Marcel Koks is the Industry and Strategy Solution director for the Food & Beverage sector at Infor, a global business cloud software company. With more than 20 years' experience in working with companies implement enterprise applications for resource planning and supply chain management, Koks shares his insights with Food & Drink Business on how AI can help businesses improve process productivity and efficiency.
- What potential do Artificial Intelligence (AI), Machine Learning (ML), and Robotic Process Automation (RPA) hold for driving profitability in food and drink businesses, and how can companies overcome perceptions of these technologies as bottlenecks?
We asked some of our prominent customers in the food and beverage industry to complete a survey with questions about the adoption of AI, ML, and Enterprise Automation (EA). Thirty percent of the responding companies said they have planned their journey to autonomous business processes in the next three years.
Sixty per cent think that autonomous business processes are achievable but haven’t detailed their plans yet. None of the respondents said that they were missing enabling technology. All these customers run cloud-native Infor applications that come with a digital business platform with Data Fabric, AI and RPA.
I think that 90 per cent willingness to adopt these innovations in their business-critical operations is very high. They said the bottleneck was quality of data, but that’s really a chicken or egg situation. ML can help assess the quality of data, it can analyse much more data and data elements than any human can.
It’s a totally different approach than in the past where people had to simplify the problem at hand and simplify the datasets so that the brain could make sense out of it.
- What strategies should food and drink companies adopt to effectively integrate AI, ML, and RPA into their operations?
It should start with the business. It’s about solving business problems, not about finding a purpose for technology. What will have the most impact on the performance of the business (with the use of data) in a smart way?
RPA has existed for quite some time to automate relatively simple and repetitive tasks; the difference is that we can now combine it with data and AI to achieve smart automation. The combination of these technologies in one digital platform is crucial to avoid businesses trying to make many puzzle pieces from multiple vendors work.
- Can you share some examples of how AI, ML and RPA are being used successfully to enhance efficiency, sustainability, reduce costs, improve yield and address workforce shortages in the food and drink industry?
For Infor, it’s not just about providing the technology platform, but also the industry-specific use case and templates of AI and EA. This speeds up adoption tremendously. As an example, Amalthea in the Netherlands implemented ML in its cheese-making process within eight weeks. The ML model provides real-time insight into the milk yield per cheese batch. The customer can see whether it's higher or lower than the target value and can see directly why a certain cheese batch has a higher or lower milk yield.
The ML model looks at 20 process parameters, such as raw milk butterfat and protein contents, milk temperature, and equipment settings, which is way more than could be done in the past manually. Also, it can be done immediately per production batch instead of an after the fact analysis a couple of weeks later. Increasing the milk yield by only one per cent saves Amalthea €500,000 ($802,000) per year.
Another example is Zeelandia, a global bakery ingredients supplier – its sales experts saved 83 per cent of time preparing product and price recommendations for their customers and improved customer experience.
- What are some specific applications of big data and predictive analytics in the food and drink industry to optimise operations and prevent potential issues?
Some interesting applications include improving the forecast accuracy and responsiveness to demand changes with the use of additional data sources, such as the effect of competitor promotions, the weather, etc.
It helps to meet demand more efficiently with less overtime in the factory and reduced food waste from the overstock of perishable products. It’s also applied at the sourcing side to predict yield of harvests. For the latter satellite images can be used. This is important because crop supply can be volatile and better predictions give food processors more time to rebalance demand and supply.
- Looking ahead, what major challenges do you foresee for the food and drink industry in 2024, and how can AI and automation help address them?
The food and beverage industry is a bit more stable than the last couple of years, during which we have seen price inflation and a huge shortage of workers.
You see that food and beverage processors are now looking ahead and strategically investing in technology, either being factory automation or the digital business platform.
This is primarily done to offset the shortage of workers, but building in resiliency to deal with the increased risk of crop failures is probably even more important.
![Infor Industry & Solution Strategy director for Food & Beverage, Marcel Koks. (Source: Infor)](http://yaffa-cdn.s3.amazonaws.com/yaffadsp/images/dmImage/StandardImage/marcel-koks.jpg)
Infor's Industry & Solution Strategy Director for the Food & Beverage sector Marcel Koks is an acknowledged food & beverage technology expert who sets industry and solutions strategy for Infor’s food & beverage business unit and its Infor CloudSuite Food & Beverage solution. He has more than 20 years of experience using, implementing, and selling enterprise applications for resource planning and supply chain management.