Fighting Food Waste With
Artificial Intelligence
An estimated 1.3 billion tonnes of food is wasted globally each year, and at the same time, around 9 million people die of hunger annually. This is due to the inefficiency of the current linear food system.
In a vision for a circular food system, food production improves rather than degrades the environment and people have access to healthy and nutritious food. Agriculture is regenerative, improving the overall health of local ecosystems and people. Discussed below are the three main ambitions that can set the transition to a circular food system on the right path.
Match supply with demand
AI can play a role in forecasting demand using historical and real time data, potentially improving efficiency of supply chains leading to less overproduction and less overstocking, caused by distortion of demand information moving upstream in the supply chain.
Furthermore, it could be used to map the current landscape of global food supply chains and determine the optimal way to ‘rewire’ them to source and consume ingredients locally where appropriate, and to process food as close as possible to where it is grown and consumed, enabling better matching of supply and demand for food.
Make the most of food
- Production and processing:
An example is TOMRA‘s food sorting solutions which use AI algorithms to analyze images and data from cameras, near-infrared spectroscopy, x-rays and lasers, to identify non-uniform produce such as carrots and potatoes, and sort each one according to its optimal use.
- Distribution and storage
The manual process of inspecting quality, safety, size and appearance of food before reaching retailers is time-consuming, expensive, inaccurate, and often subjective. Technology can enable automated, objective inspection, potentially implemented upstream from distribution hubs. For example, the Food Team at Google is exploring how visual imagery techniques could help accelerate the food inspection process to improve food supply chain efficiency, minimise waste and enable more accurate retailer planning.
- Preparation and consumption
Companies like Wasteless help retailers to sell food before it goes bad, using AI-enabled tracking and dynamic pricing based on expiration dates. Tenzo uses AI algorithms to forecast and predict sales, enabling restaurants, retailers, and other hospitality institutions to connect supply to demand more effectively when ordering food, and reducing avoidable food waste.
- Valorisation of unavoidable food waste and by-products
While reducing overproduction is the best way of reducing food waste, unavoidable by-products, along with peelings, clippings and sewage, could be turned into valuable products. AI tools can then provide suggestions for potential recipes to make use of leftovers. It could also help provide the information about the nutrient content of organic waste streams (such as inedible food byproducts, human wastes, and green waste) and the presence of any micro-pollutants, in order to valorise organic by-products in the right markets.
By using AI as a tool to help source regeneratively grown ingredients, replace animal protein ingredients with plant-based proteins, reduce processing waste, and avoid unsafe additives, food innovators and designers can make it easier for people to access healthy food products.
In the current food system, for every USD 1 spent on food, society pays USD 2 of economic, social, and environmental costs. The food and agricultural system is too complex to fully understand with traditional analytical methods. Building a circular food system adds additional complexity, this is where AI is needed. By helping design out food waste, AI can generate an estimated economic opportunity of up to USD 127 billion a year in 2030.