SEARCH

SEARCH BY CITATION

Keywords:

  • food defense;
  • interoperability;
  • product tracing;
  • traceability pilots

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methodology
  5. Results and Discussion
  6. Next Steps
  7. Conclusion
  8. References

Despite the best efforts of food safety and food defense professionals, contaminated food continues to enter the food supply. It is imperative that contaminated food be removed from the supply chain as quickly as possible to protect public health and stabilize markets. To solve this problem, scores of technology companies purport to have the most effective, economical product tracing system. This study sought to compare and contrast the effectiveness of these systems at analyzing product tracing information to identify the contaminated ingredient and likely source, as well as distribution of the product. It also determined if these systems can work together to better secure the food supply (their interoperability). Institute of Food Technologists (IFT) hypothesized that when technology providers are given a full set of supply-chain data, even for a multi-ingredient product, their systems will generally be able to trace a contaminated product forward and backward through the supply chain. However, when provided with only a portion of supply-chain data, even for a product with a straightforward supply chain, it was expected that interoperability of the systems will be lacking and that there will be difficulty collaborating to identify sources and/or recipients of potentially contaminated product. IFT provided supply-chain data for one complex product to 9 product tracing technology providers, and then compared and contrasted their effectiveness at analyzing product tracing information to identify the contaminated ingredient and likely source, as well as distribution of the product. A vertically integrated foodservice restaurant agreed to work with IFT to secure data from its supply chain for both a multi-ingredient and a simpler product. Potential multi-ingredient products considered included canned tuna, supreme pizza, and beef tacos. IFT ensured that all supply-chain data collected did not include any proprietary information or information that would otherwise identify the supply-chain partner who provided the information prior to sharing this information with product tracing technology providers. The 9 traceability solution providers who agreed to participate in this project have their systems deployed in a wide range of sectors within the food industry including, but not limited to, livestock, dairy, produce, fruits, seafood, meat, and pork; as well as in pharmaceutical, automotive, retail, and other industries. Some have also been implemented across the globe including Canada, China, USA, Norway, and the EU, among others. This broad commercial use ensures that the findings of this work are applicable to a broad spectrum of the food system. Six of the 9 participants successfully completed the data entry phase of this test. To verify successful data entry for these 6, a demo or screenshots of the data set from each system's user interface was requested. Only 4 of the 6 were able to provide us with this evidence for verification. Of the 6 that completed data entry and moved on to the scenarios phase of the test, 5 were able to provide us with the responses to the scenarios. Time metrics were useful for evaluating the scalability and usability of each technology. Scalability was derived from the time it took to enter the nonstandardized data set into the system (ranges from 7 to 11 d). Usability was derived from the time it took to query the scenarios and provide the results (from a few hours to a week). The time was measured in days it took for the participants to respond after we supplied them all the information they would need to successfully execute each test/scenario. Two of the technology solution providers successfully implemented and participated in a proof-of-concept interoperable framework during Year 2 of this study. While not required, they also demonstrated this interoperability capability on the FSMA-mandated food product tracing pilots for the U.S. FDA. This has significant real-world impact since the demonstration of interoperability enables U.S. FDA to obtain evidence on the importance and impact of data-sharing moving forward. Another real-world accomplishment is the modification or upgrade of commercial technology solutions to enhance or implement interoperability. As these systems get deployed by clients in the food industry, interoperability will no longer be an afterthought but will be built into their traceability systems. In turn, industry and regulators will better understand the capabilities of the currently available technologies, and the technology provider community will identify ways in which their systems may be further developed to increase interoperability and utility.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methodology
  5. Results and Discussion
  6. Next Steps
  7. Conclusion
  8. References

Despite the best efforts of food safety and food defense professionals, contaminated food continues to enter the food supply. It is imperative that contaminated food be removed from the supply chain as quickly as possible to protect public health and stabilize markets. In addition to findings highlighted in the March, 2009, Office of Inspector General Report “Traceability in the Food Supply Chain,” numerous outbreaks have demonstrated that expeditious identification (and removal) of potentially contaminated products is not currently possible. In this study, Institute of Food Technologists (IFT) evaluates the capability of currently available technologies to predict the downstream consequences of an upstream event, or the likely upstream source of contamination given a downstream event, and to explore the interoperability of currently available traceability technology solutions.

Scores of technology companies purport to have the most effective, economical product tracing system. IFT has worked with many product tracing technology providers and, in this study, sought to compare and contrast the effectiveness of these systems at analyzing product tracing information to identify the contaminated ingredient and likely source, as well as distribution of the product. Most importantly, IFT also determined if these systems can work together to better secure the food supply (their interoperability).

