Algorithms powered by Artificial Intelligence get better and better in their performance by continuously learning from experience, by adapting to the environment they operate in.
When it comes to food industry, artificial intelligence is able to analyze vast and complex datasets from the large consumer base that food manufacturers cater to daily. Whether it be identifying repeat customers, or personalizing offerings, or which products are better to deliver at last minute, artificial intelligence has a critical role to play.
Artificial Intelligence powered algorithms learn from a variety of factors on analyzing customer behavior, including product promotions, mapping accurate inventory levels, and identifying need for product replenishment.
Tokyo-based Kewpie Corporation deploys Google's TensorFlow to quickly inspect ingredients, including the diced potatoes it uses in baby food. Kewpie and its partner, BrainPad attuned the Machine Learning
System to recognize good ingredients by feeding it 18,000 photos, and set it to work looking for visual 'anomalies' that hint at sub-par potatoes. The result was an inspection system with "near-perfect" accuracy, culling more defective ingredients than humans alone -- even with a conveyor belt shuttling potatoes along at high speed.
CA-based Abundant Robotics is part of a new generation of hardware companies developing autonomous equipment for use on farms. Abundant Robotics has built a robot able to pick the right apples, while San-Diego based Vision Robotics is working on a pair of robots that would trundle through orchards plucking oranges. These types of solutions have the potential to save farmers millions of dollars in labor costs and spoiled fruit.
Reducing Food Wastage
California-based AgShift, a deep-learning tech startup, is building an autonomous food inspection system to reduce global waste. The AgShift solution blends deep learning with computer vision to autonomously inspect the produce and other commodities for defects. It does quality assessments and makes judgements as per USDA specifications or an enterprise’s own specifications. The patented deep learning models analyze the defects in the sample images and predict the overall quality of the sample.
AgShift’s data visualization dashboard enables drill-down analysis of inspection data captured daily across the supply chain. Indices, inspection reports, defect maps are published in real-time unearthing patterns and market conditions within the organisations for actionable decisions.
New Product Development
NYC-based Analytic Flavor Systems aims to usher in a new era of hyper-personalized food. Its product, the Gastrograph AI is a smartphone app that features a 24 spoke wheel. Each section of the graph represents a category of sensory experience relating to food. For example, “bitter” and “mouthfeel”. The user maps their perception of the flavor into this wheel by tracing the spokes corresponding to their qualities. The intensity of each is also recorded on a scale of one to five. Then, the taster gives the food an overall rating between one and seven.
Gastrograph AI, uses machine learning and predictive algorithms to model consumer flavor preferences and predict how well they will respond to new tastes. The data can be segmented into demographic groups to help companies develop new products that match the preferences of their target audience.
The Gastrograph app also gathers data about the person doing the tasting—demographic information, socioeconomic status, past experience with the product, smoking habits, and more— as well as information about the ambient environment, such as temperature, barometric pressure, and noise levels, all of which can shade our experience of how things taste.
Cleaning processing equipment
Cleaning processing equipment requires a lot of time and resources, including water. Researchers at the University of Nottingham are developing a system that uses AI to reduce cleaning time and resources by 20-40%.
SOCIP is a pioneering cleaning system that will use artificial intelligence to autonomously optimize the cleaning process for food manufacturing equipment. When the SOCIP Development Project is complete, it will enable the UK food industry to cement its global leading position of delivering world-first technology. The project is led by Martec of Whitwell Ltd in collaboration with the University of Nottingham and the Centre for Sustainable Manufacturing and Recycling Technologies (SMART) at Loughborough University.
The SOCIP system could save the UK food industry £100 million per year.