Fear of flying: Assessing drone risks

Drones have rocketed in popularity over the past decade with the use of both commercial and leisure drones taking off. Analysts are predicting that there could be more than 67 million shipments of drones—or ‘unmanned aerial vehicles’ as they’re also known—by 2021.

Andrew Chadwick from Frazer-Nash Consultancy argues that their increase in popularity has had three main drivers: usability, usefulness and cost.

“They are relatively easy to operate, have an increasing variety of applications, from aerial search and rescue to security monitoring, and the basic models are relatively affordable. Plus, of course, they’re fun,” he says.

A useful tool

The insurance sector has been one of several to start to realise their benefits. Companies are able to use drones to help assess claims, for example where there has been damage to a building’s roof or chimney, or after a fire. Graeme Anderson from Frazer-Nash Consultancy notes that not only can the drone’s aerial photography or video show the extent of the damage, it can reduce the risk to the assessor who might otherwise have had to climb a ladder or hire a helicopter to perform an inspection.

“This speeds up the assessment process for insurance companies and the claims process for customers. Drones can also be used to inform underwriting, by providing data on an asset’s condition,” Anderson says.

They also have risks and drawbacks. The upsurge in their use has already led to an increase in incidents where drones have collided with infrastructure, including power lines, trains, cars and aeroplanes. Airport runways have been forced to close and flights diverted when drones were spotted flying nearby, leading to costly delays.

Anderson notes that one drone is reported to have crash-landed on to a nuclear power station site; another is said to have collided with power lines in China, resulting in an electricity blackout that took six hours to repair.

“We’ve seen a camera drone hit by a car at the 2017 World Rally Championships, while another nearly landed on a World Cup skier in 2015,” he says.

Where there is risk there is potentially an opportunity for insurers. This is not a simple one, however. Chadwick notes that the ‘disruptive’ nature of the drones’ developing technology and regulation can make underwriting the risks associated with them extremely complicated.

Graeme Anderson He explains that commercial drones, and those weighing above 20kg, require insurance in the UK, so depending on the sector, you might be required to cover public liability risks if an insured photography or surveying drone causes injury to a person or damage to property. Equally, coverage might be needed to insure manufacturers and suppliers for product liability against failures in the drones’ control technology—failures could compromise the safety of the craft, or potentially leave them open to cyber attack or ‘drone-hacking’.

“You might even be required to indemnify against claims such as invasion of privacy,” he notes. “Where drones are being used for commercial purposes the resulting damage could be extremely costly—for example, a drone surveying an oil pipeline or electricity pylon could strike and damage the asset it’s examining, causing a knock-on effect to power and fuel networks.”

Anderson adds: “Whether you’re offering commercial or leisure drone insurance, any operator—even one who’s been trained—can make a mistake, or a technical fault can cause them to lose control of the craft.”

Modelling the risk

There are ways for insurers understand this risk better. Chadwick says that through using modelling and simulation, insurers can predict the possible outcomes of different types of incident, and can quantify the potential risks of the different types of drone impact, which can inform their underwriting decisions.

While very few studies have been undertaken to assess the effects and consequences of drone impact, so the potential risks aren’t well understood, modelling is a powerful tool that can help all the relevant stakeholders to understand what the possible outcomes might be, what the probability of impact is, and thus clarify the level of risk, Anderson says.

“It can be used to simulate the impact effects of a variety of drone types striking a range of different objects: from buildings, to aircraft, to people,” he says.

Chadwick explains that the modelling and simulation techniques his firm has developed are based on methods used for the aerospace industry, to analyse the effect of bird strikes on aircraft.

“It can identify whether the drone is likely to cause substantial damage to the object it hits, or if it’s more likely to destroy itself. It can also pinpoint where impact might cause the most problems, enabling these to be planned and minimised—for example by installing extra shielding.

“For key critical national infrastructure, these mitigations may even extend to measures that detect drones’ approach, through tools that monitor their wireless or radio frequency signals, or ultrasound transmissions,” he says.

Anderson adds: “Modelling is quick, inexpensive and flexible. It allows insurers to examine and compare a range of different scenarios, and the models are then analysed to define the risks of the impact, including any danger to life.

“By performing some physical testing on objects and materials the models’ predictions can be validated and, once validated, they can be used to look at any number of different impact conditions: angles, masses, velocities, locations.”

For example, modelling could predict what might happen if a drone crashed into the glass window of a building—a high-rise office block or shopping mall, for example. Knowing this risk, and its probability, would allow the business to develop actions to minimise the outcome of any collision, or to produce operational guidance for its staff to advise the steps to be taken during such an incident.

Chadwick explains that, financially, producing a risk profile of the impact a drone collision could have on a business’s assets can help to inform the level of investment needed to protect them, and enable it to manage the risk accordingly.

Regulations pending

The UK government ran a consultation on the safe use of drones from December 2016 to March 2017, including a specific impact assessment paper on insurance for drones. In this, it recognised that the regulations that established drone insurance requirements were created in 2004, and were generally aimed at the aviation sector.

The House of Lords recommended in 2015 that the minimum amount of public liability cover required by commercial drone operators should increase, and the Department of Transport has been given feedback by operators on the cost of policies and the levels of cover available.

The government is now in the process of analysing the consultation responses, but it seems likely that new legislation on drones will be enacted at some point in the future to reduce the potential risks.

Anderson says that regulations are almost certain to increase and will provide some guidance to the insurance sector. But he believes it’s unlikely to quantify the levels of risk that insurers are taking on when they insure specific drones for particular activities.

“The risks may even be overestimated, and modelling drone impact may show that they are actually lower than expected, and that it is more likely that a drone will be damaged than the object with which it collides.

“But until the levels of risk are understood, the potential effects won’t be comprehended and insurers will be forced to use ‘best-guess’ assessments that may have both business and financial impacts,” he says.