IFT also sought to further refine the terms “Critical Tracking Events” and “Key Data Elements” to determine critical points, throughout the food supply chain where product tracing data must be collected, and the appropriate data elements to collect. As a result, industry and regulators will understand the capabilities of the currently available technologies, and the technology provider community will identify ways in which their systems may be further developed to increase interoperability and utility.

The specific objectives of this work were to:

  • Compare and contrast the services and functionality of 9 product tracing technologies, using the same data sets
    • Examine the extent and automation of data quality assurance and standardization for each system
    • Assess the ability of an individual software system to include/exclude potential sources or recipients of contaminated product when incomplete supply-chain data sets are provided
  • Determine the interoperability of product tracing technology systems
    • Assess interoperability when full supply-chain data are provided (in pieces)
    • Assess the ability of the network of software systems to include/exclude potential sources or recipients of contaminated product when incomplete supply-chain data exist
  • Determine how to further develop product tracing systems to increase interoperability and utility
  • Identify critical points throughout the food supply chain where product tracing data must be collected and the appropriate data elements to collect.

IFT hypothesized that when technology providers are provided with a full set of supply-chain data, even for a multi-ingredient product, their systems will generally be able to trace a contaminated product forward and backward through the supply chain. However, when provided with only a portion of supply-chain data, even for a product with a straightforward supply chain, it was expected that interoperability of the systems will be lacking and that there will be difficulty collaborating to identify sources and/or recipients of potentially contaminated product.

Methodology

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methodology
  5. Results and Discussion
  6. Next Steps
  7. Conclusion
  8. References

In order to expand on IFT's initial product tracing research, IFT in collaboration with RTI provided supply-chain data for one complex product to 9 technology providers, and then compared and contrasted their effectiveness at analyzing product tracing information to identify the contaminated ingredient and likely source, as well as distribution of the product. IFT worked with RTI, who attended each meeting related to the task, so that the technological backbone of traceability solutions providers was understood and the methods used by the varying programs to establish data relationships were known.

An expert panel was assembled and met in-person to determine what key data elements (KDEs) are necessary for product tracing and those KDEs were shared with the product tracing technology providers. IFT used its membership and contacts to secure existing (not real-time) supply-chain data for one complex (multi-ingredient) food product and one product with a straightforward supply chain, thus ensuring that the necessary KDEs were part of the information collected. A vertically integrated foodservice restaurant agreed to work with IFT to secure data from its supply chain for both the multi-ingredient and a simpler product. Potential multi-ingredient products considered included supreme pizza and beef tacos. IFT ensured that all supply-chain data collected did not include any proprietary information or information that would otherwise identify the supply-chain partner who provided the information prior to sharing this information with product tracing technology providers.

The 9 traceability solution providers who agreed to participate in this project have their systems deployed in a wide range of sectors within the food industry including, but not limited to, livestock, dairy, produce, fruits, seafood, meat, and pork, as well as in pharmaceutical, automotive, retail, and other industries. Some have also been implemented across the globe including Canada, China, USA, Norway, and the EU, among others. This broad commercial use ensures that the findings of this work are applicable to a broad spectrum of the food system and enabled IFT to design and test a better interoperability framework in Year 2.

A simulated supply chain of a beef taco food product was created as an Excel spreadsheet based on the real-world data supplied by members of the expert panel. A balance was sought between minimizing the complexity of the supply chain compared with preventing one from solving the scenarios without the use of a traceability solution (for example, using Excel filter and search functions). As shown in Figure 1, 6 ingredient suppliers, 4 food processors, 6 distribution centers, and 24 retail distribution center locations were modeled, supplying the food product to approximately 100 retail store locations (based on some degree of randomness). An error rate of 1% was introduced into the data set, including errors of containers shipped before creation; shipped containers never received; and received containers never shipped. The errors served 2 purposes: to evaluate the robustness of the technology (for example, did they identify the errors) and to complicate solving the scenarios manually. A single batch of ground beef from beef supplier 1 was hypothetically contaminated, and its flow through the system was monitored and recorded. This allowed IFT to compare the trace-back and recall information provided by the participants with the expected flow of contamination. A practice test was conducted using one of the traceability solution providers to ensure the completeness of the data set and relevancy of the scenarios. Following the pilot test, the remaining 8 solution providers were given the data set at the same time, followed by approximately 1 wk to feed that data set into their solution, and then were provided 2 scenarios consecutively to test the baseline capabilities of their system.

image

Figure 1. Mock supply chain used to test traceability.

