Optimizing sample plans to improve microbiological safety

in a food processing plant

Hassan M. Masri

Thesis submitted to the Faculty of the Virginia Polytechnic Institute and State University
in partial fulfillment of the requirements for the degree of

Master of Science in Life Sciences

In
Food Science and Technology

Joseph D. Eifert, Committee Chair

Renee R. Boyer
Hengjian Wang

June 7, 2013
Blacksburg, Virginia

Keywords: sampling plan, pathogen, low-moisture food, environment

 Optimizing sample plans to improve microbiological safety in a food processing plant

Hassan M. Masri

ABSTRACT

Salmonella and Cronobacter sakazakii are two leading causes of foodborne

illness associated with low-moisture foods, including infant formula. Both causative
organisms can persist in food manufacturing processing environments and contaminate
finished product if programs are not in place to limit their introduction and control their
spread. An environmental sampling and monitoring program is an important tool that
food manufacturers use to determine the effectiveness of their sanitation practices and
pathogen control efforts. Guidance for initiating an environmental sampling plan and
evaluating the plan is lacking.
The objective of this study was to develop microbiological environmental sampling
plans based on the answers to a series of questions related to product hazards,
processing risks and controls, and knowledge of appropriate microbiological sampling
and testing protocols. Furthermore, these initial sampling plans were related to the
volume of product and size of the processing facility. An interactive spreadsheet tool for
designing sampling monitoring plans for an infant formula process was developed using
Microsoft Excel.
Additionally, the tool can be used to record qualitative and quantitative sample
test results, and to alert the user how the upcoming sampling plan will be changed, if
necessary, based on monthly test summaries. The sampling tool provides a simple
method for selecting an appropriate environmental sampling plan (samples per zone per
month) and provides a rationale and guidance for creating and modifying these
plans. Effective sampling plans and trend analysis of sample test results support the
food processors decisions for implementing controls to enhance food safety.
 DEDICATION

I dedicate my thesis work to my family and many friends. A special feeling of gratitude

to my loving parents, Mohammed Masri and Ameerah Al Mashat, whose words of

encouragement and push for tenacity ring in my ears. I also dedicate this thesis to my

many friends who have supported me throughout the process. I will always appreciate

all they have done, especially Mohammed Al Shuniaber for helping me study and pass

my classes. I dedicate this work and give special thanks to my wife Hawazen Nadrah

and my wonderful daughter Ameerah Masri for being there for me throughout the entire

masters’ program. Both of you have been my best cheerleaders.

iii
 ACKNOWLEDGEMENTS

I would like to express my gratitude to my supervisor Dr. Joseph Eifert for the useful

comments, remarks and engagement through the learning process of this master thesis.

I would like to thank my family- my mother Ameerah Al- Mashat, my wife Hawazen

Nadrah and beloved daughter Ameerah Masri who supported me during my years of

study. As well as my friends Mohammed Al Shuniaber, Matthew Schroeder, Ivan

Volonsevich and others for the motivation that inspired me. I would like to thank the

Saudi Arabia Cultural Mission for the scholarship and financial support granted to me

during my graduate life. And, thanks to Virginia Tech and the Department of Food

Science and Technology for the opportunity to be one of their team and family.

iv
 TABLE OF CONTENTS

INTRODUCTION 1

LITERATURE REVIEW 4
A. Microbial Sampling and Testing in Food Manufacturing 4
B. Microbial Sampling in Milk Powder Processing to Enhance
Food Safety 5
1. Milk powder production 9
2. Infant cereal and formula production 6
3. Foodborne illness associated with dried milk products: 7
a. Salmonella 8
b. Cronobacter sakazakii 9
4. Control of Salmonella and C. sakazakii in low-moisture foods 10

