Why we shouldn’t overlook AI’s potential to transform the way cars are built
In discussions about the potential of artificial intelligence (AI) in the automotive sector, the conversation often shifts directly towards how AI is paving the way for autonomous cars. And indeed, AI plays a crucial role in the advances of self-driving technology. However, it is equally interesting for car manufacturers to boost their investments in AI to improve how all cars (and not only self-driving ones) are built. AI can deliver benefits across every automotive business function. From a manufacturing perspective, car manufacturers can leverage insights from smart data analytics to reduce rework, accelerate systems and improve production processes.
Robots are not new in car manufacturing plants, of course, but AI-powered machines go even further in the way they improve performance. Because they are not limited to the specific scenarios and rules with which they were programmed, they devise more creative solutions autonomously over time.
Realism kicks in
But AI can’t solve every problem. One of the most crucial success factors for the wide-scale adoption of AI in the automotive sector will be the process of demystifying AI. To exploit its potential to the fullest, it is important to first educate leaders about that potential, invest in training for workers and create the right governance and mindset. Luckily, car manufacturers are increasingly aware of this and don’t appear to be rushing into it. In fact, the pace of adoption has actually slowed down in recent years. According to Capgemini research, the number of automotive companies who have successfully scaled AI initiatives has increased only slightly, from 7% to 10%, since 2017.
Besides having more realistic expectations, many organizations come up against obstacles too, many of which are related to the technological transformation: legacy IT systems, data concerns, lack of skills… All of these challenges have led to a decrease in new pilots, with the number of respondents confirming they were running AI pilots down from 41% in 2017 to just 26% in 2019.
How AI leads to real value
Capgemini estimates that AI at scale could deliver increases in operating profit ranging from 5%, based on conservative scenarios, to 16% in an optimistic scenario. One of the most obvious use cases is the use of AI to improve the quality control process. For example, AI-based object detection can check whether a car matches the specific customer requirements even when faced with about a billion different configuration options. AI-powered hardware can perform visual inspections and provide superior quality control, constantly improving its analysis based on feedback.
A second opportunity for AI is to help avoid equipment failure. AI-based algorithms can digest masses of data from sensors and other sources, detect anomalies, diagnose errors and predict if a breakdown is imminent. The potential impact is estimated at a more than 20% increase in equipment availability, up to 25% lower inspection costs and up to 10% lower total annual maintenance costs.
And there’s more. AI can also reduce forecasting errors by 30 to 50%. That represents huge gains. By incorporating near real-time data from advertising campaigns and other sources, it can truly boost supply chains, allowing them to adapt and adjust to all kinds of changes proactively. Eventually these decisions can even be made autonomously. Think of how AI-powered solutions can automate supply chains by automatically calculating the available space of the warehouse and recommending the best loading options, performing semi-automated checking and loading of parcels into a truck, or using cognitive visual recognition to help check products before receipt in the warehouse.
Last but not least, there is still considerable untapped potential in the use of AI to reduce indirect costs in finance, procurement, marketing and other supporting business functions. The main benefits in these domains lie in capacity reallocation, spending effectiveness and improved accounts receivable processes.
Protecting profits and growth
With the real impact of the corona virus pandemic still unclear, maintaining profits and growth will be more challenging than ever before. But that doesn’t mean that car manufacturers cannot take smart actions now. Leaders should (re)prioritize projects based on business logic and invest in skills and talent. Another option is to reach out to partners to access skills faster. Maybe the risk of a global recession can actually stimulate the adoption of AI in car manufacturing plants, based on realistic expectations and a solid business case. There is more to AI than its ability to disrupt an industry. By taking a step-by-step approach to create scalable AI solutions, car manufacturers can help to create a promising outlook for 2021.