Download figure to PowerPoint

After sharing the multi-ingredient food product data (simulated and real) with 9 product tracing technology providers, they had 2 wk to input data and report back to IFT on various parameters determined by the expert panel, such as time to determine contaminated ingredient, and so on. As a check for this self-reporting, IFT and RTI then conducted in-person meetings with up to 2 representatives from each of the product tracing technology providers so they could walk the research team through the steps taken to input data and achieve results, specifically their ability to predict the downstream consequences of an upstream event or the likely upstream source of contamination given a downstream event. IFT met with each of the technology providers for a 3-h face-to-face meeting where the baseline test, results, challenges, and the future direction of their technology were discussed. Results helped the expert panel to further refine critical tracking events (CTEs) and KDEs required for accurate product tracing.

Since none of the 9 technology solution providers had the ability to interoperate, Year 2 attempted to demonstrate a proof-of-concept on how interoperability could be achieved. The primary technology lacking for interoperability was a communication standard for data interchange. Once the approach and framework was decided upon by all subject matter experts, the 1st step was to formalize the standard that will be used to communicate between different technology solution providers to enable interoperability.

Due to previous experience and the ease of modification for compliance, Electronic Product Code Information Service (EPCIS) was selected as the standard protocol to record and share supply-chain data. For the purpose of this study, GS1 US, a standards setting organization, was brought in as a Subject Matter Expert (SME) to help tweak the existing EPCIS standard.

Results and Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methodology
  5. Results and Discussion
  6. Next Steps
  7. Conclusion
  8. References

IFT evaluated the traceability solutions on 3 success metrics (successful data entry, response to Scenario 1, and response to Scenario 2) and 3 time metrics (time to complete data entry, time to respond to Scenario 1, and time to respond to Scenario 2).

Table 1 summarizes the results of the success metrics. The 1st metric tested for was the successful entry of the simulated supply-chain data set into each technology provider's system. Since no major changes were made to the data set or the scenarios following the feedback received from the pilot test, the pilot test results were treated the same way as the test results for the other 8 technologies. As shown in Table 1, 6 of the 9 participants successfully completed the data entry phase of this test. To verify successful data entry for these 6, a demo or screenshots of the data set from each system's user interface was requested. Only 4 of the 6 were able to provide us with this evidence for verification. Of the 6 that completed data entry and moved on to the scenarios phase of the test, 5 were able to provide us with the responses to the scenarios.

Table 1. Capabilities test-success metrics (√ = completed, X = not completed, N/A = not attempted)
SolutionsDataResponse toResponse to
providerentryScenario 1Scenario 2
1
2
3
4
5XN/AN/A
6XN/AN/A
7XN/A
8XN/AN/A
9

As indicated above, 3 of the 9 technology solution providers failed the data entry phase of the test. The following factors contributed to their failure:

  • – Scope of the system was on internal traceability (within 4 walls of a facility).
  • – The system enabled communication between supply-chain partners during an investigation, but did not store any data itself.
  • – The data set was not in a standardized format.

In addition, one participant was unable to respond to the scenarios because some data elements required by their system were not available within the data set (like the name of the employees that handled the food product).

Table 2 summarizes the results of the time metrics used to compare the capabilities of the systems. The time metrics were useful for evaluating the scalability and usability of each technology. Scalability was derived from the time it took to enter the nonstandardized data set into the system (from 7 to 11 d). Usability was derived from the time it took to query the scenarios and provide the results (ranges from a few hours to a week). The time was measured in days it took for the participants to respond after we supplied them all the information they would need to successfully execute each of the 3 phases listed in Table 2.

Table 2. Capabilities test–time metrics (in days)
Image

It was apparent from this test that not all traceability solution providers had the capability to participate in an interoperability framework proposed for Year 2, and even of those with the capability, IFT concluded from the time metrics that not all traceability solutions were built with scalability and usability in mind. For a truly interoperable framework and the timely resolution of an outbreak investigation, including trace-back and recall, the technologies need to be fast and responsive. The time metric helped further narrow the prospective list of candidates who had the baseline capabilities to participate in Year 2 of this project.

Table 3 summarizes the accuracy of the responses received from the solution providers when compared with the simulated flow of contamination in our model.

