The second in this series of posts looks more broadly at Journey Planner, offering an introduction to how data is made available on the system, from the research stage through to release.
Because we aim to ensure the system is as accurate and as broad-ranging as possible, we have recently increased the frequency of our data updates, which are now up to five times a week.
The process of updating data has four main components.
- Research – This involves producing a detailed list of forthcoming timetable changes for all of TfL’s forms of transport, ensuring that the change is fully understood by our team, data is available and that work is prioritised so that new information is available as far ahead of service adjustments as possible. Alongside this, we research information on bus stop adjustments and Points of Interest, making changes to our system as necessary.
- Entry of data – Bus and Underground timetables are imported from internal systems, but require some further processing to ensure that they are correctly reflected in journey solutions, such as matching the destination shown on the front of buses. Other data requires manual entry into the back-office system we use.
- Build – This puts all new data into a usable format for journey solutions, creating a dataset which we can test. Each new piece of data is verified for accuracy, and corrected if necessary, to minimise the likelihood of wrong data reaching customers. The confirmed dataset is shared with the Traveline and Transport Direct journey planners which cover the rest of the country.
- Release – This is the point at which new data appears on Journey Planner and the API and features in journey solutions. Messages explaining engineering works and major service changes are also added. A weekly newsletter is produced for TfL colleagues describing forthcoming changes and new data which has been added.
We believe that new information should be made available to Journey Planner users, developers and staff as quickly as possible. This is why we’ve moved away from the fixed weekly cycle to a more flexible approach where we release new data as soon as possible.