How UK CIOs are using AI and machine learning

Virgin Trains CIO John Sullivan uses AI to automate ticket refunds for delayed trains.

“I think in the future, it will tell us the right frequency of trains to come and when to change trains,” Sullivan told CIO UK. “If we need to run more trains to Manchester, our data will be able to tell us that.”

The organisation has also transformed the ticketing experience, enabling customers to use Alexa to make bookings and receive digital tickets.

“About two years ago, less than 1% of our tickets were digital, but today it’s about 30%,” he said. “The customers’ ticket is on their mobile phone. They show the mobile phone at the gate, the gates open and the train manager scans it. It’s really convenient and they don’t have to do the queuing.”

Alongside this, Virgin will also refit all Pendolino trains to offer free Wi-Fi to all passengers. The new technology has been future-proofed to work with 5G mobile phone networks.

Read next: Virgin Trains CIO explains how he ensures adoption of new services

Volvo is exploring how AI could improve the predictive maintenance process. The automotive firm hopes to use AI to assess the condition of a car a customer is considering buying.

It also hopes to automate the whole maintenance process in the future, programming alerts of potential issues and instantly arrange a visit from a mechanic to make the necessary repairs. This will be supported by constant software updates and improvements.

“This is an industry with a heritage of long car lifecycles where the product doesn’t change for four or five years,” said Rafiq. “That is absolutely the opposite of where this industry is going.”

Read next: Volvo CIO Atif Rafiq draws autonomous car roadmap

Picsolve CTO, Dan Maunder has digitised the photography process for rollercoaster rides with the launch of a new cloud platform that adds facial recognition to existing video and photography systems.

The platform is built to automatically identify and organise multiple images from leisure attractions and send them straight to visitors’ digital albums using facial recognition technology.

“If we want to take a video of someone on a rollercoaster ride and give it back to them, or we want to take some really high quality, high definition pictures of them enjoying that moment, then we’d have invested that research and development time into new camera technology and new software,” Maunder said.

“With our business strategy of wanting to diversify in different sectors like water parks and stadiums, we needed to look at our technology in a different way. And rather than sinking lots of time into research and development for transient products that don’t last that long, we decided to create our core offering around our infrastructure and our service. I like to think of our AWS microservices platform as quite API-centric and also built in a fashion that we can plug in third-party products and diversify our suite much quicker than any of our competitors could do.”

The use of facial recognition gives consumers instant access to their photos, and provides Picsolve with a vast quantity of demographic information at different theme park around the world.

Read next: Picsolve CTO Dan Maunder redefines theme park photography

Post Office Group CIO Rob Houghton is transforming internal and external processes of the organisation by adopting digital capabilities.

“We treat digital transformation in two ways,” says Houghton. “One is how do we transform externally, so what apps do we build, how do we improve the service and make it quicker for customers, and two is how do we transform internally, and what do we need to do within the organisation to drive digital transformation.”

On the external side, Houghton is transforming the Post Office as a platform using APIs that partners will be able to use to add new innovations. For instance, CityMapper and Amazon Alexa could link to the Find The Nearest Branch service to give customers new routes to the nearest Post Office.

Houghton also plans to adopt machine learning technology in the future to add network resilience and boost security by detecting anomalies.

“If you think of our branches as a network, it’s quite a predictive network,” he says. “Most people come in, switch the lights on in the branch and operate their computer systems, and if you look at the flow of traffic, it’s quite predictive. With machine learning, if I can learn about that network, it will help me detect and analyse problems which were affecting them much quicker.”

Read next: Post Office CIO Rob Houghton fuses digital and physical customer needs

Experian CIO Barry Libenson uses machine learning to understand how applications behave and to identify any potential risks by monitoring their performance with Dynatrace and then analysing the data in Splunk.

“We can find things that are outside of normal tolerance much more quickly using machine learning than we were able to do previously because human beings just can’t process the information that way and aren’t nearly as likely to say that something is out of the ordinary as a machine,” says Libenson.

“Machines have no emotions. They don’t care. If the machine sees something that it thinks is abnormal, it’s just going to say: ‘I can’t make sense of this, somebody should look at it’. Whereas a human being is much more likely to put a subjective inference into that. That’s why the machine technology’s much better at monitoring this stuff and watching for bad behaviour and potential areas of problems.”

Read next: Experian CIO Barry Libenson explains latest changes in data analytics

Heathrow Airport CIO Stuart Birrell has overseen the development of an “insights hub” built on Microsoft’s Azure cloud that uses machine learning to predict passenger numbers and allocate resources based on the needs.

“The flights we already know. The big variability is how full each flight is, so we get some forward scheduling and forecasting from the airlines,” says Birrell.

We apply our own intelligence analytics to it, and then we know when passengers are likely to turn up, and we can plan for that. That’s where the machine learning comes in to get the specifications on quality.”

