Artificial intelligence (AI), or machine learning/machine vision, is playing a predominant role in the world of food safety and quality assurance. According to Mordor Intelligence, AI in the food and beverages market is expected to register a CAGR of 28.64 percent, during the forecast period 2018-2023. AI makes it possible for computers to learn from experience, analyze data from both inputs and outputs, and perform most human tasks with an enhanced degree of precision and efficiency. Here’s a brief look at how AI is augmenting food safety and quality initiatives.
Get Paid For Your Thoughts!
- Wiley (Food Quality & Safety’s publisher) is offering $200 to qualified food scientists who participate in research interviews about challenges facing the food industry.
Take the survey >
Round the Clock Efficient and Effective Monitoring
Sensors not only monitor temperature, humidity, pressure, and time, but they also record data, highlight areas of improvements, and in some cases, make critical decisions to ensure the safety of the products is not compromised. “Oh, my walk-in refrigerator just texted me,” is a real conversation that most folks in the storage and distribution lines are quite familiar with. Other monitoring techniques include utilizing spectroscopy, lasers, X-rays, or cameras to examine both intrinsic and extrinsic characteristics of produce from harvest to packaging. This is a huge leap from conventional sorting systems that separate what is programmed as “acceptable” from the “rejected” lot, to almost intuitively making decisions based on best yield by sorting products based on their optimized use. Tomra is an example of a business that is demonstrating effective food resource management using machine learning.
Enhanced Traceability with Precision
There used to be a time when the success of a recall or a product trace depended on simply gathering data and interpreting it accurately. Today, executing strategic safety interventions in the event of a recall relies heavily on utilizing the least amount of time possible to gather data, interpret the findings, validate them, and share the results. AI systems have made it possible to compare historical data and predict certain events across multiple timelines from different regions.
When you think food traceability, Amazon is not the most likely name to come to mind. This organization is making headlines through natural language processing—a system that continually monitors and processes customer feedback by scanning millions of emails, instant messages, phone calls, and social media platforms to detect a food safety issue even before the product gets recalled by the manufacturer. At the Global Food Safety Conference 2018, Carletta Ooton, vice president of health and safety, sustainability, security, and compliance at Amazon, shared how the leading digital retailer blocks a faulty product on an average of 50 days before the official recall, simply based on collective customer feedback.
Automation Augments Sanitation
“Smart agriculture” is a classic example of how robotics are augmenting the production, processing, and packaging of food products in a sanitary fashion. Clean-in-place (CIP) systems are programmed to clean equipment in timed cycles. The advantage of operating a self-cleaning unit is that it limits human intervention, which in turn, limits the chances of cross-contamination from foodborne pathogens. Martec of Whitwell and the University of Nottingham are in the process of conducting research for the Self-Optimizing CIP (SOCIP) Development Project. Systems like these are intended to be programmed to flex cleaning schedules based on operational use. This automated process can be taken one step further to ensure CIP systems are also sustainable by optimizing resources.
The common emotion associated with technological advancements is fear, stemming from the notion of jobs and careers being displaced. While most labor-intensive processes have been replaced or deemed redundant by automation, the flip side is there is a growing need for specialized skill sets around gathering and analyzing data. Machine learning is still in its infancy stages, and technological advancements are occurring much faster than we anticipated. From self-driving cars to self-prepping salad bars, the possibilities are definitely endless.