Inventory management challenges – the role of analytics and AI

While this helps them avoid interruptions to the assembly line, it means tying up working capital in inventory – a costly and wasteful strategy that truck OEMs, in particular, need to avoid in the longer term. Fortunately, more robust ways to add resilience are available, and newer technologies such as generative AI open up exciting possibilities. Fredrik Almhöjd, Director, Automotive & Manufacturing at Capgemini and the company’s go-to-market lead for commercial vehicles in the Nordics, and his colleague Sven Lierzer, Senior Manager, eMobility & Commercial Vehicles, discuss some options.

Supply chain resilience has been a huge issue for commercial vehicle (CV) manufacturers in recent years. During the pandemic and subsequent disruptions, OEMs experienced component shortages that had a significant impact on production and deliveries, and hence revenues and costs. Now, securing the supply chain is a top priority for leading truck companies.

In our last blog, we introduced a recent CRI report on automotive supply chain trends. This report takes a deep dive into the automotive supply chain to help companies across the industry better understand the issues they face. It reveals that faced with these supply problems, companies have often shifted from just-in-time to “just-in-case” approaches to inventory. In fact, this emerged from a survey conducted for the report as the most common approach to semiconductor supply chain issues.

In other words, many companies have been adding resilience by stockpiling parts, which means tying up working capital. It’s also undesirable from the automotive sustainability point of view since it results in resources standing idle, and at worst never being used at all.

Time to find better resilience strategies

With capital costs increasing, stockpiling components in this way is a risky strategy – particularly for CV manufacturers, we’d argue. For one thing, truck businesses operate on low margins, so cannot afford the extra cost of that unnecessary inventory. For another thing, truck customers are exceptionally price sensitive – and this extra cost will, sooner or later, affect pricing.

OEMs are already realizing that they need to pursue other strategies to increase resilience, and have taken some steps to do so. As we mentioned in our previous article, the CRI research revealed that procurement from offshore locations has declined by as much as 22% over the past two years. This change should make the supply chain more resilient and reduce the need to stockpile parts. But on its own, it’s not enough to ensure the levels of resilience manufacturers need.

What else can truck OEMs do to increase resilience without tying up capital?

To see what other options are available, let’s revisit the recommendations from the new research report, summarized by the figure below.

A couple of options in the figure seem to us to be especially relevant to this issue.

Improving collaboration and trust

The fourth bubble is about building trust to improve collaboration with suppliers. A contributory factor to the supply issues seems to be that suppliers often mistrust OEMs’ forecasts of their requirements, largely owing to a perception that OEMs overstate those requirements. Increasing transparency will help to overcome this issue.

Part of the solution to this need for better collaboration and trust could lie in the development of logistics platforms to provide reliable connectivity and visibility along the supply chain. Some OEMs have been experimenting with this type of platform, in some cases relying on blockchain technology to assure traceability. As mentioned in our previous article, initiatives such as Catena-X can also help to improve communication by standardizing the way data is used.

But even with great connectivity and communication in place, transparency is not so easy if a manufacturer cannot accurately forecast what components it will need at a given point in the future – and that brings us to another relevant recommendation.

Adding intelligence to the supply chain

By leveraging technology to make the supply chain intelligent and data-driven – the third bubble in the figure also touched on in our last article– systematic forecasting becomes possible. Tools that harness the latest AI techniques and advanced analytics can be used to build an accurate picture of future demand that enables manufacturers to share precise information with their suppliers about what they will need.

To achieve this intelligent, data-driven supply chain, it’s essential that CV businesses mature to the point where they fully understand the benefits of using advanced analytic tools, including AI-enabled ones. Talent and culture are key enablers of that maturity, as indicated by the first bubble of the figure.

Adding intelligence to the supply chain will further enhance trust, and ensure that manufacturers get the necessary parts without precautionary stockpiling. The need for better forecasting is discussed in more depth in a recent Capgemini point-of-view report, which focuses on the issue of resilience in the supply chain.

Analytics and AI to the rescue

As noted above, when it comes to this task of adding intelligence to the supply chain, advanced analytics and AI have a key role to play.

For example, building on historical data, market trends, and other information, AI-enhanced predictive analytics can help companies to accurately forecast future demand for vehicles and hence components, and continuously adjust those forecasts in the light of additional information. This way, a CV manufacturer can successfully move back from just-in-case to just-in-time inventory management, avoiding tying up excess working capital in stock.

Beyond analytics, AI can make the work of CV-related inventory management easier and faster. Generative AI – the new and glamorous kid on the block – could play an important role, most obviously thanks to two abilities: First, it can digest large amounts of information into a convenient form, and second, it can engage in natural-seeming dialogues with humans.

A genAI-supported virtual supply chain assistant could provide an intuitive, one-stop entry point enabling users to access supply chain data of all sorts (including demand forecasts, inventory insights, stock levels, supplier routes, network information, and delivery times).

GenAI could also contribute to a range of user-friendly supply chain facilities, from tendering assistants for procurement (complete with supply quality evaluation) to automated tools for ensuring the regulatory compliance of parts and associated documents.

For more insights, please see Capgemini’s reports on Harnessing the value of generative AI: Top use cases across industries. You might also be interested in The Art of Software: The new route to value creation across industries, which discusses the productivity benefits of genAI-assisted code development. We hope to return to the topic of GenAI, and AI generally, in a future CV blog article.

Conclusion

Of course, none of these recommendations can be followed without having the right skills and capabilities available, and therefore the recommendations in the first and second bubbles of the CRI report, about talent, culture, and workforce management, are highly relevant too.

For a single company to pull all of these levers is a big ask, but Capgemini is waiting to help you by providing or locating capabilities to complement your own. This is possibly the fastest way you can equip your company to deal with future supply chain challenges in the automotive industry.

Please get in touch if you’d like to know more about Capgemini’s views on how the automotive supply chain must evolve – or, more specifically, about our approaches to inventory management for truck companies.