey planners, they should be defining their own strategy for data management. Having a strategy for in-house data management guarantees:

• Full access to multimodal mobility data, especially rich data,
• Consistency of exported data regardless of the standardised format required by the journey planner,
• Differentiated data exports depending on the type of journey planner to feed (mainstream, white-label, etc.).

➡️ How Chouette and Ara help you manage your data to feed trip planners