Traceability and Recall Best Practices

When an FDA-regulated product is either defective or potentially harmful, recalling that product is the most effective means for protecting the public. Recalls are almost always voluntary. Sometimes a company discovers a problem and recalls a product on its own. Other times a company recalls a product after the FDA raises concerns. Only in rare cases will the FDA request a recall. But in every case, the FDA’s role is to oversee a company’s strategy and assess the adequacy of the recall.

Effective recalls rely on traceability, and the food industry is stepping up in providing resources. One is the Global Food Traceability Center (GFTC), developed by the Institute of Food Technologists (IFT). Deliverables include online courses, podcasts, journal articles and other knowledge items on such topics as creating interoperable, full-chain food traceability. “The main idea of traceability is to record relevant information (referred to as KDEs [key data elements]) associated with the physical goods (‘traceable entities’) as they move through specific steps (CTEs [critical tracking events]) in the supply chain,” said a GFTC spokesperson. “Therefore the architecture and the components that will enable it to function must be suited to managing “big data” while simultaneously providing granular insights as required.”

Plex, a leading provider of cloud-based enterprise resource planning (ERP), is among the latest software platforms. It has identified the following as big-data “must-haves” for food and beverage manufacturers:

1. One-up/one-back traceability both for ingredients and finished goods.
2. Real-time tracking and conformance/corrective strategies.
3. Batch recipe management, including clear alerts for every ingredient in every product, particularly allergens and other data.
4. Data access and storage – traceable, complete and user-friendly.
5. Customization and flexibility, such as adding product-specific labeling and barcoding features easily.

How Amazon Uses Technology for Monitoring Customer Feedback

At this year’s Food Safety Summit, keynote speaker Carletta Ooton, vice president of health, sustainability, security and compliance, at Amazon, described how a visionary company applies new tools for food safety.

Heartbeat, an Amazon built tool, enables Amazon to sift through over 20 million pieces of customer feedback per week to identify text specific to safety concerns. “These interactions [from customers] include publicly available data like product reviews and detail page ‘ask an owner’ correspondence as well as internal data like customer return comments, CS chat and machine-transcribed CS calls,” Ooton says. “The scale alone is not the issue. Our challenge is extracting relevant and actionable meaning from the interactions.”

“To do this, we rely primarily on Natural Language Processing (NLP),” Ooton says. “NLP is a broad field of computer science that focuses on applying machine learning to understanding language. One of the key differences between NLP and general keyword searching is the understanding of context, sentiment and sentence structure. By doing so, we are able to surface true safety-related concerns about products to subject matter expert teams for investigation and subsequent needed actions.”

On December 1, 2017, Amazon detected several food safety concerns on a dietary supplement. They removed the product from the website the same day and initiated an investigation with the manufacturer. Almost two months later, the product was officially recalled due to undisclosed allergens—and the side effects listed in the official recall nearly exactly matched Amazon’s detected feedback.

“We design safety into every process touch, and then we audit to ensure that the process is maintained,” she said. “In the rare occasion when a key performance indicator (KPI) and a customer anecdote disagree, the customer is nearly always right. These customer signals are a rare gift we obsessively seek.”