AI and Technology in Food Manufacturing: How Nestle and Others Stay on Top
Food manufacturers had to suffer another difficult year due to market dynamics in 2022, which look likely to persist this year. Food and beverage manufacturers are under pressure due to quickly shifting consumer preferences, stricter regulatory requirements, and rising commodity prices. Many companies need to upgrade their equipment and technology to meet up with rising consumer demand.
Why Food and Beverage Manufacturers Need to Upgrade Their Technology?
Food tech is revolutionizing the food sector by employing the newest technology to control production, distribution, and consumption as firms turn to technology to combat inflation and boost efficiency. According to the US Department of Agriculture, the food sector contributes more than $1 trillion to the US GDP. There are several difficulties in such a big industry, such as food sustainability.
The worldwide food industry is being transformed by food technology. With the emergence of big data, AI, and the internet of things (IoT) in recent years, food technology has established itself as a separate industry. Technology used in food industry must help the industry to Increase visibility, optimize operations, improve quality and output, increase agility, adhere to regulations, and speed time to market. Food industry uses technology at every stage of manufacturing.
AI and Technology Application in Food Manufacturing
Currently many food manufacturers use AI and combined it with existing or other technologies in the food manufacturing processes. Here are some innovative ways AI can be implemented in the food manufacturing:
AI can be used for concept research in food manufacturing by analyzing vast amounts of data to identify patterns and trends in consumer preferences, market trends, and production processes. AI can be trained on large datasets of product reviews, industry report, sales data, and consumer feedback to identify emerging food trends and new product concepts.
AI can be used for formulation development in food manufacturing by analyzing large datasets to identify the optimal combination of ingredients, flavorings, and additives to create new and innovative products. AI algorithms can analyze the chemical properties of different ingredients and predict how they will interact with each other, allowing for the development of more precise and efficient formulations.
Clinical data mining
AI can analyze large amounts of clinical data to identify patterns and trends related to nutrition and health. This can help food manufacturers to develop new products that meet the specific nutritional needs of different populations, such as athletes, children, and seniors. By leveraging clinical data and AI, food manufacturers can develop products that are both nutritious and safe, while also meeting the evolving needs and preferences of consumers.
Raw material quality assurance
For raw material quality assurance in food manufacturing, AI can be used by analyzing data from multiple sources, such as supplier performance, ingredient quality, and environmental factors, to predict potential quality issues and ensure the consistency of raw materials. AI can also be used to monitor and track the quality of raw materials throughout the supply chain, from sourcing to processing, to ensure that products meet regulatory and quality standards.
Improved process control
AI can be used to optimize manufacturing processes by analyzing data on factors such as temperature, pressure, and flow rate. from sensors and production lines to optimize and automate production processes. AI can identify patterns in the data and predict when a process may be going out of control, enabling rapid intervention to prevent product defects or quality issues.
Early problem detection
AI can be used for analyzing data from multiple sources and predict when a problem may occur. By using machine learning algorithms, AI can learn from historical data to detect early warning signs of problems, such as equipment failures, deviations in process parameters, or quality issues. Early problem detection using AI can help food manufacturers to take corrective action before problems occur, minimizing product defects, reducing waste, and avoiding production downtime
How Food Manufacturers Leverage Technology
Whether it is the production or the research, here are the examples of how food manufacturers leverage technology:
“We moved, actually, from an average project duration of 33 months to 12 months, and that's an average of different categories. In food and beverage, sometimes projects take us only six to nine months, so we are faster now than many of the startups that are out there.” said Stefan Palzer, Chief Technology Officer
Since 2016, the business claims that Nestle SA has increased the pace of its product development by 60%. Its research and development process was restructured to achieve a faster speed to market. Concept research, formulation development, plant breeding, clinical data mining, raw material quality assurance, improved process control, and early problem detection are just a few of the ways that AI is already being applied throughout the enterprise. In order to meet the increased complexity of the product creation process, where items must taste good, be regarded as healthy, be sustainable, and be inexpensive, Mr. Palzer stressed that AI and machine learning are now essential product development tools.
Not only on research, at their factories, Nestle also employ predictive maintenance automation using machine sensors that raise an alarm if something is off. The production line will halt unnecessarily if troubleshooting is not monitored, thus they also utilize predictive models to make troubleshooting more effective. Using chatbots that automate front line customer care, Nestle was able to cut costs. They also deploy chatbots to assist their supply chain partners in finding the internal resources and information they require. AI has not only sped up the decision-making process but also revealed previously hidden insights for Nestle.
