The FDA investigates bringing the IoT to food supply chains

A combination of sensors, blockchain technology, and the cloud is poised to deliver almost real-time tracking of food as it leaves a farm or fishery, and the U.S. Food and Drug Administration wants to help the technology behind food traceability along.

The FDA has named 12 companies as winners of its New Era of Smarter Food Safety Low- or No-Cost Tech-Enabled Traceability Challenge, pulling the winners from more than 90 submissions from around the globe. The challenge sought companies that could improve food traceability from harvest through packaging, shipment, and finally, consumption. The businesses also had to provide this traceability in a manner that was easy to implement without IT expertise and at a low cost.

The contest was created as part of a 2020 FDA program to make food safety and traceability a priority. The idea of tracing food “from farm to fork” has been talked about for the last decade as food distributors battled recalls and saw opportunities in providing consumers with more information about their food. Ethically caught fish or fair-trade coffee are examples of selling consumers on a traceability story.

Both those efforts are industry-specific and often driven by a larger seller or distributor as opposed to an open standard that can be used across different food producers.

There are practical reasons the FDA wants to track food. Foodborne illness sends 128,000 people to the hospital worldwide each year and costs about $6 billion according to Sharmeen Khan, chief marketing officer at OpsSmart, one of the winners. In addition to food safety, it can address counterfeit food at restaurants or a need to track the carbon footprint of an item to meet suppliers’ environmental promises around carbon reduction.

A few of the winning solutions involved sensors from Wiliot and Eniot to track the conditions of food as it moves through the supply chain. Williot was chosen by two companies because someone can attach its low-power, energy harvesting Bluetooth sensors to pallets or packages for a cost of less than 10 cents per sensor. The challenge then is ensuring there’s a gateway device in near range to “read” the sensor data. That data includes temperature tracking and GPS location.

Other projects did away with sensors, but most had some form of blockchain traceability program. The process generally involved logging a unique package or pallet ID at harvest or packaging, then storing those data points on a blockchain. As the item moves through the supply chain, each subsequent step is logged along with any relevant data. The most complete solution I saw involved Avery Dennison using its cloud resources, Wiliot sensors, and Mastercard’s blockchain to track food.

While watching all of the presentations, I realized how daunting setting up any sort of global food traceability standard would be. Getting harvest data into a traceability system requires ruggedized handsets designed for areas without connectivity and people who may not speak English and have many other tasks to handle with their hands. Data logging has to be part of the process, but can’t take too much time.

Harvesters have to collect data on where the food was picked, what it is, and the condition it was in. Once packaged, the packager (who may not be the same person who harvested the food) has to inspect the food, note where it was packed, and perhaps include notes or a photo of its condition. From there they may print a label for the package and attach a sensor. Packers aren’t necessarily tech wizards, so all the equipment has to be easy to use and durable.

From packing, food can go directly to a distributor or even a restaurant, while other food might get combined with ingredients to become part of a pizza or a salad kit. In most cases, the FDA and startups see distributors and upstream packagers as the ones who will pay to use the traceability system, not the farmers.

This means that the big distributors are choosing the technology, and they may not choose platforms that make life easiest for the farmers or downstream suppliers. This is why the FDA focused on awarding systems that were easy for farmers to use and log data into.

The final step in this journey usually ends with a QR code on a package and a consumer eager to scan the item to understand where it came from. Some of the winners even expected the consumer to download an app.  Maybe I’d do that for a piece of steak or expensive coffee, but I think the focus should be more on the grocery stores or restaurants to act as the primary last tracker in the system. This means that recall information would go to them as opposed to a push notification on my phone.

It’s also clear that any traceability system will require multiple vendors working across clouds and continents, which means the vendors who succeed should follow existing data models for food and provide robust APIs to link to clouds or log food into a blockchain.

Creating such a system will be costly, and the costs will likely get passed along to the consumer in the form of higher food costs, but the benefits of assuring an item of food is what it says it is and the ability to trace foodborne illness are compelling.  I expect the government to drive the adoption of such traceability chains as part of federal nutrition programs like school lunches. I hope they will also push it further with regulations and offset higher costs to those who can least afford it with another bump in spending on the Supplemental Nutrition Assistance Program.

For more, watch the FDA panel with the winners from Tuesday.