Artificial Intelligence (AI) is coming to the auto glass and insurance industries. Case in point, Mitchell International is exploring a computer vision application that uses image recognition to confirm repair vs. replace decisions. This is according to the company’s 2017 Third Quarter Industry Trends Report.
Computer vision researchers at Carnegie Mellon demonstrated the ability to detect and understand small movements. Mitchell says in its report, “Instead of an automotive repairer just getting guidance on the next step in a given repair procedure, they could get real time evaluation of ancillary problems detected by computer vision.
In fact, in February 2018, glassBYTEs reported that Autoglass, a vehicle glass repair and replacement company based in the UK and owned by Belron, tested the latest AI technology to assess the severity of vehicle glass damage. The company says this can be used to determine whether customers require a repair or a full replacement.
An article written by Olivier Baudox, vice president, Global Product Management, auto physical damage solutions, authored an article in the Mitchell Report, saying that AI will be used in the future of auto glass claims workflow—first, the concept has to gain traction.
“With rapidly changing conditions that put more drivers and more complex cars on the road, it’s no surprise that auto claim value and loss costs have increased substantially in recent years,” says Ryan Mandell, director of performance Consulting for Mitchell Auto Physical Damage Solutions. He adds that AI is ready to tackle these increased workloads with specific solutions.
Baudoux’s article goes on to say that “once the meaningful data is identified, AI can help to elevate the right information in a way that assists and expedites workflow processes. By leveraging AI and visual computing to analyze photos for example, AI enabled workflow solutions can use machine learning technology to minimize estimate errors and maximize review efficiency.”
In that vein, Mitchell launched the Mitchell Assistance Review Project 18 months ago to accomplish this goal. “By utilizing millions of damaged vehicle photos computers are ‘trained’ to recognize vehicle damage and use computer vision to double check repair versus replace decisions. This will help carriers achieve better estimate consistency, maintain estimate quality and be more selective about sending appraisers into the filed, all while improving cycle times and productivity.”
When did the decision of repair versus replacement taken out of the individual’s personal preference ? I do not want to look at a blemish in my windshield when I’m driving . Being a
repair or not .