Analytics create better systems to grow, distribute and purchase what we eat
By 2050, there will be more than 9 billion people on the planet.1 And because they will all need to eat, food production will have to double today’s output to keep up.2 But we aren’t very efficient with our food right now. Up to 40 percent of all food produced in developed nations never makes to a human stomach.3
While looking for ways to increase the farming capacity has gotten bulk of the global attention, it would be just as important to modernise farming in a way that minimises usage of natural resources and make decisions based on scientific models instead of leaving it to chance. We need to be smarter about food: how we grow it, harvest it, distribute it and consume it.
The resources we use for our food supply – land, water and most forms of energy – are not only finite, they are at risk. Fortunately, there is a bounty of one resource that we can use to be more efficient: information. Using big data and analytics, we can create smarter food systems across the value chain.
Analytics and agriculture
You might not think of farmers as heavy users of information technology. But from the days of the earliest almanacs, farmers have used historical data to help improve their yield. Now a variety and velocity of structured and unstructured data can help today’s agribusinesses create a more informed picture — from things such as sensors used to gather soil and weather information, posts to social media sites, and GPS signals, to name a few. New technologies, such as predictive analytics and commodity price optimisation, can also help farmers anticipate and adjust profitably to marketplace, weather and other conditions that typically leave them vulnerable.
Deep Thunder and precision agriculture
Watch the video (00:03:15)
This idea of “precision agriculture” drives projects such as Deep Thunder (US) by IBM Research. Measurements of the weather and soil, including data from sensors dotting a farm, multi-spectral images of fields taken from satellites or airplanes, characteristics of irrigation systems, requirements for fertilizer and pesticide coupled with precise weather predictions — all can help optimize a farmer's decisions about what to plant, when to plant, when to water, when to fertilize and when to harvest.
Sustenance and sustainability
When you eat a carrot, you also consume the resources that went into growing it: water, fuel, electricity.
Estimates show that the production of one calorie of food requires an average of 7–10 calories of input. Of course, this varies depending on the crop, from three calories for plant crops to 35 calories for beef.4 By leveraging sensor and traceability data, we can now remotely monitor and determine when someone has been engaged in unsustainable farming practices. Consumer products companies can have the visibility into their suppliers operations and influence them to make better decisions that have positive impact for all of us.
Sun World (US), a California fruit producer, uses an analytics system to analyse its water usage — as well as crop yields, farm labour costs, growing patterns, and an array of sales and distribution processes — to gain insights that enable it to make faster and better decisions in response to changing conditions.
Looking at historical weather, crop yield and irrigation data, Sun World has been able to modify operations to take advantage of natural weather patterns as well as gain a better understanding of the exact amount of water each type of crop needs. For example, increasing the use of drip irrigation systems has cut the company’s water use by 9 percent over four years.5 They have also been able to maximise profit for each harvest by having a better understanding of the demand and market forces that could influence how their products will be consumed.
Safety and the supply chain
Imagine knowing exactly where your snack time apple came from and just how long ago it was picked. That possibility is closer every day. The phrase “farm to fork” is more commonplace now, as is the demand for traceability as food products make their way through the supply chain. Knowing a food’s origin also supports the “buy local” movement, cutting down on transportation costs and shortening the supply chain.
But even a locally sourced produce item stands a good chance of spoiling before being purchased. In an effort to appear bountiful, grocers often overstock shelves regardless of demand, resulting in anywhere from 5 to 15 percent of produce being discarded due to spoilage... produce that has used water, labour, petrol and packaging… only to be thrown away.
But whether spoiled on the shelf or contaminated in the plant, food-related crisis situations have the potential to affect thousands of people, leading to significant healthcare costs, loss of revenue for food companies, and — in the worst cases — death. In the United States alone, one in six people are affected by food-borne diseases each year, resulting in 128,000 hospitalisations, 3,000 deaths, and a nearly $80B economic burden. 6
IBM scientists have built a system that automatically identifies, contextualises and displays data from multiple sources to reduce the time to identify the mostly likely contaminated sources by a factor of days or weeks. (US) It integrates pre-computed retail data with geocoded public health data to allow investigators to see the distribution of suspect foods and, selecting an area of the map, view public health case reports and lab reports from clinical encounters. The algorithm effectively learns from every new report and re-calculates the probability of each food that might be causing the illness.
In China, pork is a major pillar of the economy in the Shandong Province, one of the country’s most important agricultural regions. To limit the impact of porcine diseases and prevent tainted pork from being sold to consumers, experts from IBM China Development Lab and China’s National Engineering Research Center for Agricultural Products Logistics have created a pork monitoring and tracking system (US). It can extract and store information from millions of interconnected sensors. The system brings an unprecedented level of accountability and efficiency to every stage of the pork production process, from production to distribution to retailer.
Grocers, vendors, suppliers and purveyors all share data from farm to shelf to cart. Watch the video (00:06:12)
1. International Data Base, United States Census Bureau, accessed at:
2. United Nations Sixty-fourth General Assembly Second Committee press release,
“Food Production Must Double by 2050 to Meet Demand from World’s Growing Population,” retrieved at:
3. Food Insight, 2012 Food & Health Survey: Consumers Attitudes Toward Food Safety, Nutrition and Health, May 2013.
4. Institution of Mechanical Engineers, Global Food: Waste Not, Want Not, January 2013.
5. Peter Williams, Data Analytics: Delivering insight for water management, February 16, 2012.
6. U.S. Department of Health and Human Services, Centres for Disease Control and Prevention, Frequently Asked Questions – Foodborne Illness.