Food safety in supply chains is of utmost importance. As the food product changes hands from raw material to finished product, it goes through several touchpoints, including production, transportation,¹ storage, and packaging, increasing the risk of contamination and fraudulent activities. The more complex the supply chain, the higher the probability of such risks. Without accountability and transparency among different stakeholders, including the tier 1 and tier 2 suppliers, it becomes challenging to avoid or respond to food safety incidents. The risk has increased significantly with greater offshoring to different markets, including developed and emerging markets.

Accordingly, to enable higher accountability and traceability in the food supply chain, the U.S. Food and Drug Administration (FDA) enacted the Foreign Supplier Verification Program (FSVP)² via the Food Safety Modernization Act (FSMA) in 2016. Through the FSVP requirements, food importers are now liable for their foreign-based suppliers. According to FDA, they must ensure that the "…foreign suppliers are producing food in a manner that provides the same level of public health protection as the preventive controls or produce safety regulations." Importers are now responsible for hazard assessments, risk evaluation, supplier audits, supplier performance assessments, and the development of written procedures, along with extensive recordkeeping.

Accordingly, how do U.S.-based firms deal with such increased responsibilities? Also, how do they enable themselves to reduce adverse consequences for themselves, public health, and the general economy? As per the blueprint of FDA's New Era for Smarter Food Safety initiative,³ one of the core principles of modern-day food safety will be "technology enabled." According to FDA, "The world is rapidly becoming more digital. Advances in artificial intelligence, the Internet of Things, sensor technologies, and blockchain are improving business processes. New digital technologies offer the potential to help us predict and prevent food safety problems and better detect and respond to problems when they do occur." Accordingly, through this article, we assess how the move toward digitalizing supply chains may help food firms achieve FSVP compliance.

What are Digital Supply Chains?

Digital supply chains utilize digital technologies and data analytics to shape the decision-making process, enhance performance, and promptly respond to rapidly changing conditions.⁴ Digital technologies offer an array of tools to modernize supply chain operations, revolutionizing the way companies have been managing their supply chains traditionally. It not only provides an opportunity to update existing supply chain networks, but it also forms basic building blocks for future technologies. Recent estimates suggest that the digital supply chain market is expected to reach $13.7 billion USD by 2030 from a $3.9-billion-USD valuation in 2020.⁵

It is high time that companies committed to operational excellence in supply chain execution should move toward a long-term and short-term digital strategy. With ongoing uncertainties caused by the recent U.S.-China trade war, the Russo-Ukrainian war, the COVID-19 pandemic, and other events, spending on digital supply chain technologies could pay off in building a resilient supply chain network.

While many industries have embraced the digital supply chain and reaped its benefits, the food industry, unfortunately, has lagged in leveraging these technologies. This has been due to several factors, outlined below.

Complexity of Regulations

According to one estimate, it takes more than 200,000 regulatory restrictions from farm to fork to produce one ounce of beef in the U.S. (as of 2021), compared to fewer than 50,000 restrictions in 1970.⁶ These restrictions impact not only farm production but also the processing, transportation, storage, and retail of food products. The burden of such regulatory restrictions is a double-edged sword that can accelerate and impede the pace of technology adoption among food supply chains.

Consumer Activism

Heightened consumer awareness and sensitivity to food-related safety incidents have impacted how food firms respond to technology adoption. Consumer activism challenges the practices of food businesses to change corporate practices for good. This increased social pressure from food activists lead to a different dynamism within the food industry. This level of scrutiny affects the ability of food firms to experiment with such technologies.

Low Gross Profit Margins

The increased costs of transportation, storage, and procurement of perishable products, along with high labor costs (especially in cold chains), as well as the associated regulatory burdens, have decreased the profit margins of the food industry significantly. To maintain competitiveness, food firms constantly compromise on margins. According to one estimate, the food industry experienced a gross profit margin of about 22 percent in 2019, compared to approximately 49 percent for the rest of the market.⁷

Perishable Nature of Goods

The food industry deals with perishable goods that have a limited shelf life. This poses challenges in maintaining product quality, managing inventory, and ensuring timely delivery. The complexity of handling perishable items adds a layer of difficulty in implementing digital solutions effectively. The need for such procedures extends beyond the FSVP. For example, the lengthy customs clearance process and economic impact caused by perishable products have led many countries to use a single-window system, where digital information is shared to facilitate the prompt trade of perishable goods.

