Big data: The next frontier for food safety risk assessment

China is considered as the superpowers of global food production, having undergone instrumental changes over the past 50 years in the food safety regulatory system. The economic importance of the Chinese food sector globally and domestically has put food safety as a national priority for China, as stated in the Healthy China 2030 plan.

Since the identification of food safety as a national priority, China has become one of the key players in modernized food safety approaches. This is evident not only by the update of food safety legislative framework along with organizational change, but also by the increasing number of public and private initiatives fostering food safety modernization. China’s National Center for Food Safety Risk Assessment (CFSA) established the National Food Safety Standards (NFSS) Framework with a clear focus on risk analysis principles and in particular on risk assessment based on scientific information and data collection that reflects China’s context (food production practices and consumer patterns).

But the modernization of food safety and regulatory systems doesn’t end up with having a robust risk assessment framework. The continuously evolving supply chains need adaptive frameworks that allow for timely response to food safety problems and moving from reaction to prevention.

What if we could perform live risk assessment and predict what will be the next critical incident in the supply chain?

A wide variety of tools is used today by Food Safety and Quality Assurance professionals to assess risk ranging from literature-based classification of risks to Excel files and software tools such as the JIFSAN’s iRisk Tool and the University of Tasmania and USDA-ARS Combase. Ingredients and raw materials are analyzed against biological, chemical and physical hazards based on historical data and literature.

One of the main limitations of these approaches is that the estimated risk remains static. It is reviewed on an annual basis or every time that there is a significant change in the supply chain of each company. In our global and complex supply chain, emerging risks and increasing hazards are identified in most cases after the quality or food safety incidents have already affected consumers. This is far too late if we want to prevent risk and not just to react to incidents.

The two key steps that our team considers as the next era of risk assessment are the live risk assessment based on live data streams and the risk prediction using supply chain data and machine learning (computational) models.

Thus far, a large number of computational models have been developed by the research community and can be found in catalogues such as OpenFSMR. The majority of models are hazard- and product-specific and can predict the growth of bacteria in specific conditions. Such models could be fed with incredible amounts of data and (most of the time) live data so they can support real-time risk prediction.

Besides live data stemming from sensors installed in different steps of the supply chain, there is a vast amount of data available from official national and international sources. The plethora of data that is available on a daily basis about food recalls and border rejections can be used to estimate the frequency of hazards and fraud cases in different types of food and beverages. This enables the live risk assessment allowing early highlighting of the increasing and the emerging risks. In that way, a food company can have a live risk for its ingredients, incoming raw materials and product recipes. Moreover, any change in the risk trends can be tracked and all responsible departments are notified so as to deploy corrective actions.

Our team will present details about these new approaches specifically focusing on how live risk can be estimated based on real-time data for food recalls and border rejections from more than 45 data sources.Come and join us in Beijing at the China International Food Safety & Quality Conference (Stand B27) along with 700 regulators, scientists, industry executives, and academics.FOODAKAI, is an intelligent data service that employs intelligent scanning services to collect all the available relevant data and provide support to food safety and quality assurance professionals. With its intelligent assessment services it can perform live risk estimation for a food company’s raw materials, ingredients and finished products.

div#stuning-header .dfd-stuning-header-bg-container {background-color: transparent;background-size: initial;background-position: center top;background-attachment: scroll;background-repeat: initial;}#stuning-header div.page-title-inner {min-height: 100px;}

By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.

Close