Read next: Heathrow Airport CIO Stuart Birrell’s new IT operating model delivering digital workplace and business innovation

Bloomberg CTO Shawn Edwards has driven the development of a range of new machine learning tools and techniques that support global investors, such as a data science platform for building machine learning models

Machine learning plays a growing role in the company’s iconic Terminal’s development. Clients can now access now rapidly receive high-quality financial data that’s automatically extracted from text and tailored to their requirements.

“We know your portfolio – it’s in our system,” says Edwards. “We know who you are, we know what kind of role you play, we know the functionality that you use, so we can target information to you in a really, really powerful way. We can bubble up the events and the data that is pertinent and relative to you.”

Read next: Bloomberg CTO Shawn Edwards explains how data science is guiding the finance industry

First Central Group CIO John Davison applies AI and machine learning to augment the traditional insurance underwriting process.

“The way the business works is you have an underwriting assessment of a risk and you perform a series of calculations. When they’re different, we have a team of people who do statistical modelling. We’re using analytics and AI for that process,” Davison tells CIO UK.

“We continue to perform the standard, actuarial calculations that are done and the modelling that’s done, and we then augment that with data enrichment machine learning to help us pick the right models.”

Read next: First Central Group CIO John Davison explains how data science and AI are transforming insurance industry

Belron Chief Information and Digital Officer Nick Burton has been working with startups in the vehicle glass repair business’ accelerator to develop AI systems that detect hail damage, understand recordings of customer phone calls, and make decisions on customer claims.

He has also recently deployed an AI-powered interactive voice response product that customers can use to independently change their appointments and a visual recognition tool that lets them spot vehicle damage in photos.

“That’s when we recruited the first AI person in my team,” says Burton. “We’ve now got an AI solutions manager who really understands the different platforms who works with the countries to help them shape how they’re going to try and solve the given business problem using AI. It’s not always the most obvious way but it’s the way that works.”

Read next: Belron CIDO Nick Burton applies AI across the vehicle repair business

Photobox Group CTO Richard Orme is developing “emotionally intelligent AI” that helps clients create better photo albums by suggesting photos and page arrangements based on an analysis of historical data. The objective is to create a “magic book” that will guess what clients will want and help tell their photo story.

“When we bring that kind of insight and that kind of assistance to the creation process, we actually see a higher conversion rate from customers,” says Orme.

“They do go on to buy more from us. We believe that there’s a great opportunity for us here, to create AI that doesn’t shorten that experience. It still takes you two weeks, but at the end of it, you feel like you’ve told the best story you can.”

Read next: Photobox Group CTO Richard Orme developing emotionally intelligent AI and realising DevOps squad goals

Southampton FC used data analytics to recruit the players that formed Europe’s most profitable youth academy. The club is now using machine learning to find even deeper insights.

“We use Python and R and we’re moving to machine learning for even better efficiencies,” says IT Director Matthew Reynolds. “When you start surfacing data initially, you find things you never knew and a lot you need to clean up.

“We’ve found players we had scouted that used to be on a written piece of paper and then dropped off the radar because they weren’t the hot topic at the time. Now we’ve resurfaced them and they’re looking at those players again because it was so misdated and hidden in a spreadsheet somewhere.

“Now it’s more centralised. We can see who’s been looked at, who hasn’t been looked at for a little while, and whether we go back and look at those players. It is quite exciting really and it’s something that the states have been doing in American football with analytics for quite a while. Football now is just seeing this transformation of data and insights.”

Read next:  Southampton FC IT Director Matthew Reynolds kicking tech into football

Lufthansa uses machine learning to determine ticket prices, determine flight schedules and plan staffing requirements. The airline has also created a predictive maintenance platform called Aviatar that analyses data from aircraft parts to continuously assess the condition of each component.

“We have close to 2,000 aircraft in the Aviatar who are constantly monitored,” says CDO Christian Langer. “Every second we ingest something like 60,000 data points in our systems coming from all these aircraft in the air, parts in the workshops or aircraft on the ground. We combine all this data and build our models and are able to predict the future behaviour of this specific spare part.”

Read next: Lufthansa CDO Christian Langer harnesses AI to compete with startups

Rolls-Royce uses data analytics to predict future engine faults and schedule inspections around their predictions.

“Ninety-seven percent of the faults found on our engines are automatically predicted,” says CDO Neil Crocket. “By planning and understanding how our engines work, we have reduced disruption to our customers by 40% in the last 13 years, and we’ve reduced our maintenance burden by 30% since 2012.”

Read next: Rolls-Royce CDO Neil Crockett drives data into engine design

Amnesty International is using AI to identify violence and abuse against women on social media platforms and track how the organisation is represented in the media.

“There are many media monitoring services available, and they are great at tracking sentiment and reporting how much is being written about an organisation,” Amnesty International CIO John Gillespie told CIO UK.

“This is sufficient for a company that is sending out a handful of press releases each month, but when you are issuing four or five a day and you want to know the impact of each one individually, you need something more sophisticated.”