Fragmented Supply Chain

Food supply chains are often broken into several small and informal value chains, including several small players. This fragmentation results in various complex layers that can make it hard to trace the products from their place of origin to the end consumers. Primarily, food supply chains in emerging economies are entangled, often resulting in silos that are difficult to integrate into the mainstream digital ecosystem. Digitization requires considerable time, money, and effort, making it challenging for small firms to achieve.

Resource Constraints

Most businesses in the food industry are small- and medium-sized enterprises struggling to keep up with large firms. As discussed, they are already working to keep afloat and face resource limitations in budgeting, retaining, and attracting qualified employees, which makes their technology learning curves much steeper. These firms also need help accessing loans, insurance, and other investments. Moreover, technology adoption is a function of economies of scale, limiting food firms from multiplying the impact of technology assimilation.

Despite these challenges, the benefits outweigh the costs. Digital technologies are playing a significant role in shaping the landscape of food supply chains, especially compliance with FSVP. These technologies enable food firms to achieve transparency, traceability, and accountability at the heart of FSVP requirements. The digital supply chain toolkit comprises several technologies; this article focuses on some of the major ones, discussed below.

Digital Supply Chain Solutions for Complying with the FSVP

Blockchain

Blockchains are shared and immutable ledgers that are used to record business transactions. They are linked to peer-to-peer networks where each transaction is uniquely recorded. Blockchain has some unique properties, such as once data is recorded, it is almost impossible to change it. Each block has data information, a unique code, and a code of its previous block. Data stored inside the blockchain depends on the type of blockchain. Once a code has been generated, changing any information will change the foundation of the block and disrupt its chain of information.⁸

Blockchains allow for improved visibility and traceability⁹ across the supply chain, essential for FSVP compliance. As a result of improved access to accurate information, tamper-proof storage of digital records, and digital access to this information, blockchains enhance the traceability of products from their origin to the consumer with verifiable proofs. This helps increase the bargaining power of firms, improve the credibility of information-sharing, and negotiate transparent contracts. Transparent access to reliable information helps firms better manage their inventory.

One of blockchain's most important benefits for fulfilling FSVP requirements is that it provides end-to-end data encryption. This helps firms avoid food fraud and safeguard their supply chains from counterfeit products. Once generated, it is tough to alter these information blocks, and any alteration triggers a new blockchain, keeping the original information intact. Furthermore, since this is a peer-to-peer open network system, the data on the blockchain is stored across a network of computers, making it impossible to alter any information via cyberattack.

Blockchains eliminate the use of intermediaries, thereby significantly increasing the rate of information gathering, flow, and access. They help digitize the record and eliminate any manual processing of information. For example, in case of a recall or a food safety incident, a firm can trace the problem to its source of origin in seconds. The speed of information access within the supply chain decreases information asymmetry among the firm and its suppliers, leading to more confident transactions. Firms can safely meet the regulatory requirements of recordkeeping and monitor suppliers more accurately using blockchains.

Overall, blockchain can help firms improve transparency and traceability. They can better rely on auditable information. Blockchain also helps increase stakeholder confidence and trust in the mutual relationship. The immutable nature, decentralized approach, and increased privacy of blockchain make it an intelligent choice for FSVP compliance.

Artificial Intelligence (AI) and Machine Learning (ML)

Artificial intelligence (AI), in simple words, is machine intelligence used to mimic human intelligence. On the other hand, machine learning (ML) is a sub-field of AI that comprises models that train themselves without being programmed by humans.¹⁰ These capabilities have already found significant applications in the areas such as natural language processing, machine vision, speech recognition, etc. For example, the rise of ChatGPT¹¹ as an AI chatbot that can synthesize existing information on the internet and intelligently interact with its user is a testament to the potential of AI models. Responsible use of these models can help firms comply with the requirements of FSVP in several ways.

In the 21^st century, the problem is not the need for more data, but rather over-information. With technology, firms are generating or collecting a huge amount of data without knowing how to drive actionable insights from it. Managers in the food industry face the challenge of accessing the correct information at the right time. AI and ML models can help make sense of large data streams by helping identify patterns, highlighting critical information, and pinpointing data inaccuracies and duplications. For example, AI and ML models can be used to understand legacy databases, help process certification records, streamline written procedures, and help optimize supply chains for FSVP compliance.

