Steak & chips - how IoT and machine learning will disrupt risk in Animal Insurance

How IoT and machine learning will disrupt risk in Animal Insurance

A partnership between IoT and animals is not an obvious one.

On the face of it, a partnership between the Internet of Things (IoT) and animals is not an obvious one. However, a number of trials and larger-scale implementations of IoT use with household pets and in farming are showing that connected ’Smudge’ and ‘Daisy’ can provide real benefits.

This should not be surprising because animals are big business, both personal ownership and commercial farming. Mintel, a market research company, expects UK pet insurance premiums to grow from £ 976 million in 2015 to £ 1.6 billions by 2021. Animal farming, in particular, tends to come under fire when it comes to welfare, with farmers always seeking new ways to increase productivity. Anything therefore, that makes it easier and more efficient to care for animals, while also potentially improving their health & well-being, is likely to be of interest.

Connecting the dots

How, then, are animal farmers using IoT?

Cows are perhaps the most connected farm animals — possibly not surprisingly, since they are potentially the most valuable. A number of companies have developed sensors to measure and assess particular indicators. For example, Irish company Moocall has a sensor that attaches to a cow’s tail and detects movement. Tail movements apparently change when the cow is in labour, meaning that farmers can be alerted when this is happening, and also if there are problems or the labour is lasting too long. In the UK alone, around 50,000 cows and 110,000 calves die each year because of problems during delivery. The app therefore has huge potential to save animal lives and therefore farmers’ money.

Tail movements apparently change when the cow is in labour, meaning that farmers can be alerted when this is happening, and also if there are problems or the labour is lasting too long.

Several sensors from different manufacturers can detect changes in how much a cow is walking, or monitor eating patterns. This information allows farmers to predict when cows are most fertile, so they can time insemination appropriately. They can also see which cows are eating more and/or producing less milk, and therefore which are most productive. Well Cow, a British company, provides systems to monitor stomach acidity, and help to detect digestive problems. Sensors from Lely, in the US, can monitor the quality of milk.

Sheep are also getting in on the IoT act. A pilot last year in Norway tracked 1,000 sheep in their summer pasture using Narrowband IoT technology. This enabled farmers to check the location and well-being of sheep remotely, without having to go out and round them all up. This pilot is interesting because of the potential for the technology to be used for other purposes.

For example, how many ‘lost cat’ posts or posters have you seen in your area, or on social media, in the last month? Small-scale local tracking devices would eliminate this problem by ensuring that owners could locate their animals, whether dead or alive or locked in a garage somewhere. Microchipping helps, but only if the animal is found and taken to a vet to be scanned — and even then, only if a) it has been chipped, and b) the owner has kept the chip information up-to-date.

Getting value: insurance and analytics

What all these applications have in common is money – but not all financial implications are obvious. Preventing cows from dying means that insurance companies will not have to pay out for the cost of the animal. Keeping sheep healthy saves on vet bills, and again, potentially, insurance. Even tracking pets has insurance implications. As more of us have pets while vets’ bills do not get any cheaper, insurance companies can start taking a preventative approach to animal health by spotting problems before they become serious. Being able to identify small changes in behaviour that predict illness or other potential problems, could enable proactive herd or flock management to prevent later issues.

Alternatively, I can imagine many pet owners (my dog obsessed Mum for one) would certainly use an app to track their pet’s health to indicate potential problems that could occur, much in the way we do with fitbits etc. With the Association of British Insurers announcing recently that pet claims in the UK surpassed one million in 2017, with £ 775 million being paid out, there’s understandably a lot of interest in this topic from Insurers too.

Insurance companies have a long tradition in the use of analytical methods. Where does the insurance industry stand in terms of modern analytics and artificial intelligence? This is what our industry experts found out in personal talks with insurers from all over Europe. Read the survey results.


Unlocking value with analytics at the edge

The key to unlocking value with all these applications is analytics, and particularly if deployed at the edge. Connecting animals is going to create a vast amount of data, but how do you get the intelligence out of that data? We believe that only with “edge analytics” can rapid, actionable insights be identified and used at the point it’s meaningful for the organisation. By “edge analytics” we mean taking analytics directly to the source data as it comes off sensors, so insights can be extracted immediately – rather than data being moved somewhere else to be stored and then analysed.

The partnership between IoT technology and animals may be driven by finance and insurance, but it depends heavily on a ‘third leg’ - analytics so that business benefit to be realised. And let’s face it, who doesn’t want to give a better life to Smudge & Daisy?

Connecting animals create a vast amount of data, save animal lives and therefore farmers’ money and give Smudge & Daisy a better life. #Insurance #IoT #Analytics #MachineLearning Click To Tweet

Why not check out how analytics is helping the World Wildlife Fund to protect animals across the planet, or more broadly how analytics can deliver value to IoT data at the source.


About Author

Adam Goldsmith

Helping Insurers turn Data into Actionable Intelligence

Adam has worked in the enterprise analytics arena for 8 years and has a background in Capital Markets, Banking & more recently in Insurance, specialising in high performance analytical applications. Having spent time at IBM & TIBCO in the past, he has been working in the Insurance world for the past 4 years and since joining SAS in January 2018, he now runs the relationship with some of SAS’ strategic Insurance customers in the UK.

Leave A Reply

Back to Top