C. Microbiological Environmental Sample Selection 11

1. Zoning in a processing environment: 11
2. Sample type and frequency 12
3. Number of sample sites 13
4. Sample analyses 14
D. Sampling Plans in Food Manufacturing 14
1. Sampling plan rationale 16
2. Sampling plans for monitoring product and process
safety and quality 18
a. Attributes sampling plans: 19
b. Variables sampling plans 19
3. Sampling for routine inspection 20
4. Investigational sampling 22
5. Tightened inspection/ skip lot sampling 23
E. Development and Evaluation of Environmental Sampling Plans 23

v
MATERIALS AND METHODS 25
A. Development of an Interactive Tool to Establish an Environmental
Sampling Plan for an Infant Formula/ Infant Cereal Processor 26
1. Initial sampling plan based on product and process food
safety risk 26
2. Questionnaire to determine relative product/process risk 27
a. Step One: Food safety procedures 27
b. Step Two: Processing hazards 28
c. Step Three: Microbiological Sampling and Testing 28
3. Questionnaire to determine relative production volume/
plant size 29
4. Question response scoring 30
5. Sampling plans based on process risk and production volume 31
B. Development of a Spreadsheet Tool to Record and Summarize
Environmental Sample Test Results for an Infant Formula/ Infant Cereal
Processor 32
1. Recording environmental sample test results 32
2. Sample Test Result Summaries 33
C. Sampling Plan Modification based on Monthly Evaluations of
Test Result 34

RESULTS 37

DISCUSSION 40

SUMMARY 44

REFERENCES 59

APPENDICES
A. Sampling Test Improvement Workbook 63

vi
LIST OF FIGURES

Figure 1. Questions related to product and product risk determination

(food safety procedures, processing, sampling /testing). 45

Figure 2. Questions related to relative production volume. 46

Figure 3. Current sampling plan view describes the sample plan category code,
risk level, current guideline state, production volume level,
data entry and data report view. 47

Figure 4. Sample site view describes site code, zone,

sample description or location. 48

Figure 5. Summary report pivot table for both enterobacteriaceae

“EB” count and Salmonella “S” test result, including zone,
total number of samples and site location 49

Figure 6. Sample data entry window includes (date sample taken, date sample
analyzed, sample zone, sample site code, and enterobacteriaceae
“EB” count and Salmonella “S” test result. 50

Figure 7. Initial sample plan software interface. 51

vii
LIST OF TABLES

Table 1. Summary of question response scores required for each of the

9 possible starting environmental sampling plans 52

Table 2. Summary of total environmental samples required, and maximum

number of unacceptable (failed) samples, for each of the 9 possible
starting environmental sampling plans. 53

Table 3. Data collection report view includes date sample taken, date sample
analyzed, sampling zone, sample site code, enterobacteriaceae “EB”
test result, Salmonella “S” test result, and sample test result validation
(Pass/Fail). 54

Table 4. Number of samples required for each of 9 sampling plans at sampling

state 0 (initial) and state 1 with guideline for adjusting sample plan totals
by zone (“if # of failed Salmonella samples > 1 sample per zone (1, 2 or
both 1.2), then shift to higher risk sampling plan (low --> High or med. -->
high) for 1 months, until you receive a negative result for 2 weeks”). 55

Table 5. Number of samples required for each of 9 sampling plans at sampling

state 0 (initial) and state 2 with guideline for adjusting sample plan totals by
zone (“If # failed samples in a month > than limit, then increase EB samples
by 2X each zone for 1 month until you receive under norm results”). 56

Table 6. Number of samples required for each of 9 sampling plans at sampling

state 0 (initial) and state 3 with guideline for adjusting sample plan totals by
zone (“if # of failed samples > limit for 3 consecutive months, then shift to
higher risk sampling plan (low --> med., or med. --> high) for 1 months till
you receive under norm results”). 57

Table 7. Number of samples required for each of 9 sampling plans at sampling

state 0 (initial) and state 4 with guideline for adjusting sample plan totals
by zone (“if # of failed samples < limit for 3 consecutive months, then you
may reduce total # of sample analyses by 10% (by pooling samples)
if you have a good history of your results”). 58

viii
INTRODUCTION

The annual number of foodborne illnesses in the United States is estimated at 49

million cases, and Salmonella bacteria are responsible for more than 1 million cases of

illness each year (Sallan et al., 2011). The medical care and lost productivity cost for

Salmonella between $6.5 and $34.9 billion annually (IFT, 2004). Salmonella is a

generic name applied to a group of approximately 2,000 biochemically related serotypes

responsible for foodborne illness. Salmonella can cause illness with an infectious dose

as few as 15 cells (IFT, 2004).

