At a glance
- Zurich has developed a cutting-edge computing solution that uses elements of artificial intelligence in combination with other technological tools to improve its claims processes
- The system is designed to ease the administrative burden on Zurich’s claims handlers – giving them more time to speak to customers about their claims
- We look at how the system works and how customers reap the benefits
In the space of a few years, robotics and artificial intelligence have gone from being confined to the pages of science fiction novels to being referenced almost daily in news reports and opinion columns.
One reason for this is that there is now a far better understanding of the practical settings in which this kind of technology can be applied. Driverless cars, for example, use computer software to make sense of multiple sources of data in an instant – a driver indicating left, a pedestrian using a zebra crossing and so on. The more that driverless cars learn about the behaviour of other road users, the more intelligent they become.
Artificial intelligence is also being applied in many other settings – including within the insurance industry, as Neil Fraser, Claims Proposition Manager, Zurich, explains.
“For some years now, insurers have been using computer technology to automate many of their administrative processes,” he says. “One example would be the use of computers to perform repetitive tasks – such as transferring numbers from a spreadsheet to a database and so on – in order to free up claims handlers’ time.
Zurich’s cutting-edge approach
“This is essentially a computer doing what it is told to do. However, Zurich has been doing something in recent years that is far more cutting edge, in relation to cognitive analytics. This is where a computer is able to look at lots of sources of data, interpret that data and then make assumptions based on it.
“This provides a number of benefits; for example, we are using cognitive analytics to spot fraudulent activity and also to identify emerging claims trends. This is information we can then use to help manage risks relevant to their sector.
“Most recently, we’ve been working on a project to pull together a variety of technological tools – including robotics, cognitive analytics, and optical character recognition – into a single cognitive computing solution.
“This is where we really stand out. Very few, if any insurers in the UK, are using technology in this way.”
A new approach to assessing claims
The system is currently being used to simplify certain claims processes.
The technology uses optical character recognition and natural language processing to scan and read documents, and extract relevant information. The analytics element of the system then takes over and evaluates the information and makes a decision on how to proceed with the claim.
“While our system is not currently used for every type of claim, its biggest advantage is that it is continually learning and therefore has the potential to be used with more and more cases over time,” says Neil.
Keeping the personal touch
While Zurich’s cognitive computing solution is able to assess claims in a fraction of the time it would take a claims handler, the human touch is not lost, as every assessment made using the cognitive computing system is overseen by a Zurich expert.
“Our technicians still have oversight, but this system means the computers do the legwork,” says Paul Dixon, Head of Claims Relationship Management, Zurich. “It means our claims handlers have more time to speak to customers to discuss any concerns or queries they may have about a claim. This isn’t about using technology to replace people – it’s about using technology to improve our offering to customers.”
Zurich’s cognitive computing solution was a finalist in the Insurance Times’ 2017 Tech and Innovation Awards, which took place in September.
For more information please speak with your local Zurich contact