Advanced data analysis and machine learning will allow OEMs to analyze real-world data on how vehicles behave under specific driving conditions, writes Ron Soriano
Thirty days in June, as the old saying goes. But in June of 2022, automakers issued 31 recalls in the United States, making an average of more than one recall per day. The recalls were made by nearly all foreign and domestic automakers — from Fiat-Chrysler to Hyundai to Porsche and Lamborghini — and the individual recalls included numbers ranging from 2.9 million vehicles to just one, on issues from hardware safety to software problems in hardware devices. Vehicle computer. It’s just been a typical month for automakers, which have been issuing recalls at the same pace for years now.
Is there any way out of this recurring summoning dilemma? How is it that after building cars for nearly a century, automakers still aren’t making them right? But a solution, or at least a partial solution, is emerging: modern technology in the form of artificial intelligence (AI) will in the future be able to help manufacturers build better and safer cars, reducing the likelihood of having to issue recalls. Through advanced data analysis and machine learning, OEMs will be able to analyze large amounts of real-world data about how vehicles behave under certain driving conditions, taking into account the impact of weather, road conditions, driver habits, wear and other factors that can influence on the performance of the vehicle. Although there are many challenges, including production process modification and privacy concerns, OEMs must take more steps to incorporate this data into vehicle design and construction.
How is it that after building cars for nearly a century, automakers still aren’t making them right?
The first step is to take advantage of and use the big data collected by a large number of sensors in modern vehicles, especially with the advent of connected autonomous and semi-autonomous vehicles. This, along with information on weather, traffic and the condition of the road itself, as well as data collected by repair shops on specific physical and operational problems, can provide valuable insights into the vehicle’s function and performance. This will allow manufacturers to better understand how to avoid issues that may cause a recall. They can use this data and factual information to help them with things like designing better parts or the best way to write or upgrade vehicle software. Ultimately, data-driven automated production systems can rapidly change manufacturing processes to improve products coming off the assembly line.
And while this is a vision for the future—most cars haven’t yet carried the advanced set of sensors needed for this kind of analysis—smart AI systems are already doing such predictive proactive analysis in a variety of fields, from medicine to machine repair. Of course, OEMs already use some data for these purposes, but the advantage of advanced data analysis systems is that they can engage in machine learning, honing their knowledge of what makes a car work — and what might prevent it from working — to build a model that OEMs can use To help get rid of problems.
It is already clear that data analysis works. In 2012, General Motors used a database that tracked parts used in its cars and collected manufacturing records from suppliers in order to track down the defective part on some Chevy Volt models. As a result of the investigation, GM was able to avoid a mass recall—bringing only four volts to service, for NHTSA approval. It took GM investigators a month to analyze the data in order to come to their conclusion — and in an era before the proliferation of sensors, applications and other data-collecting sources, at a time when AI systems were less advanced than they are now. If GM was able to reduce recalls a decade ago, current technology should be enough to avoid recalling tens of thousands of vehicles and saving companies millions of dollars. The data can be used to improve the manufacturing process, and reduce the number of recalls in general.
But analyzing AI data to improve engineering and processes has yet to become a standard among OEMs. While manufacturers are already using AI in some production processes, OEMs still have to build systems that can quickly act on data collected from a large number of data sources, including connected vehicle sensors, potentially resulting in to interruption of the production process. Thus, along with AI systems, OEMs will need to invest in automated systems to work on data and quickly pivot production processes to prevent production problems.
In addition to this logistical challenge, the McKinsey report attributes the slow adoption of AI analysis to several factors, including the traditional culture of the auto manufacturing sector, where data is often siled; Few, if any, OEMs have been able to develop dedicated multifunctional monetization modules that can effectively leverage AI-generated data to change production and engineering systems. OEMs are also struggling to recruit the talent needed for advanced data analytics, and they struggle to partner with outside organizations, which is essential to truly benefit from the data. In addition, OEMs will need permission from consumers, many of whom are not interested in giving away data on driving habits or vehicle condition.
Besides AI systems, OEMs will need to invest in automated systems to work on data and quickly pivot production processes to prevent production problems.
However, the data that OEMs can collect is too valuable to be overlooked, and once manufacturers develop the correct methods for collecting and using data, they will be able to protect themselves from major problems, and identify design and mechanical problems that occur more quickly. . The data collected can include details of the condition of the parts when the vehicles are maintained as well as their condition after an accident or other accident. For example, if repair shops find that 60% of the fender flares cause the passenger side mirror to break, this may indicate that the way the vehicle is made makes it more suitable for such damage.
Manufacturers can also use a data-driven design approach to increase consumer confidence. Research shows that large or highly advertised recalls hurt sales of not only specific OEM nameplates, but even vehicles manufactured by their competitors in the same country; A Suzuki recall, for example, will affect Subaru sales as well. By identifying and addressing problems before a mass recall is needed, OEMs can show consumers that their quality control is good enough to catch and fix problems before they get out of hand. In addition, it increases consumer confidence in the brand in the used car market, dispelling persistent concerns among buyers that dealers do not always ship recalled cars to the manufacturer, but instead try to sell them as used vehicles.
Big data has had a huge impact on dozens of industries, and it is time for OEMs to use big data to improve the manufacturing process, as well as increase consumer confidence in their brands. Fortunately for them, a lot of the data they need to analyze is already being collected and used for various purposes; All they need now is to integrate it into the production process, and implement systems to work on it quickly. Why not use that to save themselves – and consumers – the trouble of having to deal with paybacks, and ultimately produce better and safer cars?
About the author: Ron Soriano is Vice President of Operations at Raven AI