To develop the tools, the human rights NGO turned to ASI Data, a London-based startup with the mission statement of “AI for everyone”. One of the ways it pursues this is through its ‘Data Science Fellowship’, a six-week programme for PhD graduates and software engineers that helps them apply data science to real-world problems.

The fellows developed mathematical models that evaluated where stories in the press related to Amnesty’s recent releases.

“The success of the fellowship project far exceeded our expectations,” said Gillespie. “The opportunity for Amnesty to experiment with data science in a low cost, low-risk way was perfect for us.”

Any organisation that needs to analyse a vast volume of data could save much time and money by deploying AI. The Serious Fraud Office is already reaping the benefits, thanks to an AI robot called RAVN that helps share the burden of reviewing more than 100 million documents annually in its investigations into major cases of fraud and corruption.

“In a large [case] such as Rolls-Royce, which resulted in a £671m settlement, we had 70 investigators working to review over 30 million documents,” Ben Denison, the Serious Fraud Office’s CTO, told CIO UK.

“It’s just not possible to manually review that amount of data, so we worked with our technology partners to develop an AI robot to assist with that. We were able to prove that this approach is both more accurate and much more efficient than human review alone – in some instances at one-fifth of the cost.”

Virtual assistants have been a popular early use case for AI. Financial services companies such as Capital One Europe have been exploring how they could improve better customer service.

“I’ve seen some early prototypes in our North American labs of virtual agents – be that chatbots, be that the recently announced integration into Amazon’s Alexa product – and I think we’ll see a lot more of virtual agents in the financial services industry and other industries,” Rob Harding, Capital One Europe’s Chief Operations and Technology Officer, told CIO UK.

“I think it’s a good example of helping customers interact with financial services companies with a lot less friction.

“We could use machine learning where we currently have manual intervention in aspects of our workflows, and we could even get to the stage where we use a lot of machine learning in our underwriting algorithms.”

Allied Irish Bank has already found a wide range of uses for data science, from a scanning system that digitises the information in customer documentation to finding errors in tax deductions on mortgages that can arise due to regular changes in the rules.

“It’s usually something we figure out and fix, but it annoys customers and it annoys the regulator and it costs us effort and time,” said the bank’s CIO Tim Hynes at the AI Congress London.

“We applied artificial intelligence looking backwards, and we discovered that if we had an AI watching what was going on and dealing with this for us, it would have identified over 90% of the errors that had slipped through the net.”

He believes that the true potential of AI will be unleashed through the combination of quantum computing and 5G.

“In the future, with quantum computing on the back-end and faster communication, the processing doesn’t actually have to be on the robot, so you start getting even more intelligent or clever activities and uses for the robots,” he added.

“As you look at this stuff, think about where we’ve come from, understand the pace of technology, understand it’s going to keep happening and ground it in practical use with a view to the future.”

AI could threaten jobs in all manner of sectors, and journalism is sadly no exception. Supporters of the technology often emphasise that it will augment rather than replace human workers, as the Financial Times Chief Product and Information Officer Cait O’Riodan told CIO UK.

“AI is coming on in leaps and bounds. Things that were ropey not that long ago are getting good very quickly. Speech-to-text recognition is really phenomenal now; there’s really good text-to-speech that we are experimenting with at the FT,” she said.

“All of this with AI is not aimed at replacing journalism but augmenting it, how can we use those things to make sure our journalists are concentrating on the high-value content that’s going to really drive engagement while removing some of the repetitive steps they may have concentrated on in the past.”

Working with startups can add expertise in AI to make deployments successful in large enterprises, as professional services and real estate investment management JLL found through a partnership with ‘proptech’ startup Leverton.

JLL turned to the Berlin-based startup to increase automation of its lease management operations and digitise key processes. The startup used machine learning and deep learning to optimise the review of lease documents by identifying, extracting and managing key terms such as rental values, dates and figures contained in JLL’s contracts.

“Our work with Leverton on machine learning technology implementation across our lease administration business is transforming the way we do things,” Chris Zissis, JLL’s EMEA CIO, told CIO UK.

“Lease contracts can comprise between three and 15 documents, hence, digitising the process significantly reduces the time spent reading and reviewing each document. It also allows the lease administrator to spend time applying subject matter expertise in recognising patterns, anomalies and opportunities.”

Healthcare organisations have provided some key proving grounds for AI. Data analysis can provide personalised treatment for patients and scan test results to discover early signs of diseases. Rachel Dunscombe, Salford Royal Group’s Director of Digital, told CIO UK she has high hopes for the technology.

“For me, the coming year is all about the big data side of it. And that’s going to be the new frontier along with artificial intelligence, machine learning – which allows us to automate more of the diagnostic process,” said Dunscombe.

“Those are challenges we’ve started work on and in the next year we’re going to see those actually bringing benefits and operationalised into care settings. It’s going to take time before we are doing this en masse; these are very early days – but it is real now.”