CFSAN logo

Risk-Ranking Model for Food Tracing

This site is best viewed using current versions of Google Chrome, Mozilla Firefox or Microsoft Edge.

The FDA Food Safety Modernization Act (FSMA) section 204 (21 U.S. Code § 2223), requires the Food and Drug Administration to designate high-risk foods for which additional recordkeeping requirements are appropriate and necessary to protect the public health.

The FDA developed a Risk-Ranking Model for Food Tracing (RRM-FT) as a data-driven science-based decision support tool to assist the Agency in the process of designating a Food Traceability List as required by FSMA Section 204.

This website is designed for users to view the RRM-FT results as well as the risk-ranking methodology. Users can navigate the RRM-FT website through the following tabs:

  • Methods and Criteria: Click on the “Methods and Criteria” tab to view a description of the seven criteria in the RRM-FT, to download the “Methodological Approach to Developing a Risk-Ranking Model for Food Tracing FSMA Section 204” document, and to download the full list of references evaluated and used to obtain data for the scoring of the seven criteria for commodity-hazard pairs in the RRM-FT. This tab also provides an example of how a risk score is calculated for a commodity-hazard pair or for a commodity, as well as examples of how criteria scores are calculated;
  • The methods document “Methodological Approach to Developing a Risk-Ranking Model for Food Tracing FSMA Section 204 (21 U.S. Code § 2223))” is downloadable here.

    Download the current methods document

    The full list of references evaluated and used to obtain data for the scoring of the seven criteria for commodity-hazard pairs in the RRM-FT is downloadable here.

    Download the full list of references

  • RRM-FT Results: Click on the “RRM-FT Results Table” tab to view a list of the RRM-FT commodity categories, commodities and commodity scores, and commodity-hazard pairs and their scores. Interactive graphs are provided in the “RRM-FT Results Figure for Commodity-Hazard Pairs” and the “RRM-FT Results Figure for Commodities” tabs.
  • The results can be downloaded at the Commodity level or at the Commodity-Hazard pair level.


The RRM-FT is a semi-quantitative risk-ranking model that scores commodity-hazard pairs according to data and seven criteria. Click on a Criterion to obtain a highlight of the scoring definition, the data sources and an example.

These criteria are consistent with the requirements in section 204(d)(2)(A) (21 U.S. Code § 2223(d)(2)(A)).

The methods document “Methodological Approach to Developing a Risk-Ranking Model for Food Tracing FSMA Section 204 (21 U.S. Code § 2223))” is downloadable here.

Download the current methods document

The full list of references evaluated and used to obtain data for the scoring of the seven criteria for commodity-hazard pairs in the RRM-FT is downloadable here.

Download the full list of references

Calculation of the Risk Score for a Commodity

To calculate a risk score for a commodity (fictitious example: “Commodity A”), we first obtain the risk score(s) for the commodity-hazard pair(s) (e.g. “Commodity A - Hazard 1”, “Commodity A - Hazard 2”, “Commodity A - Hazard 3”, etc.) and then aggregate the commodity-hazard pair risk score(s) into a commodity risk score.

1. Commodity-Hazard Pair Risk Score

A risk score for a commodity-hazard pair is calculated by summing weighted criteria scores across all seven criteria. Equal weighting (criteria weight = 10) was used for each criterion (see § 5.4 of the method document). For example, say that the pair “Commodity A - Hazard 1” has criterion scores as follows:

Commodity Hazard Criterion 1: Frequency of outbreaks and occurrence of illnesses Criterion 2: Severity of illness Criterion 3: Likelihood of contamination Criterion 4: Growth potential, with consideration of shelf life Criterion 5: Manufacturing process contamination probability and industry-wide intervention Criterion 6: Consumption Criterion 7: Cost of illness
Commodity A Hazard 1 9 9 1 3 3 9 9

The Risk Score for “Commodity A - Hazard 1” is then:

Commodity-Hazard Pair Risk Score = (10 × 9) + (10 × 9) + (10 × 1) + (10 × 3) + (10 × 3) + (10 × 9) + (10 × 9) = 430

2. Commodity Risk Score

For a commodity associated with multiple hazards, i.e., multiple commodity-hazard pairs are identified, a Commodity Risk Score is calculated through an aggregation method according to the following formula:

$$RS_i\_C=10\times\log_{10}(\sum_{j=1}^{n_i}10^{RS_{i,j}/10})$$

where,

RSi_C is the aggregated risk score associated with commodity i;

RSi,j is the risk score for commodity i and the jth hazard;

ni is the number of hazards associated with commodity i.

(The division of the individual risk scores by 10 is for computational reasons. To return to the original scale, the result is then multiplied by 10.)

For example, say that the commodity “Commodity A” is associated with the following commodity-hazard pairs:

Commodity Hazard Criterion 1: Frequency of outbreaks and occurrence of illnesses Criterion 2: Severity of illness Criterion 3: Likelihood of contamination Criterion 4: Growth potential, with consideration of shelf life Criterion 5: Manufacturing process contamination probability and industry-wide intervention Criterion 6: Consumption Criterion 7: Cost of illness Commodity-Hazard Pair Risk Score
Commodity A Hazard 1 9913399430
Commodity A Hazard 2 1939391350
Commodity A Hazard 3 0939390330
Commodity A Hazard 4 0910190200

The Commodity Risk Score is then:

Commodity Risk Score = 10 × [log10 (1043+1035+1033+1020)] = 430.

The Commodity Risk Score is driven by the risk score for the highest-ranking commodity-hazard pair.


Click on a row to open a sub-table that shows the commodity-hazard pairs evaluated in that commodity including their model criteria scores. Users can search for a commodity category or commodity of interest by typing the name in the search box. Users can visualize the results in alphabetical order by commodity category and by commodity, and in numerical order by commodity score by clicking the corresponding column name.

The results can be downloaded at the Commodity level or at the Commodity-Hazard pair level.


View the full graph by selecting “All” for Commodity and Hazard. View a sub-graph by selecting one Commodity or one Hazard. Users can select from the dropdown lists, and search for a Commodity or a Hazard of interest by typing some letters in the search box. For mobile devices, the full graph may not show, and the sub-graph view is the option to use. The x-axis of the graph shows risk score for the commodity-hazard pair.

Commodity

Hazard

 

Click on a bar to obtain a graph showing the corresponding commodity-hazard pairs.