“How can we allow food to be thrown away when more than 820 million people in the world continue to go hungry every day?”, asked FAO Director-General Qu Dongyu in the foreword to his agency’s report?
One-quarter of food bought by private households in the U.S. gets wasted. Also 6.7% of the greenhouse emissions come from the rotten wasted food in landfills. Food waste is a social, humanitarian and environmental concern. Reducing food waste can significantly narrow the huge gap we need in future food production.
AI can help ensure that ZERO good food actually reaches the land waste
Machine Learning can be the force which will empower the food supply chain system to operate without waste. This will trigger them to become competitive when pricing. The predictive data will enable wholesalers and retailers to order based on sales patterns. It’s a win-win situation for all! Deepnify is a top player in this area.
One of the solutions use the computer vision technology of AI. This Singapore based solution “Good for food” by a young team, make use of cameras set up in the waste bins to detect the food being thrown away. This would help the hotels to study the data of the kind of food versus quantity being wasted. The hotels can then plan their buffet meals accordingly, to prevent wastage. They understand consumer behavior without explicitly asking questions to the customers on their likes and dislikes. The same can be applied to supermarkets to monitor the wastes. This can be extended to office and household also!
The training initially will manually tag and monitor to identify the food rightly, to the extent that it becomes smart enough to differentiate the curry from the actual meat for creating structured data. These solutions will slowly mature and reuse learnings to become smarter. It will consolidate the learnings with more data patterns and scale smartly to be used on a wider range of scenarios.
Similarly, Winnow through its AI platform helps provide insights on wastes, to chefs, on a daily or weekly basis helping chefs plan how much to order saving expenses of the hotel and at the same time preventing food wastage.
But AI can extend its capabilities to many more business cases. AI can detect and predict the threats to the food, at each level of the supply chain, to prevent food wastage or deterioration in quality. Predicting and giving visibility on the shelf life accuracy will be a crucial way to save some food from still making its journey till a retailer. The predictive analysis of how fresh ‘x food’ will be after ‘y days’ will be useful. This helps retailers buy food by demand patterns, with the least wastage. It can also help food move out of the supply chain at the right moment to reach the mouth of the hungry rather than get wasted and rotten and reach landfills. Agshifts deep-learning platform is working towards one such solution.
AI solutions all along supply chain
The food wastage can also be stopped at the food inspection and grading stage with the help of automation through AI. Trained people or inspectors currently help to determine the grades based on patterns that come from the education, experience, and interpretations that have evolved across the years of experience. Which means it’s biased. The neural network algorithms can help grade and provide insights to assist inspectors in making decisions instead. These gradations would be more reliable in unbias, consistency, and standardization. This is further enhanced since the same solution and reusable algorithm will be used by different humans to predict the quality of the same edible item, at different levels of a supply chain – starting from Production>>Harvesting and transit>> Primary processing>>Secondary processing>>Distribution, packaging, handling>> to Wholesale and Retail markets.
The rejections at different levels are unavoidable to ensure quality. For example, in storage and shipping, there is a 17% loss of global production. Nevertheless, it will surely help wholesalers and retailers save money and prevent food wastage through near accurate analysis and predictions.
Another objective of AI solutions can be to prevent food deterioration in transit or shipping before it reaches retailers. This will also reduce food wastage. The factors causing deterioration of products are detected using sensors for temperature and humidity as an example. Data collected with the help of sensors is used to train the AI-enabled platforms. The algorithms would help warn beforehand, preventing aggravated product deterioration. These advanced alerts from the AI systems can also help to plan and speed up the delivery to the end destination, in standard edible condition. These were simply a few use cases.
There are still lots to be done to help retain post-harvest quality and prevent edibles from perishing prematurely. And different AI solutions can definitely ease things for all stakeholders in the food industry. All it needs is a mindset change.
From farm to fork, AI can enable reduce food wastage at consumer level and food loss at production level, if adopted rapidly in all remote corners of the globe.
The Author Satarupa Sen is a graduate from BHU. She has 19 plus years of IT experience in providing digital solutions including AI. She has now taken a break from her work to explore pragmatic ways to apply AI in Agriculture.