Nadja De Maeseneer
ChatGPT as our Virtual Panelist, Part II: Generative AI in Automotive - Myth or Opportunity?
Today we continue our virtual panel discussion with our industry experts along with ChatGPT generated responses.
Mark Seng, Aftermarket Expert and VP of Business Development at Predii recently spoke about the latest trends at ETI Tool Tech 2023. He outlines three major disruptions caused by AI: the use of insights from historical data to provide smart products and solutions; the democratization of expertise, with AI providing a platform for sharing knowledge among technicians, service centers and customers; and the use of AI-powered predictions to ensure the right part is in the right place ahead of time, reducing inventory costs and increasing supply chain efficiency. Lastly, Generative AI continues to fascinate the AI community - with good reason! Enjoy reading.
Nadja: Mark, AI-based solutions continue to disrupt the Automotive Service Industry. You just came back from this year’s ETI Tool Tech event where you were one of the speakers. Can you share with us what your presentation was about?
Mark: Yes, of course. First, I want to thank ETI for the invitation to address the conference regarding how AI is being leveraged to improve the automotive repair service experience. It was Predii’s first time attending as a member, and I thought it was a great event.
Artificial Intelligence is becoming more and more prevalent across just about every industry and is affecting everything from planning, production, supply chain management, and product development. The Automotive Service Industry is no exception.
The Automotive Aftermarket repair experience is an extremely complex ecosystem. With nearly 300 million vehicles in operation and over 200 million repair orders annually in the US alone, there is a huge amount of data that can be leveraged to optimize operations or power data-based products. Unfortunately, much of this data is messy and not actionable, which is where AI comes in.
At Predii, we have observed how AI is impacting the Automotive Service Industry for nearly a decade. Generally speaking, we are looking at 3 major disruptions:
Insights from historical data are key to powering smart products and solutions
Democratization of expertise changes the dynamics between shops, technicians, and customers
AI-based prediction enables ‘right part, right place, ahead of time™’
Nadja: Thank you, Mark. I would like to get back to each of these three disruptors in a bit. Before we do so, let me ask ChatGPT what they think. ChatGPT, please look at Mark’s expert statement about AI disrupting the Automotive Service Industry and add your thoughts to 1,2 and 3.
Insights from historical data: The use of AI in analyzing historical data can provide valuable insights into the performance of vehicles and their maintenance needs. This information can be used to create predictive models that enable the early detection of potential problems, reducing repair costs and downtime. The integration of AI-powered predictive maintenance can help increase the efficiency and accuracy of service operations.
Democratization of expertise: AI can democratize expertise by providing a platform for sharing knowledge and best practices among technicians, service centers, and customers. This can help level the playing field and promote fair competition among service providers. Customers can benefit from access to a wider pool of expert knowledge, leading to improved customer satisfaction and loyalty.
AI-based prediction: AI-powered prediction can enable the right part to be in the right place at the right time. This can lead to reduced inventory costs and increased efficiency in the supply chain. Service providers can benefit from improved operational efficiency and reduced downtime, leading to increased profitability.
Overall, the use of AI in the automotive service industry has the potential to bring significant benefits to service providers, customers, and the industry as a whole. However, the adoption of AI requires careful planning, implementation, and training to ensure that it is integrated seamlessly into existing operations and yields the desired results.
Nadja: We might need some panel training for our friend here - no introduction at all - but all in all, this sounds pretty reasonable to me. Mark, what do you think?
Mark: Absolutely. I would like to add little more depth here. Let me start with historical data. Harnessing the power of historical data provides a strategic advantage. Prediction of repair demand - as ChatGPT mentions - is one dimension. I would like to highlight another, much more ‘hands-on’ use case. Today, nobody knows which vehicle is coming into the shop on any given day or what repairs may be needed. But whatever vehicle or whatever repair it is, the shop has most likely seen it before, heard the symptoms before, done similar diagnostics, ordered the same parts, and performed the same repair and related maintenance before. The challenge is - will you remember all of this for every situation? You might not, but an automotive-specific AI solution surely does. By tapping into historical data it can augment the process of making the right diagnosis, ordering the right parts and performing the correct repairs to solve the problem the first time and not have the driver bring the vehicle back again the next day. Basically, AI is being used to leverage intelligence from large amounts of historical repair data to make inexperienced technicians act like a veteran mechanic.
