Food Protection Trends - July/August 2017 - 282

the optimal processing in a tuna canning factory. For an
example of an attribute tally sheet, see Appendix C-4. The
disadvantage of the attribute sampling plan in use with the
precooked fish is the sheer volume of core temperature
measurements that need to be collected quickly before
the fish start cooling. From Table 10, for a confidence level
and percent acceptable of (95%/99%), the sample size
for attribute sampling plan is 299, versus a sample size of
35 for a variable sampling plan using the calculated mean
and standard deviation. If the attribute sample size is 300
pieces and 299 pieces pass, but the 300th one fails, what is
the corrective action? By the time the 300th fish has been
measured the cooling has started, so a processor would have
no proof that the fish attained the proper temperature, and
the entire batch will have to be recooked as a corrective
action, and all 300 pieces would have to be resampled.
CONCLUSIONS
Uniform initial core temperatures (ITs) at the start
of precooking will shorten the precooking times in a
conventional atmospheric precooker (CAP) because
the fish was thawed properly outside of the precooker.
A good thawing setup provides uniformly thawed fish to
the butcher table and the precookers. The importance
of properly thawing the fish is discussed in DeBeer et al.
(8). The initial core temperature will affect the speed of
precooking and will need to be measured and verified for
each precooking cycle.
After precooking, the fish must be cooled quickly. If
the fish continues to heat up after being precooked to a
safe core temperature, the fish will be overcooked, and
usable fish recovery and yield will suffer. For this reason,
the processing facilities should have cooling and chilling
equipment with sufficient capacity to cool the fish as
quickly as possible. If there are delays in the cooling area,
the factory should delay the precooking process until the
needed cooling capability is available.

The core temperature method of testing for the
determination of whether an adequate precooking final
core temperature has been attained is a very effective tool.
The testing, which is non-destructive and relatively quick,
is actually only a refinement of the widespread practice of
collecting core (backbone) temperatures after precooking
and therefore should be relatively simple to implement. This
paper provides the scientific basis to support the historical
practice. The technique gives the factory management and
precooker operators sufficient time to optimally process fish
of any size in a safe and effective manner.
The sampling plans(s) presented in this paper are conservative; they are based on the lower tolerance limit of three
standard deviations below the average core temperatures of a
normally distributed set of temperatures to determine whether a Critical Limit is met. In all cases, lethality is counted only
while the temperatures are increasing or remaining the same,
not during the cooling phase (19). In a CAP, it actually takes
longer to cool the fish than it does to heat it, so more log
lethality accumulates during cooling. With this approach, the
fish is safe for further processing past the original 12 hours
recommended by the Seafood HACCP Guide (15). Precooking is and has always been a good tool for preventing histamine formation in commercial tuna processing by precooking
to the proper minimum backbone temperatures. A uniform
core temperature range with very little variation is not only
safe but also maximizes overall recovery and workability of
the fish, making it economically feasible as well.
ACKNOWLEDGMENTS
We thank Lisa Weddig for critical editing of the manuscript. We acknowledge Minitab, Inc.; at the time this paper
was being developed and written, Mr. Colton was a statistician employed by Minitab, Inc. We acknowledge Gerson Hernando Correa Gonzalez, Rick Heroux, and Dan Brooks for
helpful discussions and suggestions. We thank the reviewers
for suggestions to improve this document.

REFERENCES
1. Anonymous. 1986. Microorganisms in
Foods. 2. Sampling for microbiological
analysis: Principles and specific applications.
2nd ed. ICMSF. Blackwell Scientific
Publications, Oxford.
2. Anonymous. 1995. Federal Highway
Administration. Publication FHWARD-02-095. Available at: http://www.
fhwa.dot.gov/publications/research/
infrastructure/pavements/ pccp/02095/
apph.cfm. Accessed July 4, 2015.
3. Anonymous. 50 CFR 260.61 - Sampling
plans and procedures for determining lot
compliance. Available at: https://www.law.
cornell.edu/cfr/text/50/260.61. Accessed
January 7, 2017.

282

4. Anonymous. Normal probability plots and
the "fat pencil test." Available at: http://
support.minitab.com/en-us/minitab/17/
topic-library/basic-statistics-and-graphs/
introductory-concepts/normality/fat-penciltest/. Accessed July 10, 2015.
5. Colley, J. 2015. Personal communication.
[email: javier.colley@thaiunion.com].
6. Colton, J. 2014. Minitab Blog: What does it
mean when your probability plot has clusters?
Available at: http://blog.minitab.com/blog/
the-statistical-mentor/what-does-it-meanwhen-your-probability-plot-has-clusters.
Accessed August 17, 2015.
7. Colton, J. 2015. Personal communication.
[email: james.colton@rocket.com].

Food Protection Trends July/August

8. DeBeer, J., F. Nolte, and C. W. Lord. 2015.
Precooking tuna: A study of the factors
impacting the time required for precooking.
Food Prot. Trends 35:448-460.
9. DeBeer, J., F. Nolte, C. W. Lord, J. Colley, and
L. Weddig. 2016. Setting HACCP Critical
Limits for the Precook CCP of commercially
processed tuna. Food Prot. Trends. 37:176-188
10. Dittami, S. 2009. Shapiro-Wilk normality
Test. Available at: http://sdittami. altervista.
org/shapirotest/ShapiroTest.html. Accessed
June 6, 2015
11. Duncan, A. J. 1974. Quality control and
industrial statistics, 4th ed. R. D. Irwin, Inc.
Homewood, IL.

http://support.minitab.com/en-us/minitab/17/topic-library/basic-statistics-and-graphs/introductory-concepts/normality/fat-penciltest/ http://support.minitab.com/en-us/minitab/17/topic-library/basic-statistics-and-graphs/introductory-concepts/normality/fat-penciltest/ http://support.minitab.com/en-us/minitab/17/topic-library/basic-statistics-and-graphs/introductory-concepts/normality/fat-penciltest/ http://www.fhwa.dot.gov/publications/research/infrastructure/pavements/pccp/02095/apph.cfm http://www.fhwa.dot.gov/publications/research/infrastructure/pavements/pccp/02095/apph.cfm http://sdittami.altervista.org/shapirotest/ShapiroTest.html http://www.fhwa.dot.gov/publications/research/infrastructure/pavements/pccp/02095/apph.cfm http://sdittami.altervista.org/shapirotest/ShapiroTest.html http://blog.minitab.com/blog/the-statistical-mentor/what-does-it-mean-when-your-probability-plot-has-clusters http://blog.minitab.com/blog/the-statistical-mentor/what-does-it-mean-when-your-probability-plot-has-clusters https://www.law.cornell.edu/cfr/text/50/260.61 https://www.law.cornell.edu/cfr/text/50/260.61

Table of Contents for the Digital Edition of Food Protection Trends - July/August 2017