Furthermore, AI and ML models can be trained to use historical data points and help understand patterns to make predictions. These forecasts can be much more precise than the comparable alternatives. AI and ML have vast potential in inventory optimization and procurement decision-making, especially with perishable products. These tools can help warn supply chain managers of potential glitches in the value chain that can help them develop mitigation plans. AI and ML can also help with strategic sourcing decisions such as multi-sourcing, nearshoring, offshoring, insourcing, or co-sourcing arrangements for FSVP compliance. They can also help firms pre-emptively evaluate existing and potential suppliers by assisting them in understanding the risks to food safety and quality.

In addition, AI and ML tools can help augment human judgment on quality inspections. For example, machine vision models enabled by AI and ML can precisely detect bacterial colonies¹² in food. This is a much faster and more efficient way of pathogen detection compared to more expensive and time-consuming laboratory procedures. Machine vision can also identify different types of bacteria and tell them apart based on their features. Such optical capabilities can aid in understanding the patterns of food composition and help increase the confidence of both consumers and producers in the food supply chain.

In summary, AI and ML models have enormous potential to process vast data streams and detect anomalies during food inspections. These models hold great promise in augmenting human judgment by helping them process accurate information more efficiently.

Customer Relationship Management (CRM) Tools and Automation Software

Customer Relationship Management (CRM)¹³ tools help firms manage their customer interactions and relationships with clients, as well as manage end-to-end supply chains by focusing on upstream and downstream stakeholders. This technology can also optimize the sales experience by enhancing visibility into customer data and improving the efficiency of contract management and performance. There are several advantages of using CRM tools for FSVP compliance, including streamlining the flow of information, creating a central repository of information, and efficient task management.

CRM tools help integrate several information channels.¹⁴ They enable effective follow-up on potential leads by profiling customers and suppliers. They also integrate website information, telephone, chats, and social media data points. CRM tools can help employees with external-facing roles to be more effective in relationship management by providing intelligent insights. They also help with better data organization and act as a centralized database for supplier information.

Furthermore, CRM tools make sharing information with internal and external stakeholders much more convenient by giving them access to this central repository of information and helping them improve team collaboration. This enhanced collaboration leads to better coordination among employees and suppliers and effective FSVP compliance. It enhances communication with the stakeholders by automating reminders, task follow-ups, and email generation. Such automation not only takes away non-value-added tasks from employees, but it also creates more accurate follow-up and helps generate automated reports on supplier and customer communication, thereby improving the visibility and efficiency of reporting.

Overall, CRM tools are an effective way to automate redundant tasks while managing relationships with external stakeholders. They help communicate effectively with suppliers and customers, improving coordination among partners and leading to efficient compliance with FSVP requirements.

Robotic Process Automation (RPA)

Robotic process automation (RPA)¹⁵ is a software technology that can be used by businesses to create "bots" that can replicate human efforts, especially for rule-based, mundane tasks. RPA can help food safety managers automate data entry processes for digital recordkeeping, update written procedures in real time, and generate reports for audits and inspections.

One of the best uses of RPA for FSVP compliance can be to automate data entry tasks that humans otherwise carry out. One of the requirements of FSVP is enhanced recordkeeping. Firms must maintain data records for all business transactions and activities. These recordkeeping requirements could include data on supplier audits, supplier performance, inventory levels, warehouse temperatures, incoming material details, etc. For example, RPA can automate data entry tasks by enabling bots to scan bill of lading paperwork, extracting required information such as the type and volume of products being transported and shipper information.

RPA bots can also update written procedures required for production. For example, in a production unit, due to demand and supply uncertainty, RPA bots can automatically adjust recipes based on these daily production plans. The operators would not need to manually adjust the quantity order for each sub-unit of ingredients that go into the final product. The RPA bots can also communicate with the equipment to adjust temperature, pressure, and humidity requirements for the quantity of products produced.

Finally, RPA bots can help generate real-time reports for audits and inspections. For example, during a surprise visit, a consumer safety officer may ask for a summary of food safety performance over the past few weeks. Instead of grappling with the information, an RPA bot can generate real-time performance reports for the inspector. The RPA-generated pieces can have pre-populated fields connected to multiple databases to help gather, append, clean, and analyze data to produce insightful reports in seconds.

In summary, RPA is a powerful technology to automate several mundane yet rule-based tasks. It can help bring efficiency to otherwise non-value-added tasks and help operators and managers focus on more value-added activities. Beyond food safety, the RPA bot can automate procurement order generation, inventory management, and many other tasks.

Internet of Things (IoT) Solutions

IoT solutions¹⁶ comprise physical devices, such as sensors, actuators, and vision systems, that gather information from the real world and transmit that information to the digital world via the internet. These sensors are the eyes and ears of the digital world in the real world. They are feedback loops through which the digital ecosystem adapts its responses. They enable FSVP compliance by helping firms increase real-time visibility into critical parameters of their products and processes.

