The current global political crisis is already having a direct impact on the food manufacturing sector, with the prices of many raw ingredients reaching new industry highs.

Commonly known as “the bread basket of the world”, Russia and Ukraine together supply the world with around a third of its wheat. The recent conflict in the region has resulted in wheat reaching a record high price increase of over 50%.

Besides wheat, Russia and Ukraine are also major suppliers of corn, sunflower oil, and barley; while Russia is also a significant source of potash – the vital ingredient in fertilizer.

Food manufacturers are now in a tough position, having to deal with drastic price increases of all of these commodities, on top of the existing energy crisis.

So while the seriousness of these price increases is clear, is there any room for optimism?

Let’s break down the problem and explore how food manufacturers can turn this crisis into an opportunity.

The 2 fundamental challenges of the raw material crisis for food manufacturers

Due to the current situation, the agricultural sector in the affected regions is completely unpredictable at the moment. Since farming requires strict scheduling, this disruption will continue to affect the supply chain well beyond the conclusion of the conflict.

In manufacturing terms, we can narrow this challenge down to 2 main problems:

  1. Raw material scarcity. A shortage in supply of raw materials means that manufacturers will have to begin working with different suppliers. This means different types of corn, for example, with different consistencies, quality, etc. This will raise the variance of the raw materials used in production, affecting production and throughput.
  2. Steep rise in costs. While the market has absorbed recent increases in prices, there is a limit, and manufacturers should be aware of consumer behavior with regards to pricing changes. If manufacturers continue to raise prices due to their incurred cost increases, consumers won’t hesitate to switch brands, while general consumer spending will drop, greatly disrupting the economy.

How Artificial Intelligence can solve these challenges

With these 2 major challenges outlined, we can begin to focus on what we can do to mitigate the effects on production, and explore how to improve performance despite these trying conditions.

To deal with raw material scarcity, manufacturers will have to source their wheat, corn, sunflower oil, barley and other raw substances from suppliers they haven’t dealt with in the past. This means greater raw material variance. To make matters worse, they might have to combine multiple types of wheat, or switch between suppliers.

The solution to this is to improve flexibility to suit this new dynamic state of production conditions. In other words, manufacturers will have to improve their ability to adjust their production processes as needed in a timely manner, to cope with the dynamic, unpredictable reality, and to meet demand while keeping waste and quality issues to a minimum.

To deal with rising production costs, manufacturers will have to make significant improvements in overall efficiency in order to cut costs. From the preparation of raw materials to quality control, energy efficiency and everything in between, production should be analyzed in great detail and no stone should be left unturned in seeking out new areas for optimization.

Becoming more flexible to suit sudden condition changes, and seeking out new areas for optimization are both excellent objectives – but don’t they conflict? After all, food manufacturing processes are dynamic, and optimizing for one objective (e.g. reducing costs to offset high raw material prices) can often harm other key objectives (e.g. maintaining or increasing quality and yield, or reducing waste).

One tool that has already demonstrated major value in efficiency improvement and cost reduction in the food manufacturing industry is artificial intelligence. Many leading food manufacturers are already harnessing the power of AI in production, and this will put them in an advantageous position to deal with the current crisis.

AI enables the kind of continuous, multivariate analysis that is needed to optimize multiple, conflicting objectives simultaneously. Food manufacturers can maintain or even improve their existing performance in areas like quality, waste and yield – while also optimizing other areas of their process to drive down costs and remain competitive.

But is Industrial AI relevant if you don’t have data?

While the benefits of AI in manufacturing are increasingly clear, many food manufacturers feel that they “don’t have the data” to implement AI. Fortunately, this is a misconception.

Related content: How Barilla went from zero data, to reducing waste 37% with AI
(in just 4 months!)

The assumption is that in order to implement AI in production, a food manufacturer will already need to have a robust data infrastructure in place throughout the entire organization. So when considering AI adoption, many companies envision a daunting, long-term plan that seems highly demanding on resources.

While data is of course needed for any AI system to work, it’s not necessary to collect huge amounts of data from every aspect of an operation’s activity.

The smart approach to AI deployment is to focus on a specific business problem as the ones outlined above, and collect the data that is relevant for that problem. In this way, a defined data infrastructure can be built that can supply information regarding a specific challenge, instead of attempting to capture all the production data. This method drastically simplifies and shortens the process, enabling production lines to literally go from zero data to successful AI adoption in a matter of months.

At Seebo, we refer to this approach as the Lighthouse Strategy, and it is an excellent way for manufacturers to implement AI in a way that allows for an early ROI (read more on that topic here.)

“AI” can only provide value if the algorithms understand your unique production process

While building a data infrastructure is far less difficult than many food manufacturers imagine, there is one additional condition for successful AI adoption – and that relates to the AI technology itself.

Since manufacturing data is uniquely complex, a generic Machine Learning algorithm won’t be able to make sense of the data in-context. Buffers, loops, dynamic traceability, parallel processes and multi-product lines are just some of the “curve balls” that a typical algorithm will miss, thereby providing out-of-context and ultimately inaccurate insights.

Only an algorithm with embedded process expertise, that understands your unique production process, can provide long-term value. The advantage of “process-based AI” of this type is that it has the power and objectivity to reveal hidden causal relationships and suggest direct action that will lead to optimal performance.

An added advantage is that teams on the production floor now have an extremely powerful tool to help them deal with challenging conditions. Process-based AI is not only an analytical tool, it is designed to be understood by humans – to suggest actions and when to take them in order to maintain a high level of performance.

So for example, if production conditions change – say, due to raw material variance – the algorithm can suggest specific parameter changes that can be made to maintain optimal performance.

Food manufacturers can emerge more competitive from the raw material crisis

So on the one hand, it’s clear that manufacturers have no choice but to step up their game starting now.

But this could also be used as a catalyst for innovation that will in fact improve food manufacturer’s competitiveness in the long-term. Specifically through the use of AI, companies can not only weather this storm, but actually thrive, harnessing sustainable energy, and becoming more flexible to deal with disruptions in raw materials supply, waste reduction, and quality in the future.

Artificial intelligence is by no means an instant “just add water” solution, but with a smart, strategic and focused approach to this challenge, food manufacturers don’t need to undertake a lengthy plan to start reaping the benefits of this technology. And there’s no better time to start than right now.