Operational research computerised allocation of tickets and services (ORCATS) is a centralised legacy computer system used on passenger railways in Britain. The system can be used for real-time reservations and revenue sharing on tickets between train operating companies.
However, the system has been facing setbacks – primarily due its age – such as not being able to account for online booking or identify split ticketing issues. This has created a significant loss in revenue for operators around Britain.
James Bain, CEO at Worldline UK & I, explains the current shortfalls of ORCATS, as well as some possible new management technology that could replace it.
Frankie Youd (FY): Could you tell me about the history of ORCATS?
James Bain (JB): ORCATS was created in the 1970s by British Rail. It is a computerisation of the railway clearing house, which was invented in 1842 to make sure the private companies back in the 19th century shared passenger fares between them equally.
It was based on demand profiles. They went out and surveyed how many people were on trains and came up with a statistical model you can use. The model allocates revenue between train operating companies and from the fares the customer pays to travel on the network.
These demand profiles were updated in the 1990s as an early part of privatisation; the industry has been using those demand profiles for nearly 30 years.
Every day, when a rail ticket is sold, it gets put through the ORCATS system. It gets divided up against the operators and the proportional part of the fare gets shared between whichever operator the ORCATS calculation says should have that part of the fare.
FY: What are the benefits of having a system like ORCATS in place?
JB: For the public sector, the historical benefits of having the system in place was getting people from one end of the country to the other [without them having to buy] multiple tickets to travel on the 122 private companies that founded the railway.
British Rail were able to allocate the revenue to the operating regions to allow them to create a profit and loss, which then allows them to service their costs and deliver them to the network.
From a private sector perspective, it was used to generate the revenue lines for the franchises that the Conservative government created in the 1990s that allowed private sector companies to bid for it. It underpinned all of the revenue lines and revenue modelling.
FY: What are the most prominent issues currently surrounding the ORCATS booking system?
JB: It’s on an old mainframe, written in old computer language. It only gets calibrated twice a year at timetable change. It takes about three to four weeks to calibrate it for every route and possible journey.
ORCATS takes a typical day of the week, normally Wednesday, and then it takes a typical Saturday and a typical Sunday as we call it. We use the word typical because there can be swings and roundabouts on demand due to seasonality (timetable change in December is one run, and the timetable changed in May is the other run). The rail industry has a 52-week planning horizon for the long-term timetable.
That gets pumped through the ORCATS model, but does nothing to take tickets sold into account – for example, nothing from smart cards [or] digital sales. One of the biggest risks in the industry that comes from ORCATS, as well as split ticketing, are ghost journeys, [which are] journeys that don’t really exist.
If you went from Crewe to London, you could split that journey at Stafford. You’re only one person, but the systems will record you as two journeys and two people. So that’s why we call them ghost journeys, [and] they are a significant issue in the industry.
FY: Is there any new data management technology that is being considered to replace ORCATS?
JB: There’s no reason in the world we live in today why ORCATS must exist. ORCATS runs twice a year, [but] you can run it daily with the technology that we have; you can set all the revenue daily based on true demand and true ticket sales.
The technology that we are using to move this forward is graph database technology. We work with Neo4J, one of the world’s leaders in this space.
We baseline the real data and then bring in extra datasets to improve the model, whether that’s tap data from smart cards, contactless payment [or] ticket sales from online from sites like Trainline. The whole point of pulling that together is to get a true understanding of real demand, real journeys and real people on real trains. The output of it will allow better focus on price instead of fare.
We need to move away from fare and ticket and move to price in the same way the airlines do and as the service industry does. That then allows generation of yield and the use of more technology to provide better offers to the customers that are not constrained by ORCATS running twice a year.
FY: How do you think that British railways will benefit from this modernisation?
JB: For me, its demand-led timetabling. Being able to provide a timetable against the real demand that is going to come. If we do that as an industry, it will improve performance and reliability. The other benefit is the flexibility it will bring in on price. It will allow for more choice and flexibility for the customers to access the service.
The two benefits are the ability to create a timetable based on true demand and provide the price point to a customer that really gives value.
FY: What do you think the future of rail technology holds?
JB: There’s a lot of technology that exists in other markets that should exist in the railway. Look at the hotel industry – if we use Booking.com, we are kept up to date on the booking and anything that may change around it.
We are also provided offers for other sites and destinations. That’s the technology the railway industry needs to be focusing on, [as it offers] more external focus to customers.
The industry needs to be more open and use technology to open itself to innovation from other markets, other companies and creators of real customer propositions. I think the industry needs to open up, take a few more risks post-Covid-19 and learn as it goes.