A significant food safety risk may occur in contaminated foods produced as

ready-to-eat with no additional Salmonella kill step in the process. This type of product

includes low moisture products that do not support Salmonella growth. However, low

cell counts of Salmonella in foods can cause illness, and the presence of this organism

in low moisture ready-to-eat foods must be prevented. A number of outbreaks of

salmonellosis have been associated with the consumption of ready-to-eat and low-

moisture products, including chocolate, powdered infant formula and, more recently,

peanut butter. Although foodborne illness outbreaks are rare due to Salmonella from

low-moisture products, they often impact large numbers of people (GMA, 2009).

Another pathogen associated with low-moisture foods is Cronobacter sakazakii.

Cronobacter spp. infection has been associated with powdered infant milk formula, and

several voluntary recalls of powdered formula have been issued because of

contamination by this pathogen (Norberg et al., 2011).

Control of these pathogens in low-moisture foods often requires sampling and

testing of the product, process environment or both. Many microbiological tests are

available to detect these pathogens directly or other indicator organisms such as

1
Enterobacteriaceae. Most pathogen tests for S. enterica and C. sakazakii are

qualitative since we are usually expecting an absence of these organisms in foods and

the process environment. Tests for indicator organisms or microbial tests to verify

sanitation are often quantitative.

The food industry has many specific tools and procedures for conducting

microbiological tests on food samples or the food process environment. Unfortunately,

procedures or guidance for selecting appropriate environmental samples is lacking.

Very little guidance is available that specifically describes appropriate sample sizes,

numbers, frequency, type, location, etc. Food industries often design microbiological

sampling plans that can range from excessive to minimal since they may not have a

rationale for creating a sampling other than cost, convenience or caution.

Technology provides us with suitable analytical tools which can assist us in

developing a new environmental monitoring sampling plan module that can be

applicable to several food commodity processes. Designing a monitoring sampling plan

software program with consideration of multiple factors of the operation will ensure the

adequacy of the collected environment samples. One aim of this study is to gather and

analyze the collected data from the process line over a period of time, on a statistical

basis.

Another step is to determine accept and reject criteria of the tested environment

samples. The limit of those criteria must be established and evaluated periodically.

A written environmental sampling monitoring program is imperative. The significance of

the environment samples collected from the production line depends on the zone

classification, type of cleaning and the exposure of the product. Designing an

2
environmental sampling monitoring program tool based on risks associated with

processing in a specific facility and the ability to control those risks can help food

processors carry out this important food safety program. The tool could assess the food

manufacturers’ risk that can pose threat to the food during the production cycle time.

Adapting this type of tool can support the decision of releasing the finish product

or control the activity in the high risk operation area. Furthermore, sampling plans

should be modified at regular intervals, based on analysis of previously collected

samples, to facilitate the detection and control of target microorganisms.

Data collection is significant to the processer to determine the weak point and

what major controls must be consider improving the design tool program. Trends of

collected samples can be analyzed and decisions can be made to restrict the control

level of the production line from high to low. Based on the data collected from the

sample test results, the quantity and frequency of future samples can be changed. In

other words, processors could have a way to justify an increase or decrease in the

number of environmental samples collected based on previous test results.

3
 LITERATURE REVIEW

A. Microbial Sampling and Testing in Food Manufacturing

The food processing industries can establish checks on their food products, food

processes or both to ensure the safety and quality of the food they produce. For

example, lot acceptance sampling and testing programs and environmental sampling

and testing programs can be used to enhance food safety, but these programs differ in

many respects. An environmental sampling program can be used to monitor and verify

whether a processor’s Good Hygienic Practices (GHP) are effective and being correctly

applied to prevent unacceptable contamination. GHP’s can be an effective way to

control certain pathogens and spoilage microorganisms within the process environment

and to prevent contamination of foods and then estimate whether food is acceptable.