Table 3. Accuracy of responses (number of lots identified/actual number of lots affected)
 Scenario 1 trace-forwardScenario 2 trace-back
 Accuracy (lot # identifiedFalse positivesFalse negativesAccuracy (lot
 as contaminated(lot # identified(lot # identified# identified as
 by solution provideras contaminatedas safe whencontaminated by solution
 compared with modeledwhen theythey wereprovider compared
Solution providercontaminated lot #)were not)contaminated)with modeled contaminated lot #)
  1. a

    For all the data fields that have “N/A,” either the solution provider was unable to complete that phase of the test or the results were in a format where further analysis was not possible (for example, a geographic map of the flow of contamination or when responses were provided in PDF format).

1N/AaN/AaN/Aa6/6
2200/200056/6
4130/200670N/Aa
9200/200006/6

IFT manually mapped the flow of products, ensuring accuracy, to determine the “correct” routes, lot numbers, dates, and so on that should have been identified by the technology provider analyses:

  1. Solution Provider 1: For Scenario 1, this solution provider successfully traced back the contaminated lots to their origin. However, the format that was used for presenting the trace-forward results prohibited further analysis. Thus, its accuracy was not evaluated. For Scenario 2, the solution provider successfully identified the likely sources of contamination with 100% accuracy.
  2. Solution Provider 2: For Scenario 1, this solution provider successfully traced back the contaminated lots to their origin. For tracing forward, they identified 5 lot numbers related to 90% lean beef trimmings that were not included in the actual contamination data. Manual searching for those lot numbers did not relate them to the source of contamination. The solution provider successfully identified the correct contaminated lot numbers for 50% lean beef trimmings and the seasoning. For Scenario 2, the solution provider successfully identified the likely sources of contamination with 100% accuracy.
  3. Solution Provider 4: For Scenario 1, this solution provider successfully traced back the contaminated lots to their origin. They also successfully traced forward the contaminated lots to the distribution centers for 90% lean beef, 50% lean beef, and seasoning. However, their results at the retail level did not fully match those manually identified. The summary of their accuracy is given below:
    • Seasoning: total number of lots in the manual response: 80; total number of lots in their response: 59; numbers matching: 56
    • 90% lean beef: total number of lots in the manual response: 40; total number of lots in their response: 18; numbers matching: 18
    • 50% lean beef: total number of lots in the manual response: 80; total number of lots in their response: 59; numbers matching: 56
    • This solution provider was unable to provide a response for Scenario 2
  4. Solution Provider 9: For Scenario 1, this solution provider successfully traced back the contaminated lots to their origin. They also successfully traced forward the contaminated lots to the retail centers for 90% lean beef, 50% lean beef, and seasoning with 100% matching with actual contamination flow. For Scenario 2, this solution provider “only” linked one ingredient (taco shell) to the location ID identified in this scenario. For the selected ingredient, they successfully traced back to the likely source of contamination. However, it was not clear why other ingredients were not listed (such as diced onion, shredded lettuce, seasoned beef).

Two of the 3 technology solution providers successfully implemented and participated in the proposed interoperable framework during Year 2 of this study. While not required, they also demonstrated this interoperability capability on the FSMA-mandated food product tracing pilots for the U.S. FDA in the 1st quarter of 2012. Details about the participants of this supply chain can be found in the IFT report to FDA (McEntire and others 2012). This has significant real-world impact since the demonstration of interoperability enables FDA to obtain evidence on the importance and impact of data sharing moving forward. Another real-world accomplishment is the modification or upgrade of commercial technology solutions to enhance or implement interoperability. As these systems get deployed by clients in the food industry, interoperability will no longer be an afterthought but will be built into their traceability systems. The food industry that provided supply-chain data for the purpose of the NCFPD study have also realized the benefits of participating in an interoperable framework and have started inviting IFT to present at their conferences on this important study.

When discussing the successes and failures of the various participating technology solution providers in the face-to-face meetings, several common overarching challenges emerged that have an impact on interoperability and whole-chain traceability. These challenges are summarized below.

Private sector challenges and opportunities for improved traceability

Several participants noted that significant market drivers for the adoption of traceability technology are mandatory requirements by large companies within the food system, including retailers and processors. At the same time, concerns about confidentiality of data, protection of formulation information, and potential loss of competitive advantage create corporate resistance to more actively engaging in whole-chain traceability. This concern over privacy (as well as cost) was why some small companies within the food system still use a paper-based system for tracking and tracing. However, several traceability solutions can accept faxes or scanned paper-based records and use optical character recognition (OCR) software to recover KDEs and CTEs.

