Taking the pain out of auto insurance in India using AI

Auto insurance is becoming a lot easier and quicker – with the help of artificial intelligence and smartphones.

India is one of the world’s largest and fastest-growing auto insurance markets – but until now, the sector has had to rely on traditional ways to renew lapsed policies or make repair claims. Both services have required inspectors to physically look over vehicles and make damage assessments.

In a country with more than 230 million vehicles and 1200 auto accidents every day, scheduling inspections and getting approvals can keep cars and policyholders off the road for days or longer. A more convenient way was needed.

To ease the pain, ICICI Lombard has partnered with Microsoft to develop India’s first AI-enabled car inspection feature in its mobile app, “Insure.” The app allows customers to buy or renew policies anytime, anywhere. And, soon it will also simplify the process of making a repair claim.

In case of lapsed policy instead of a physical inspection, customers can simply take images of their vehicle and upload them with Insure. The app then uses artificial intelligence (AI) and machine learning to divide the images into frames and identify the various parts of the car to look for damage. In most cases the AI module can make a judgement very quickly, reducing the time from days to mere minutes.

Proving it’s possible

“We had been envisioning something like this for a long period of time,” says Girish Nayak, Chief Customer Service, Operations & Technology, for ICICI Lombard. “We were not really successful working with some start-ups. Therefore, we approached Microsoft to see if we could do this together.”

The two companies started collaborating where software engineers from both companies met and worked on a prototype.

“The prototype gave us confidence that something like this is possible,” said Nayak.

Real time renewals of expired policies makes the customer experience consistent and convenient.

Using the Azure platform, computer vision, and machine learning (ML) technologies, they refined the process until the accuracy was fit for purpose.

The feature was launched in December of 2018 and has worked as expected. Nayak said, “real time renewals of expired policies makes the customer experience consistent and convenient.”

With the app, insurance personnel are more productive as they no longer have to be physically present to inspect vehicles and can focus on less repetitive tasks. With improved efficiency, customer satisfaction has also increased considerably.

According to a new study by IDC Asia/Pacific and Microsoft, “Future Ready Business: Assessing Asia’s Growth Potential Through AI”, organizations in India that have adopted AI have seen tangible improvements of 22% better customer engagement and 21% more productive employees. Further improvements of at least 2.1 times are forecasted over a three-year horizon.

Expanding AI in auto insurance

Renewing policies online has proven to be successful — the company is currently processing between 150 to 200 cars per day — but as Nayak pointed out, using AI for renewals is just a “stepping stone” for the company. AI is a major thrust for the company, and they are very near to using AI to process repair claims as well. The solution is currently in beta stage and is expected to be launched in 2019.

AI will soon assess images of damage uploaded by the customer and provide an estimated repair cost within seconds

When a customer’s vehicle is damaged in an accident, they must surrender it to a service center so a service engineer can assess the damage and provide an estimate for repair. Then, an insurance personnel examines both the car and the estimate, and either approves, rejects or modifies individual parts of the estimate.

Compare that to the new, AI-driven process; a customer will use the app to take photographs of car’s damage. Once uploaded, the system’s deep learning model and computer vision identifies in real time all the parts of the vehicle, like roof, window or bumper and then spots all the different types of damage – be it scratch, dent, crack, and so on. Most importantly, the app replies with an estimated cost quickly using historical data.

With AI, the company aims to process over 80,000 simple claims every month with same-day turnaround when the module is launched towards the end of 2019.

Improving customer experience and business outcomes

The advantages to the customer are clear. They can file claims whenever and wherever is convenient and will receive estimates much faster than before.

The app is also valuable to ICICI Lombard from a business perspective. Automating the process reduces the possibility of inaccurate assessments due to human error. And, increased efficiency and productivity improves the bottom line.

AI is expected to free up insurance inspectors’ time to focus on complex claims like head on collisions and spend quality time with vehicle owners to help reduce their stress

The role of the human insurance inspectors is changing as well. AI is good at quickly handling the routine claims. That allows insurance inspectors to attend to more complex claims where their experience really matters, like those that involve lot of interior damage, for instance.

Nayak said inspectors are being trained in new skills. “With AI coming in, it frees up their time to prioritize complex cases and personal customer interaction, as well as providing career paths in learning new techniques as well.”

With the launch of the app, the company aims to substantially increase employee productivity and customer satisfaction.

As AI remakes their auto insurance business, ICICI Lombard is keen to employ the technology wherever it aligns with their vision of increasing efficiency and improving customer experience. Currently, they are working towards launching similar AI-powered apps for their health and marine insurance customers.

Article, images and video kindly supplied by Microsoft Asia News Center.

Likes
Please feel free to share

Be the first to comment

What do you think?

This site uses Akismet to reduce spam. Learn how your comment data is processed.