One of the most significant advantages of IoT devices is that they can gather real-time information from products and processes. For example, milk containers can be installed with temperature and humidity sensors to monitor the variation in these two parameters from the farm to the processing plants. These IoT devices can log information locally and push the data online. Users can program these devices to adjust the frequency of data collection. Such information can be used to ensure product integrity and delineate potential deviations in the process. This can help firms take appropriate remedial or proactive actions, as required.

Furthermore, these IoT devices can ensure protection against food fraud and economically motivated adulteration. For example, these devices can be used as door alert sensors to record the times and duration for which the storage room door was accessed. Similarly, other sensors, such as vibration sensors, can record the transportation conditions of the food and create a data log for recordkeeping. Another example is IoT devices for odor and gas monitoring in warehouses and production units. These sensors can alert operators and managers about potential risks to products and processes.

IoT devices combined with IoT platforms can gather, compile, and visualize data in real time. Integration with IoT platforms can enable firms to develop data dashboards to monitor performance across critical control points. Such data can be used for compliance, audit reporting, and supplier monitoring. Overall, they can help comply with FSVP requirements by assisting firms to monitor suppliers and their activities from anywhere and anytime.

Augmented Reality (AR)/Virtual Reality (VR) and Digital Twins

Augmented Reality (AR) is an experience that combines real-world content with virtual content generated with the help of computers. AR can help improve the real-world experiences of humans by augmenting reality with different sensory experiences, such as visual or olfactory. On the other hand, in Virtual Reality (VR), a human is immersed in a simulated environment using headsets or other wearable technology.¹⁷ Furthermore, digital twins are a virtual representation of real-world objects or systems that are updated in real time throughout their lifetimes. Coupled with digital twins, AR and VR can help improve human efficiency.

AR technology can be used for the repair and maintenance of machines. Using headsets or intelligent devices such as smartphones, operators can overlay schematics and drawings on the machines to troubleshoot them efficiently. It can also help technicians refer to the service manuals in real time. This can reduce production downtime and help operators avoid errors. Moreover, using live feed, equipment service providers can generate precise instructions by guiding the in-house operators to conduct efficient repairs.

The disruption of the COVID-19 pandemic accelerated the use of virtual inspections and auditing. Remote audits and inspections can be significantly improved via AR/VR technologies. Inspectors can use AR/VR headgear or smartphone applications to digitally enter and assess buildings, storage spaces, and vehicles. AR headsets can help remote inspectors provide exact directions to the operator on the ground for a comprehensive assessment. Although virtual inspections are not intended to replace in-person inspections, they can help increase monitoring of suppliers.

VR headsets can be used to train new employees in a virtual environment where they can experience the machines and processes. They can also be used to conduct simulations to ensure compliance with safe operating procedures, and they can help in identifying potential risks and developing mitigation plans. Simulation of emergency response plans can help prepare employees for high-risk, low-probability events. VR headsets can also be used to simulate regulatory inspection experiences.

In summary, AR and VR technologies and digital twins can help humans experience the world more efficiently. Regarding FVSP requirements, AR and VR technology can aid in auditing suppliers remotely, as well as allow access to expert technicians and service providers remotely.

Summary

In a nutshell, digital supply chains hold great promise for FSVP compliance. They have the potential to strengthen the foundations of food firms by helping them build transparency, traceability, and accountability. Technologies like AI and ML can help firms improve risk evaluation by using legacy data to pre-empt issues. Blockchain, with the use of tamper-proof ledgers, can help generate more confidence in the supply chains. AR and VR can help with remote supplier audits. Technologies such as RPA can automate mundane recordkeeping tasks and document written procedures. IoT devices can help generate real-time information, leading to enhanced visibility across the internal and external stakeholders. These benefits far outweigh the costs of assimilating these technologies. Digital supply chains are a win-win solution for FSVP compliance and beyond.