Very closely related to this is what we see under ‘Democratization of expertise’. Not only technicians benefit from AI-powered expertise, but so do consumers. Repair and maintenance information is becoming more accessible. Personalized insights and recommendations based on a customer’s specific vehicle, mileage, and geography, increases transparency and helps build trust between shops and the consumer. Consumers who can see these more personalized repair and maintenance recommendations - and perhaps see them even before dropping the vehicle off at the shop - are more likely to agree to the additional recommended work, leading to increased revenue potential for the shop.
Nadja: This sounds pretty much in line with ChatGPT’s ideas! I’m curious about the last disrupter. Right part, right place, right time is a known pattern - now we’re going further and placing parts where they need to be ahead of time?
Mark: Exactly. And this applies to both maintenance and general repair - for predictive maintenance or even avoiding vehicle breakdown events. Specialized AI allows us to leverage data across multiple sources - repair orders, parts transactions, customer concerns - even connected-car telediagnostic data when the consumer gives permission - and enables shops to predict maintenance and repair needs well before the customer sees any type of warning light on their dashboard or brings in their vehicle. Think about it - it’s a game changer! We’re taking our aftermarket goal from right part, right place, right time – to right part, right place, ahead of time™.
Nadja: Pretty much the nirvana state of any industry - knowing demand ahead of time. We briefly touched on this scenario in our last conversation so I am not going to go into too much detail here. I am curious about your thoughts though as to how realistic this state is. Are we looking at a concept? Ideas? A solution?
Mark: Predicting repair issues before they happen is way more than a concept. We have all the puzzle pieces we need today to tell a shop or dealership - or even customers themselves - when the vehicle is likely to experience a serious issue. As often, it comes down to data. Whoever has access to the best data and can see the patterns and trends in the data has the advantage. AI will take care of the rest.
Nadja: But not just any AI. I heard you say ‘specialized AI for Automotive' a couple of times - in out last conversation and today. I would like to go even further here and explore the idea of generative AI in Automotive. What are your thoughts?
Mark: Well, I will leave the details of this conversation to the AI experts in our circle - maybe in a follow-up discussion with Tilak Kasturi or Tirthankar Ghosal from our Predii Labs Team. I will say that the idea of generative AI in Automotive is very interesting. Generic solutions such as ChatGPT can provide some level of insight into repair procedures or recommendations, but it is not very good at more personalized or vehicle-specific recommendations based on how I drive my car, where I drive my car, or the combination of DTC codes my vehicle is reporting on any given day. In addition, ChatGPT is open AI and not a secure environment.
The automotive-service industry is still very human-centric, with the brain of technicians being essential for the work. Again, ChatGPT can provide generic answers for repair and maintenance inquiries but cannot provide personalized repair information for a specific VIN, mileage, and set of codes. Generative AI solutions in the Automotive Service Industry become interesting when they can support more complex, vehicle, and enterprise-specific use cases.
Nadja: So, is Generative AI in Automotive a myth or an opportunity?
ChatGPT: Generative AI in the automotive service industry is not a myth, but rather an opportunity that holds great potential for improving diagnostic accuracy, optimizing maintenance schedules, and providing personalized repair recommendations.
Nadja: I have a feeling, we’re not looking at an entirely unbiased answer here. Thank you, Mark, for your time - always great hearing your share your knowledge. I will make sure we’ll have a follow-up on Generative AI in Automotive Service. Stay tuned!
We’re happy to share summary of Mark Seng’s presentation at ETI upon request.