Cost was another major factor preventing small to mid-sized firms from adopting electronic whole-chain traceability. To minimize the cost overhead, several solution providers have designed their systems to work directly with existing internal systems at these firms. A majority of the participants in this study noted that whole-chain traceability is going to be treated as a cost of doing business and generating return on investment rather than just as an expense.

Technology challenges and opportunities for improving traceability

It was clear from these interviews that for any interoperability framework to be successful, it needs to ensure minimal changes in information technology (IT) at the small to mid-sized firms and be capable of importing data via a variety of methods, including:

  • Microsoft Excel
  • Extensible Markup Language (XML)
  • Web services with batch uploads
  • Manual form-based uploads
  • Integration with enterprise resource planning (ERP) systems like SAP
  • Faxed form using OCR
  • Extract-Transform-Load (ETL) approaches

Some solutions use a centralized database approach in which every KDE for each CTE is sent to and collected by the technology. This kind of a centralized approach has some serious scalability and performance issues when attempting to interoperate with a system as complex as the food system where there is a need to store and analyze large data sets.

To mitigate the negative effects of a centralized system, some companies have used cloud/distributed computing for their technology. This fragmented approach works well if all members of the supply chain use the same technology solution provider, but fails to enable true interoperability due to lack of standardization. The lack of standardization not only applies to the type and scope of data collected, but also standards related to data sharing and communication, as identified in this project.

Discovery services are being developed and introduced to overcome this challenge. Such services act as links or interfaces between 2 different traceability solutions and provide a standardized way for them to communicate and share data (for example, electronic product code information services or EPCIS). However, for this approach to be effective, a widespread adoption of those standards needs to be in place, but currently is not.

Finally, the majority of the participants noted that pedigree is an inefficient method to accomplish traceability. Pedigree in food traceability refers to continuously adding CTEs and KDEs to the recordkeeping of the product as it moves through the supply chain, thereby documenting the entire history of the food when it reaches the retailer or consumer.

Whatever the technical approach, each of the 9 systems evaluated could be categorized into one of the following 3 categories:

  • Some solutions work only for internal traceability (inventory and warehouse management systems).
  • Some solutions work only for external traceability.
  • Some solutions work for both internal and external traceability.

Next Steps

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methodology
  5. Results and Discussion
  6. Next Steps
  7. Conclusion
  8. References

To generate better return on investment for the food companies, several traceability solution providers are working toward optional value-added services such as the following:

  • Providing ability to conduct mock recalls through the click of a button within the software
  • Identifying criminal activity, negligence, and counterfeiting for food defense-related incidents
  • Serving as an insurance policy
  • Measuring sustainability metrics such as energy and water usage
  • Providing expiration and shelf-life analysis
  • Tracking temperature abuse (dwell time)
  • Producing transportation reports (shipping and receiving)
  • Using quick-response (QR) coding, videos, and pictures in the database
  • Incorporating hazard analysis and critical control point (HAACP) concepts into the system
  • Collecting weight and volume data for mass balancing
  • Overlaying quality and sustainability metrics on traceability
  • Providing other visualization tools
  • Providing supply-chain optimization tools

To enhance collaboration and interoperability, a majority of the participants in this study identified the following broad areas as challenges that need to be resolved:

  • Define a minimum data set needed to enable connectivity (KDEs and CTEs)
  • Identify an approach to protect data and maintain privacy while sharing data with supply-chain partners and regulatory agencies (when appropriate)
  • Define governance of data as well as standards to adopt; establish standards of data formats and interchange protocols
  • Design the interoperable framework to accommodate a wide variety of platforms, technologies, and business practices and to be inclusive rather than exclusive of anyone willing to follow the standards identified and KDEs and CTEs required
  • Address scalability of existing systems
  • Address the need for business practices to change in order to implement whole-chain traceability (there is no technical limitation)

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methodology
  5. Results and Discussion
  6. Next Steps
  7. Conclusion
  8. References

The significance of this study is improved understanding of the capabilities of product tracing technologies, and identification of how these systems can have increased interoperability and utility. This will allow for faster, more efficient product tracing, which in turn will allow the food industry and governments to identify and remove unintentionally or intentionally contaminated food from the supply chain as quickly as possible to lessen catastrophic impacts on public health. Given the difficulty in conducting outbreak and trace-back investigations from food safety incidents, where the food and agent is typically known, it would be exponentially more complex to resolve outbreaks due to food defense incidents, where the food or the agent or both are unknown. Interoperability, if implemented pragmatically, helps the industry, regulators, and consumers respond with more visibility, agility, and trust in the food system.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methodology
  5. Results and Discussion
  6. Next Steps
  7. Conclusion
  8. References