References

1. Grover, A. K. "Out of the frying pan and into the fire? Uncovering the impact of FSMA's sanitary food transportation rule on the food logistics industry." Business Horizons 66, no. 2 (2023): 203–214.
2. FDA. FSMA Final Rule on Foreign Supplier Verification Programs (FSVP) for Importers of Food for Humans and Animals. Current as of January 10, 2023. https://www.fda.gov/food/food-safety-modernization-act-fsma/fsma-final-rule-foreign-supplier-verification-programs-fsvp-importers-food-humans-and-animals.
3. FDA. "New Era of Smarter Food Safety: FDA's Blueprint for the Future." 2020. https://www.fda.gov/media/139868/download.
4. Jenkins, Abby. "Digital Supply Chain Explained." Oracle NETSUITE. 2022. https://www.netsuite.com/portal/resource/articles/erp/digital-supply-chain.shtml.
5. Allied Market Research, "Digital Supply Chain Market by Component, Organization Size, and Industry Vertical, and Region: Global Opportunity Analysis and Industry Forecast, 2023–2030." 2022. https://www.alliedmarketresearch.com/digital-supply-chain-market.
6. Malone, T. and Staples, A. "Regulatory restrictions are making food supply chain disruptions worse." The Hill. November 20, 2021. https://thehill.com/opinion/energy-environment/583454-regulatory-restrictions-are-making-food-supply-chain-disruptions/.
7. Segal, T. "Profit Margins for the Food and Beverage Sector." Investopedia. February 28, 2023. https://www.investopedia.com/ask/answers/071015/what-profit-margin-usual-company-food-and-beverage-sector.asp.
8. Simply explained. "How Does a Blockchain Work—Simply Explained." 2017. https://www.youtube.com/watch?v=SSo_EIwHSd4&t=201s.
9. Pratt, Mary K. "Top 10 benefits of blockchain technology for business." June 2, 2021. https://www.techtarget.com/searchcio/feature/Top-10-benefits-of-blockchain-technology-for-business.
10. Microsoft Azure. "Artificial intelligence (AI) vs. machine learning (ML)." 2023. https://azure.microsoft.com/en-us/resources/cloud-computing-dictionary/artificial-intelligence-vs-machine-learning/#introduction.
11. Hu, Krystal. "ChatGPT sets record for fastest-growing user base—analyst note." February 2, 2023. https://www.reuters.com/technology/chatgpt-sets-record-fastest-growing-user-base-analyst-note-2023-02-01/.
12. Dobbyn, Tiffany. "How Artificial Intelligence May Improve Food Safety." February 24, 2023. UC Davis. https://www.ucdavis.edu/blog/how-artificial-intelligence-may-improve-food-safety.
13. Salesforce. "CRM 101: What is CRM?" 2023. https://www.salesforce.com/crm/what-is-crm/.
14. Zoho. "What is CRM? A guide to CRM software by Zoho CRM." 2015. https://www.youtube.com/watch?v=hnEQq7kNFWo.
15. Automation Anywhere. "Robotic Process Automation (RPA)." 2023 https://www.automationanywhere.com/rpa/robotic-process-automation.
16. Arm. "Glossary: IoT Devices." 2023. https://www.arm.com/glossary/iot-devices.
17. Tulane University. "What's the difference between AR and VR?" 2023. https://sopa.tulane.edu/blog/whats-difference-between-ar-and-vr.

Abhay Grover, M.S., is a Ph.D. candidate in Supply Chain Management at the Robert H. Smith School of Business at the University of Maryland, with a research focus on food safety and quality management in agrifood value chains. With over seven years of experience working with Fortune 200 firms such as John Deere and Whirlpool Corp., he has collaborated closely with suppliers, production, and customers to drive product and process improvements. He has also provided consultation on Industry 4.0 technology adoption for manufacturers, including food facilities, at the Center for Industrial Research and Service (CIRAS) at Iowa State University. Abhay is a Preventive Controls Qualified Individual (PCQI) for Human Food. His research has been published in several international journals, and he has received several awards for his research and industry practice.

Geetanjali Menon, B.S., is a student in the master's program in Business Analytics at the Robert H. Smith School of Business at the University of Maryland. Previously, she worked at Deloitte as a technology consultant, specializing in developing cloud-based document management solutions. She is also a certified AWS Practitioner. Her areas of expertise include product development, requirements engineering, technology product management, and data science.

Clare Narrod, Ph.D., is the Director of the Risk Analysis program at the Joint Institute for Food Safety and Applied Nutrition (JIFSAN), which is a public-private partnership between the Food and Drug Administration (FDA), the University of Maryland (UM), and the private sector. She leads the monitoring and impact effort associated with the evaluation of JIFSAN's capacity-building efforts. Prior to coming to JIFSAN, she worked at the International Food Policy Research Institute, the U.S. Department of Agriculture, and at the Food and Agriculture Organization. She has consulted for the World Bank and the Inter-American Institute for Cooperation on Agriculture.