AI in Practice: Agriculture
Democratization & adoption of AI to make food supply reliable and efficient can not only increase revenue for the industrial farm participants, but it also will help small farmers to have a decent improvement in standard of living by having direct access to the consumer.
Photo by Amanda Easley on Unsplash
Farm to Fork
Agriculture is not merely about growing stuff on the farm. There are a myriad of factors that play a crucial role in the ecosystem of activities around growing fruits, vegetables, grains, raising cattle for meat, or shipping the produce directly to your doorstep.
The current world population as of Jan 2021 per United Nations estimate is 7.8 Billion and is projected to reach 10 Billion by 2057. AI becomes immensely important to feed a growing population in light of depleting natural resources and the environmental strain caused by growing population and rapid urbanization across the globe.
It is important to note that food production is not the only issue, packaging, distribution, and consumption all are equally important. For example, in the United States alone, per the US Department of Agriculture, food waste alone is estimated to be around 30-40% of the food supply.
How much food waste is there in the United States?
Based on estimates from USDA’s Economic Research Service of 31 percent food loss at the retail and consumer levels, corresponded to approximately 133 billion pounds and $161 billion worth of food in 2010.
This amount of waste has far-reaching impacts on society:
1. Wholesome food that could have helped feed families in need is sent to landfills.
2. Land, water, labor, energy and other inputs are used in producing, processing, transporting, preparing, storing, and disposing of discarded food.
The number of variables involved in the food delivery chain makes it a very challenging and yet very rewarding problem to solve.
We will focus on three key areas where AI-enabled technology is playing a critical transformative role today and has a promising growth prospect in the future.
Weather Forecasting
Disease Prediction
Distribution
Weather Forecasting
Past, present, and future predictions of Moisture, Light, and Temperature are inputs that help in decision making on the farm. Accurate weather monitoring and actionable data can drive activities like:
The satellite weather data provided by NOAA along with local on the farm weather stations provides a very precise actionable no-frills information to the farmer to make day-to-day decisions.
If the prediction is for rainfall to occur, then the farmer can save on water by not watering the farm and refining the irrigation schedule. The amount of rainfall becomes crucial in deciding whether to fertilize or not, if it will be washed off.
Humidity prediction can help make a decision about harvest and feed storage. Wind and Sunlight data aids in using renewable energy in a very fine-tuned approach and can result in substantial savings on power cost.
Disease Prediction
Competitive pressure to lower the price of goods, increasing cost of insecticides & fungicides makes it essential to track and fix infections as soon as possible. Being informed about a potential outbreak of disease can help make proactive decision making versus reactive action after the outbreak. For example:
A combination of humidity, temperature, and image processing can predict the formation of mildew in strawberries. Based on the prediction a selective application of preventive fungicide can prevent the mildew to develop.
An image captured by a smartphone on the field can be mapped to a pre-processed corpus of images to relay the health information of the plant almost instantaneously. Along with the health diagnosis, additional valuable information like recommendation of pesticides, prevalence of the disease in the local area, and action taken by other farmers can be very useful.
Prediction of the onset of infection can alert local farmers, businesses, and administration to gear up and act, thus preventing massive losses.
Distribution
Production of produce is only one part of the food supply chain. The gap between production and consumption is the waste of about $161 billion worth of food per a 2010 USDA study. The loss of nutrition value, additional preservatives needed to keep produce fresh and health implications it will have makes efficient food distribution a necessity. Some interesting ideas that are addressing this issue are:
Establishing a platform to meet the demand-supply gap in a local zone. Keeping the customer informed about the upcoming seasonal food production from farmers in local areas, help generate interest and orders to motivate the farmer to be a part of the platform and sell direct.
Digital platforms enable small, medium, and large farmers to part of the same distribution channel, especially if the network is localized.
Reduction in need of packaging, storage, and fuel needed to transport not only helps improve the nutritional value of food consumed, it reduces waste and carbon emission, thus helping the planet.
AI in Practice:
IBM: Make faster, smarter decisions for agriculture
Agribusiness is on the cusp of digital transformation
IBM Agriculture helps overcome obstacles to digital transformation by combining the power of Artificial Intelligence (AI), data analytics, and predictive insights with unique agricultural Internet of Things (IoT) data, the expertise of veteran food and agribusiness industry leaders, and decades of IBM research. The result is a platform of customized low-cost solutions that help stakeholders across the ecosystem to make faster, more informed agricultural decisions.
Azure Beats: AI-driven insights for the agriculture ecosystem
Azure FarmBeats enables building data-driven digital agriculture solutions
Azure FarmBeats is a purpose-built, industry-specific cloud platform built on top of Azure to enable actionable insights from data. With Azure FarmBeats, you can
1. Aggregate agricultural data from different sources
2. Fuse different agricultural datasets from sensors, drones & satellites
3. Rapidly build AI/ML models using the fused datasets
4. Build your own customized digital agriculture solution
AgroStar: Cloud-based mobile app
Helping Farmers Win Combining Data and Technology with Agronomy
AgroStar has launched a cloud-based mobile app that is helping to boost crop yields and encourage best practices for small farmers in India. Launched as an on-premises ecommerce platform selling farm tools in 2008, the firm turned to Google Cloud Platform (GCP) to expand its offering. It now uses cloud-based analytics and is deploying ML models to provide timely advice in five languages on everything from seed optimization, crop rotation, and soil nutrition to pest control.
Summary
Agriculture is emerging from age-old traditional practices into a well-oiled sophisticated Agri-tech industry. The value that advanced analytics brings to the farmer, the distributor, and the consumer is phenomenal. Timely access to information can keep the cost of operations low while increasing revenue and making the process efficient and reliable.
