According to a report by Accenture, 85% of organisations have planned to adopt digital or AI technologies in their supply chains during the last year.

However, AI is not necessarily a new concept. Various names refer to more or less the same thing. For example, machine learning is used to steer self-driving cars. AI is proving instrumental in healthcare for identifying and diagnosing complicated ailments and even everyday search engines like Google use AI to refine and improve results the moment you tap in a few keywords.

The early adopters of AI include mainly pharmaceutical, healthcare, cosmetics and retail industries. The food and beverage industry is not quite so advanced in terms of AI adoption in its supply chain planning processes.

The obvious benefits of machines over humans are efficiency and speed. However, the majority of companies we speak to about AI are more driven by the promise of additional revenues, better margins and lower costs. Increased efficiencies in the food supply chain are getting harder to achieve. Many companies think they’ve already done most of what can be expected using yesterday’s technology solutions and supply chain optimisation processes and this is where AI comes in.

AI relies on a continual process of technological learning from experience and getting better and better at answering complex questions. Algorithms powered by AI can rapidly come up with alternative options which are otherwise much more time-consuming and laborious using conventional computer-powered A/B testing. Like the human brain, AI adapts to the environment and gets better the more you use it. But unlike humans, the capacity for improvement is unlimited. What’s more, boring, repetitive tasks are never a problem.

Unravelling data
With most food manufacturers and retailers having thousands of customers and products to deal with on a daily basis, machine learning is proving much more efficient at unravelling complex data quickly and meaningfully. For example, retailers want to be able to cluster and identify who are their main customers - who are repeat purchasers, browsers, or so-called aliens. Or they might need to know which products are better to deliver last-minute; or which core lines should regularly be in stock.

And with many retailers dependent on promotions for contributing up to 30% of sales machines can tell which promotions work best. Machine learning is more effective at clustering promotions based on looking at similarities and many more variables than is otherwise possible using traditional, linear-based forecasting techniques.

Software starts with creating baseline forecasts for what particular products should ideally be stocked, where and when. AI-powered algorithms learn from a multitude of factors that are likely to influence buyer behaviour - including promotions, social media, or the weather - which then are used to more accurately manage inventory levels and replenishment. Not only will such advanced technology know when shelves are empty, but more importantly, it will predict what will happen next.

Waste not, want not
In food industries in particular, many are looking at new ways of reducing waste. Some food processing companies have already turned to artificial intelligence as a means to better calibrate their machines in order to manage several products sizes and reduce waste and costs. Others have been able to identify the optimal use of raw materials dependent on size and varieties. For example, one potato processing company has turned to AI to define which potatoes will produce the least waste when cut into French fries and which ones would work best for potato crisps. Some beverage companies have even used AI to rank flavour preferences among consumers - armed with an app on their phone in front of a self-service machine, customers could opt to change the flavour of their chosen drink - as part of developing new product ranges.

One of the biggest growth areas where AI can make a significant difference is in intelligent forecasting systems. Previously, retailers and manufacturers were only able to predict roughly the quantities of products to order to keep shelves fully stocked using (often out-of-date) inventory levels and historical sales data (usually going back a few years, at best). Today AI can develop a more accurate picture of what types of products are likely to sell, by looking at multiple scenarios in real time (suppliers’ data, consumer behaviour, the weather etc.) and drawing on data from the internet. This means forecasting is no longer so much “stab in the dark” guess work.

Stephanie Duvault-Alexandre is a business consultant at FuturMaster.

Source: Control Engineering Europe - All Articles
More...