Due to the advent and implementation of cutting-edge technology, analyzing data for various operational purposes has become commonplace. In fact, many argue that espousing data analytics into operations is a must for an organization to remain relevant in an increasingly competitive marketplace. Some of the more well-known companies such as Google and Amazon already depend on data analysis to gain a competitive advantage.

The rationale for using and leveraging data stems from the speed and scale at which information is produced and consumed. Large and small enterprises are coming to the realization that our information-based economy will only expand in scope, which is leading many to allocate significant resources to developing comprehensive analytic infrastructures.

The proliferation of data and its subsequent impact on operations pertains to organizations outside of the corporate environment. For example, the health care sector is using data analytics to for a variety of purposes, including clinical trials, cost-reduction efforts, and treatment protocol, as well as others.

Data analytics is also being increasingly recognized as an important component of food safety efforts. Specifically, the definition of big data analytics as “datasets so large or complex that traditional data processing applications are inadequate” is of considerable relevance to food safety and quality efforts.

What Are Data Analytics?

While it is beyond our scope here to provide an in-depth perspective of analytics and its myriad of applications, a general overview will be helpful as a reference point going forward.

Put simply, data analytics involves collecting, filtering, interpreting, and disseminating data for organizational purposes, usually for potential or proposed strategic changes. Additionally, data analysts are often responsible for communicating recommendations—based on conclusions drawn from data projects—to decision makers for consideration.

Data Analytics Initiatives in Food Safety

  • Supply Chain Analytics

As a result of the Food Safety Modernization Act (FSMA), which further regulates the food supply chain, the industry is leveraging data analytics to assist in the safe transport of its product. The food supply chain is a relatively lengthy process involving the movement of foodstuffs from producers to consumers through production, processing, distribution, retail, and consumption. Food supply mechanisms directly relate to many cases of foodborne illness, as human resources are often required during each step of the process. Relatedly, food may become contaminated at any point in the food supply chain via human interaction.

Understanding the link between supply and contamination, food producers and transporters are increasingly reliant upon data analytics to manage complex supply chains. One proliferating use of analytics involves the use of sensor and radio-frequency identification (RFID) technology to monitor and track food shipments through logistical channels. For example, sensors are placed on containers of perishable foods with the purpose of monitoring the temperature and humidity of transport vehicles. If set environmental parameters should fail, the appropriate people are notified.

The accumulation and automation of data, such as that generated by sensors, is being used to pinpoint operational efficiencies, avoid wastage, and strengthen the supply chain. In effect, these efforts help to ensure that food products arrive at distribution centers, warehouses, and retailers safely and on time.

  • Spatial-Temporal Data Analysis

Time is of the essence in the event of a foodborne outbreak. According to Kun Hu, an IBM public health research scientist, “When there’s an outbreak of foodborne illness, the biggest challenge facing public health officials is the speed at which they can identify the contaminated food source and alert the public.” Swiftly identifying the source of contamination is crucial for a number of reasons: isolating the product(s), initiating recall efforts, informing the public, and minimizing episodes of illness are all considered key priorities throughout an outbreak investigation.

Tasked with helping to expedite the investigatory process, Hu and his team developed a new process. Hu theorized that leveraging retail scanner data to spatial information—a type of technology already used my most grocery stores and markets—with confirmed genomic-coded case information at public health agencies, can speed up the investigation process.

Using these tools, Hu and his team conducted a series of experiments using spatial-temporal data, synthesizing retail and geospatial information. They discovered that one of two outcomes is 80 to 90 percent probable:

  • Identifying the source of contamination in the early stages of an outbreak, or
  • Narrowing potential contamination sources to a few “suspect products” prior to outbreak onset

Hu and his team concluded that available electronic retail scanner data—in conjunction with advanced metagenomics laboratory methods—can successfully assist and expedite future foodborne outbreak investigations. Providing a consensus is reached affirming Hu’s findings, this is a potentially-momentous development for all stakeholders. Currently, outbreak investigations can last between a few weeks to a few months, perhaps longer. Currently, IBM is working in partnership with other stakeholders to implement and scale their proposals.