What F1 racing teaches us about AI for business

Modern business leaders understand how important it is for themselves and their employees to continuously learn based on insights generated from the data at hand. This extends to being able to readjust predictions based on streaming data received in real-time. Adapting theory into practice is challenging for business, but nowhere is this illustrated better than in the fast-paced world of Formula One (F1) racing.

For instance, take the rules around practice. Teams are only allowed a limited amount of practice, which means running simulations becomes an essential component to success. In a sport where every millisecond and performance improvement counts, having the ability to run through 100 000 practice laps with unique scenarios can mean the difference between success and failure.

In the data science era, simulations have become essential for all digital businesses. Examples can range from urban development (infrastructure planning), education (training via simulations), economics (government spending), and others. Nearly every industry sector can benefit from analysing data to predict change and run optimally.

For F1 teams, this means having an immense focus on simulation, data analysis, and data gathering. But to do so requires support from the highest levels of management. After all, those teams who want to remain leaders of the pack must fundamentally understand what makes a car fast. To do that, they need to have the simulation capability to investigate the future, make accurate forecasts and predictions, and use these tools to develop the best car.

Navigating the simulation process

Let’s equate “making cars fast” to making your customer personalisation more agile. In F1, the simulations help with the process, but these are not only for high-paced cars, and the same practices can be used in your supply chain, for example. Here are some ideas of what our strategic partner TIBCO is doing for the Mercedes-AMG Petronas F1 team to keep them in pole position and adapt to their business clients.

Curating raw simulation material (data) is one way that F1 racing cars, which have over 300 sensors that emit data from power steering, power brakes, and gearboxes, learn from analytics. Each practice lap generates millions of data points that can be used to simulate car configuration variations. In every business, every “thing” is loaded with sensors and data. Every website log shows how customers tried to find what they are looking for, and this data needs to be curated effectively.

Hire a simulation team or customer experience experts, people who can build models to test outcomes. In the case of F1 teams, they have dedicated simulation engineers focused on designing and interpreting new tests. An organisation should consider installing a simulation or customer experience team to plan, implement, and optimise digital business testing.

Employ offline simulation or scenario planning. Offline simulation or scenario planning is like running a million ‘what if’ scenarios with different parameters to uncover which combination of actions is likely to yield the best results. F1 teams use stream-based and visual analytics tools to assess offline results that they can then digest and use to influence future developments for future races. If companies are to get value from their simulations, they need to implement offline simulation to evaluate strategic options and anticipate the optimal reactions.

Employ human-in-the-loop simulations or testing of new products and services. Driver-in-the-loop simulation tests the human factor, how drivers might break best-laid plans, and how best to train them on what is to come. Any digital business can steal this idea. For example, emergency responders can use human-in-the-loop simulation to practice reacting. Airline pilot simulation is well-entrenched in that industry. Sales teams can predict the best way to handle objections and improve their ability to choose them.

Democratise simulation data and analytics and let more people in your business use it and experiment with it. An F1 team shares simulation data with drivers, race strategists, the CFO, and the CEO so everyone can sing from one data-driven sheet of paper. By making simulation data visual and accessible, diverse views can help a business team better prepare for the future and collaborate more effectively. The concept of democratising data is growing, especially in businesses that are using Ai-based analytics tools that perform dynamic learning. The more accurate the data and the more people engage with it, the better the AI model responds and directs the outcome.

Data-rich strategies

A digital business can become an industry leader that outpaces all the competition with the right data strategies that connect seamlessly with any data source; unify data intelligently for better control, and predict confidently with real-time intelligence.

The FIA, the sport’s governing body, implemented a new budget cap for the 2021 season. With new constraints, which can include stiff penalties for exceeding the budget cap, it is more important than ever to extract maximum performance from the cars. To meet this challenge, F1 teams use cost visualiser tools to better control costs during a race car’s life cycle.

To build such a tool requires the engineering, finance, and operations teams to work together to understand where they are spending their time and money. Once they understand their biggest cost drivers, they can then use the tool to visualise the information for faster, smarter insights.

Anticipating and predicting costs not only helps F1 teams but is essential in any business that wishes to optimise spending. Clearly, it is not about building the cheapest car but building the fastest one at the most optimal cost. This optimisation is true of many other industries that rely on tight margins and maximum product performance, like manufacturing and retail.

All told, using data to train, predict, and better manage business performance becomes mission-critical in today’s connected environment. The beauty is that there are tools out there, like TIBCO’s Predict portfolio, that effectively use AI and machine learning models that learn as they go. This dynamic learning ensures a cycle of continuous data-driven improvement and lets businesses respond quickly based on what the data reveals.