One of the biggest food and beverage corporations in the world, Pepsi, consistently invests heavily in artificial intelligence technologies to improve almost every element of its operations. Shameer Mirza, senior research and development engineer at PepsiCo, was aware of the various ways artificial intelligence could be applied to enhance the management of the manufacturing process. Mirza developed a machine learning method that could be used in conjunction with a visual system to calculate the amount of potatoes that had been treated. The company was able to save a sizable sum of money because it was no longer required to purchase measuring equipment for $300,000 per line (PepsiCo have 35 in the US alone). The only tools used by Mirza in his answers are a camera, a computer vision model, and extra sample points that were collected for free.
The manufacturing division of the PepsiCo subsidiary Frito-Lay is benefiting from machine learning. One prototype strikes chips with lasers, then detects texturing by hearing the noises the chips make. Software analyzes audio and calculates chip texture to digitize the performance analysis for Frito-Chip Lay's manufacturing facilities. Besides that, many of Frito-manufacturing Lay's facilities use technology to monitor and track data on the equipment there in order to foresee mechanical issues before they occur. The plants experienced no unforeseen failures or outages after a year. Mechanics were able to use their time more effectively with the aid of AI so they could concentrate on planned maintenance and take preventative action before equipment failed.
The J.M. Smucker Company
“We dabbled in a small portion of the business and saved $500,000. If we keep getting organized around that, we can save even more.” — Baier, the Senior Manager of IS Operations at Smucker’s.
J.M. Smucker is a well respected pet food, coffee, and consumer food manufacturer in North America. Smucker's lacked a data analytics team two years ago despite having a vast amount of operational data. It now has a four-person team working to extract value from the data collected from its production sites. What significantly changed? The leadership of Smucker understood the enormous potential of using big data to investigate production issues including product overfill, hidden plant capacity, and equipment downtime.
Smucker's sought to make it possible for facility managers to get information in almost real-time in order to maximize output and provide everyone the chance to make strategic modifications, from top leaders to workers on the shop floor. A production system needs time to process some adjustments. Some processes are batch-based, thus the faster a troubling data pattern is identified, the fewer batches will be discarded. After getting insights from this real time data, Smucker can drive process and people changes, and even saved $500,000 a year.
Digital Transformation Prevented Million Dollars Loss for T. Marzetti
At the specialized food company T. Marzetti Company, predictive analytics and connected kitchens unlocked millions of dollars in savings. With its headquarters in Westerville, Ohio, T. Marzetti Company is a division of Lancaster Colony Corp. Data analysis wasn't a common practice at Marzetti. Standards for connecting data to its supply chain didn't exist, and information wasn't always easy to obtain. The Marzetti's partners suggested starting with a single facility in Kentucky and digitizing data communication among important floor assets before contextualizing the information acquired to speed up the Marzetti Operational Excellence (MOE) effort. One of the main goals was to provide more people the authority to improve the company. For the first time, employees at all levels of the business had the linked, data-driven tools to identify efficiency-improving opportunities that might have gone missed otherwise.
As the first milestone in their digital transformation, this effort allowed them to discover where waste can be cut out and money can be saved. Overfill was one of the biggest issues they discovered. Four pounds were handed away for free for every 100 pounds of packaged goods because packages were being overfilled. Little surpluses of sauces, dips, and dressings were sneaking out as stowaways. It accumulated to a lot over time.
Marzetti and their partners made the necessary changes to capitalize on the information the data showed about their cryovacs, which are automated weighing devices that seal food in airtight packaging. Performance comparisons by product code, equipment, order, rationale, and other settings are possible with the system. When an exception is found, staff are automatically emailed, the system is even able compare the accuracy of fills made during the "restart" time to those made while the filling mechanism was operating at "cruising pace."
Marzetti estimates that the savings might increase three to four times when the upgrade initiative is expanded to additional locations, as Marzetti aims to accomplish. Marzetti has already saved millions of dollars annually at just one of its plants. The initial achievement was achieved by cutting product waste by at least 50%.
In conclusion, the food and beverage manufacturing industry would benefit from riding the wave of digital change. A high-quality product and meeting the customer demand are essential, and without technology, there would be much more space for production errors and unproductive processes that went unnoticed and affected overall business. As what the technology discovers is frequently more extensive than what the human eye can see.
Experiencing challenges with efficiency, quality, and profitability in your food manufacturing production? We can help and assist you in your digital transformation journey, contact us now!
#food #manufacturing #AI #MachineLearning #DataAnalytics #productivity #foodandbeverage #automation