How Big Data Analytics is Benifiting: Supply Chain Business
Towards data science
While loT technology is a powerful tool that can improve the efficiency of food manufacturing businesses, some of its greatest potential is found in its ability to collect and interpret data to improve operations. Data science, also known as data analytics, involves the collection of data to help identify customers' needs with immense accuracy to help businesses grow and succeed. Business functions such as product develop-ment, inventory levels, delivery procedures, and employees' staff schedules can all be improved when businesses harness the power of data science. Fortunately, the extensive food industry holds massive data waiting to be pro-cessed and analyzed, providing undeniable advantage to those who apply it. Figure 2 shows how big data analytics are able to benefit supply chain businesses.
Improvement in customer service and demand fulfillment46%
Faster and effective reaction time to supply chain issues41%
Increase in supply chain efficiency of 10% or greater36%
Greater integration ecross the supply chain36%
Optimization of inventory and asset productivity33%
Better customer and supplier relationships28%
Improvement in demand-driven operations20%
Shortened order-to-delivery cycle times14%
- Figure 2. How big data analytics are able to benefit supply chain businesses
Source: ByteAnt
Being able to analyze data is especially impor-tant to ensure products meet certain quality control standards as it can cover all the informa-tion regarding price, condition, and quality of products. By analyzing components along the entire supply chain, data analytics can help Identify ways to improve quality control. Con-sumers have become more demanding by wanting to know how food was processed..how the livestock was treated, and what chemicals were used in the food. Data science, particularly data visualization, can build transparency within supply chains by allowing customers to under-stand where each supply comes from allowing more confidence in the quality of the product. Being able to control nearly every aspect of food production can help food producers produce high-quality products and minimize food-related health hazards.
At the same time, predictive analysis can help detect contamination in food packaging by tracking food temperatures beyond their in-house operations. This data can be used to help identify temperature fluctuations and predict potential contamination issues, conse quently avoiding food recalls.
Data science does not just focus on production and supply chain processes, data science also involves analyzing market trends and consumer patterns that can help identify stock and menu choices that customers will naturally gravitate towards. Data, such as when products are relevant, where target customers spend their time, what marketing platforms can cater to a product, and factors that affect a customer's decision to buy can help assist in designing effective and targeted marketing campaigns.
Carrefour, Europe's largest retailer, imple-mented blockchain traceability to track the diffe-rent production stages of free-range chickens across France. Walmart, on the other hand, part-nered with IBM and Tsinghua University to run pilots using blockchain and loT sensors to track pork from China and Mexican mangoes shipped to the United States. These trials reduced the time to track supply chain information from one week to just a few seconds.
While it may take years to develop, the long-term vision is for companies to produce fully transpa-rent food supply chains. With an industry with over 500 food recalls annually in the United States alone, blockchain's transparency makes it possible to identify the safety of food produc tion by accessing food safety information should an outbreak or contamination occur at any point.
Subsequently, machine leaming is a method of data analysis that automates analytics model building. As a branch of artificial intelligence, this method is based on the idea that systems can learn from data and industry patterns to make decisions with minimal human interven-tion. Optimum sorting solutions, supply chain optimization, predictive maintenance, and self-serving robotic solutions are some examples of machine learning applications. With the immense amount of data present in the food industry, machine learning allows computers to analyze data and make recommendations based on input data.
Sentiment analysis is the process of analyzing a customer's inclination, emotions, and feelings towards a band based on data gathered from different social media platforms. Technologies such as natural language processing allow data analysis tools to categorize text gathered into positive, negative, or natural reviews. Marketing campaigns then become less about guesswork and more about studying customer's needs.

The more notable and state-of-the-art technol-ogy involved in data science is predictive ana-lytics. Using the unique capacity of Artificial Intelligence (Al) algorithms, predictive analysis is capable of determining and forecasting possible outcomes and issues from gathered data allowing businesses to correct and adjust strategies before undesirable outcomes. For example, predictive analysis allows manufac-turers to identify time-to-repair and cost-to-re-pair indicators through categorizing issues and predictive alerts, being able to save up to 50% maintenance time and reduce costs by 10%.
From supply chain to blockchain
Food supply chains are vast and spread over the world making it crucial yet nearly impossible to trace the entire network. Fortunatelyihlockchain technology allows a powerful opportunity to trace the lifestyle of food products from every point of contact from farm to table. By a simple scan of a QR code, users can get access to relevant information in just a matter of seconds.
Rapid alert systems
When public health is at risk, the flow of informa-tion to enable swift reaction when an outbreak or contamination occurs is of utmost importance.
Once presented with that information, knowing what to do with that information is crucial.
Created and operated by the European Commission in 1979, the Rapid Alert System for Food and Feed (RASFF) was put into place to provide food and feed control authorities with an effective way to exchange information between Member States in case of any risk detected in food or feed. The online system allows the Commission, European Food Safety Authority (EFSA), European Economic Area (EEA), Norway, Liechtenstein, Iceland, and Switzerland to submit notifications on the withdrawals of unsafe or illegal products in the market allowing Member States to act more rapidly and in a coordi-nated manner in response to a safety threat. The RASFF portal features an interactive online database that gives access to summary information as well as a search function for information on any notifi-cations issued in the past.
On the other hand, the RASFF consumers' portal was launched in June 2014, which allows consumers access to the latest information on food recall notices, and public health warnings issued by foo safety authorities and companies. In the same way, consumers are able to identity food that has been tagged in the system allowing them to make more informed choices. Thanks to RASFF, many food safety risks have been averted even before they could be harmful to European consumers.
In the Philippines, the Rapid Alert System for Food & Feeds (Phil RASFF) was established, which is a web-based system tool for rapid exchange of food and feed safety information among competent authorities involved in food safety control. Phil RASFF was modeled from the ASEAN RASFF and EU RASFF to institutionalize an alert and notification system for food and feed safety concerns, and improve coordination amongst food safety control agencies at national and regional levels.
Auditing goes remote
While food safety relies on straightforward principles, its verification requires a complex system of specialties, auditors, and certification bodies to ensure processes are practiced consistently and appropri-ately. Food business certification bodies have explored new ways to gather, analyze, and report data while keeping auditors and personnel safe. Since physical visits have become more difficult today, auditors have migrated to hybrid and remote auditing arrangements.
Remote video auditing through installed cameras is an approach that analyzes segments of live and/or recorded video feeds selected at random by a remote auditor. According to an article entitled "The Rise of Remote Food Safety Audits" from Food Quality & Safety, cameras deployed by various locations within the operation area are conti-nuously running, or at least transmitting a video feed during opera-tional hours.
Another remote auditing approach is via walk-through video where an employee may facilitate an audit by recording a walkthrough of the operations from receiving goods until dispatch of the final products. Remote auditors have used different devices such as hand-held cameras, helmet-mounted cameras, or even glasses equipped with cameras to carry out this type of audit.
Figure 3 shows how the different certifications programs recognized and implemented different remote audits during the pandemic.
