No one can deny that the concept of mobility has evolved. There is a revolution taking place. Nowadays, mobility takes on advanced forms such as fleets, autonomous, connected, electric and sustainable vehicles, involving many stakeholders in a multifaceted ecosystem.
Autonomous vehicles, whether cars or trucks, are connected and become part of the Internet of Things. One day, vehicles will be connected to each other, to roads and other infrastructures. Goodyear aims to be part of that ecosystem and plays a major role once tires are on the market, focusing on what goes beyond tires. In this sense, tires are becoming an integral part of the mobility ecosystem within cars and trucks.
Besides developing tires, which remains Goodyear’s core business, the company has adapted its strategy to be at the forefront of tire innovation to enable mobility of the future. To anticipate what lies ahead in terms of mobility for customers and fleets, Goodyear is developing leading-edge technologies as well as products and services. To achieve that, the company has set itself a bold goal: reinventing tires and services for all upcoming products by 2027. As part of the vision, the aim is to be at the forefront of transforming tires into intelligent systems for cars.
Using data, Artificial Intelligence (AI) algorithms, and sensory data, the company aims to deliver unique products and value-added services that enhance the customer experience and generate value for industry partners, as well. A smart strategy cannot be implemented without equipping tires with sensors, not even without designing them. This is what Goodyear plans on doing.
This broader vision allows Goodyear to develop technologies not only for incumbents, OEMs (Original Equipment Manufacturer) which it already works with, but also for disruptors, startups and companies in the automotive sector that challenge the status quo (e.g., companies working on autonomous vehicle platforms).
Several years ago, Goodyear began building a team bringing together new skills around AI, sensors and cloud computing. This will enable the company to develop in house these cutting-edge technologies for intelligent tires.
Furthermore, developing new AI products requires a well-defined process in a step-by-step manner:
- 1st step: Data collection and gathering in compliance with data protection regulations. Data are the foundation on which to generate algorithms. In order to ensure access to a comprehensive set of data, this process requires a strong collaboration with the sensor development team and external stakeholders like fleet managers.
- 2nd step: Analyzing data. Assess quality and quantity of the collected data to have a good representative set of data on which the AI model can be built.
- 3rd step: Building models. Depending on which algorithms and types of solutions are available (prediction, recommendation systems, optimization, etc.), the data scientists will decide which approach to take.
- 4th step: Developing a proof-of-concept (POC). The POC will be shown to the team developing the product and to customers as well. POC allows for ensuring that the model meets the requirements in terms of performance and accuracy.
- 5th step: Scaling-up. From the POC, make an upscale version using cloud technologies. The cloud is an important piece of the puzzle as well to deliver the solutions to the customers. From the POC, the team will adjust the model, the algorithm, to make it fit for different types of conditions under which it will be used, e.g., a high number of vehicles, different types of sensors, different roads or weather conditions, etc. and become a real product that can be sold to customers.
AI algorithms developed within Goodyear are applied to some use cases which represent business opportunities.
- With autonomous cars, safety and performance are of utmost importance. Using tire data, Goodyear is building out AI algorithms aiming at informing the vehicle about road conditions. As a matter of fact, a human driver has a lot of perception power, making decision when looking outside if it is snowy, icy, to accelerate or break slow. However, the decision to be made by autonomous cars needs to be automated. For that, Goodyear AI algorithms are taking in all the information available which includes data from 3 sources – tire sensors, vehicle sensors and external weather information – then combining them in an efficient manner and helping the virtual driver decide what the limits for today’s driving conditions are, assuming that the road conditions are changing, e.g. what maximum acceleration should be permitted, what distance the autonomous car should keep from the car in front of it when driving, what maximum speed the car should be going at when making a turn. Goodyear algorithm gives the car the ability to “sense” what is happening at the tire level and provides an extra level of confidence for some of the applications. Several car makers are very interested in adding this extra level of intelligence to the cars. Indeed, these cars often come with camera systems which are looking around the car, but cameras have inherent disadvantages also when the weather conditions are quite bad or when it is very dark. Likewise if the sun is glaring on the camera, it gets blinded quite easily.
- With fleet management, minimizing downtime is key for cost saving and revenue benefits. Goodyear develops predictive AI algorithms collecting data and providing information on the health of the tire, enabling then to assess what is the likelihood of the tire or the vehicle breaking down and proactively informing the driver and the fleet manager. The definition of the health of a tire is quite broad and can be around pressure monitoring. For example, has the driver picked up a leak? If yes, has the driver picked up a puncture and is the tire leaking? If yes, how much time does the driver have before it becomes a critical issue? Another example is when driving through a pothole which generated big impacts. Is the tire still in a good state or does it need to be changed immediately? A third example is estimating what the remaining tread depth is on the tire without physically measuring it. Often, with a fleet of 300 vehicles, people don’t have the time to check it every day. Goodyear’s algorithm can predict what the remaining tread depth is on the tire based on the driving pattern and the vehicle and other parameters. It can also make some intelligent recommendations. If winter is approaching, it could not be safe but risky to drive with tires with low tread depth. With the help of the algorithm, intelligent recommendations will be made to change the tire sooner and not wait until the last day to change tire. This will help the fleet remain safe on the road by minimizing downtime, which represents money loss. Goodyear offers a suite of tire intelligence technologies, which powers solutions for today and the future and Goodyear SightLine is the company’s global tire intelligence platform (www.goodyearsightline.com).
- Goodyear wants also to set up a recommendation system, adding a layer of AI, which will enable the customer to pick the right tire. The AI model could base the recommendations on information such as place of living, weather conditions, driving style, etc.
Taking advantage of its almost 125 years of experience and knowledge in tires, which represent a fundamental and strong competitive advantage over new players, Goodyear now develops « physics-driven AI models’. The methodology consists in embedding tire knowledge into the AI models.
It is as well worth adding that, over the past five years, certain enabling technologies valued a dramatic improvement, e.g. chips are becoming cheaper and much more efficient, more computing power being available on the chips. Sophisticated algorithms can now be run on the tire chip itself. Therefore, the AI algorithms can also run embedded in the tire sensor. Thus, sensor data can be consumed in 2 ways. A first option is to process it in the tire sensor. Cellular connection has some latency, depending on the application. Autonomous cars can’t afford to have any delay. Process must be fast, and answer must be given in milliseconds to the control systems. There, Goodyear is working on embedded solutions. In a second option, the data is streamed into the cloud and the AI algorithm is run there. As for certain applications like fleet management, Goodyear relies on cloud solutions since a certain amount of latency is acceptable when tracking fleets. In certain situations, cloud connectivity may not be available, so both must be considered.
As mentioned already, delivering solutions to customers relies on cloud computing technology. It is involved primarily in the scaling up of AI algorithms into a product, whether it is a software, a mobile app, or a web application. Using the cloud, released algorithms can run at large scale, on many vehicles or fleets for example, securely and stably. Being agile through the cloud and reducing time to market to weeks or days rather than years is the foundation for a robust digital product life cycle.
The company is focused on establishing itself as a leader in the tire industry’s digital products and services. Goodyear aims to make the tire buying experience as satisfying for the customer as possible, from the purchase of the tire to its replacement.