A plan to sample the plant environment can be based on the information needed

by the processor, the layout of their process and processing plant, and by their previous

experience with sampling and testing the environment. A variety of sample locations or

sites may be relatively straightforward to select. The number and frequency of samples;

how, when and where they are collected; how they are handled between sampling and

analysis and the sensitivity of the analytical method are all crucial to accomplish the

goals of the environmental sampling program. The cost of the environmental sampling

program is one limiting factor that manufacturers must address (Tompkin, 2004).

Microbiological samples of finished product, in-process product or the plant

environment can be analyzed qualitatively and/or quantitatively for pathogenic

4
microorganisms, spoilage microorganisms, microbial indicators of microorganisms of

concern or microbial indicators of insufficient sanitation.

The objective of the sampling plan should be defined before choosing the

sampling plan. Ensuring the complete absence of the defect cannot be defined by a

sampling plan. When a food safety objective or other limit exists, choosing a sampling

plan will be easy. The stringency of the sampling plan then can be determined to detect

the defect levels. The potential source of the problem and the population to be sampled

must be determined when selecting a sampling plan. Developing an appropriate tool to

collect samples can be meaningful to gain information (ICMSF, 2002).

When data is acquired and frequently reviewed, the environment sampling plan

can be useful. Reviewing data from the recent and past test results taken from the

processing environment to detect weaknesses and trends which might be evident are

helpful. Data from environmental sampling tests result within the acceptance limits can

support the decision that a normal routine level of sampling be continued. The reason

should be determined if an increased risk of contamination is indicated on trend or other

information. The detection of the contamination and its causes is the goal of the

intensified sampling program (ICMSF, 2002).

B. Microbial Sampling in Milk Powder Processing to Enhance Food Safety

1. Milk powder production

Raw milk obtained from cows undergoes a wide variety of technologies and

processing to be included in commodities such as fluid milk, cheese, fermented milks

and milk powder. Using appropriate technologies on raw milk such as spry drying or

5
roller drying can produce many products of milk including whole milk, skimmed milk and

cream (ICMSF, 2011).

These dried milk products can be used either directly after reconstitution or as an

ingredient in many products. Infant formula and infant cereals are produced using the

same technologies and process line usually but differ in regulatory requirements

(ICMSF, 2011).

2. Infant cereal and formula production

Infant cereals are made from one or more grains, usually rice, oat or wheat, and

supplemented with calcium, iron, and vitamins. These cereals are sold as a dry or

flaked product that must be reconstituted with water or milk. Many of these cereals are

developed for babies between 6 to 12 months in age. Examples of such foods include

cereal based infant foods, uncooked breakfast cereals, or products designed to be

baked at home (ERS, 2010).

Infant formula can be manufactured in many forms like ready-to-feed ultra-high

temperature products or concentrated sterilized products. Infant formula can be

manufactured in three process types (ICMSF, 2011):

1- Wet mix process: which all the unprocessed ingredients are separately handled

in a liquid form, which then be heat treated and dried before filling stage is

completed.

2- Dry mix process: which all separately processed ingredients are dry blended

before the filling stage.

6
 3- Combined process: which under the wet mix process both unprocessed raw

material and part of the ingredient are processed to form a base powder.

In manufacturing infant cereal a cereal soup is heated before further processing.

After the heating step is completed the cereal soup is transferred to the roller-dryer.

In this step the cereal soup is evenly distributed in a thin film on the roller-dryer

drum. Powder or small flakes are then obtained from the cereal film to form a base

powder. The obtained base powder from the previous step would be used directly or

mixed with other dry ingredient such as vitamin, fruit or vegetable flaks and powder

(ICMSF, 2011).

3. Foodborne illness associated with dried milk products:

Outbreaks have been linked to dried milk products including infant formula and

cereals in many countries. In some cases addition of hydrating ingredients such as

water or milk can assist the growth of bacterial pathogens. Powdered infant formula is

not a sterile product and can get contaminated with pathogens. Correct preparing and

handling of the product can reduce the risk of illness. Manufacturing commercially

sterile powder infant formula is not feasible with the current processing technology.

During the production of powdered infant formula it can become contaminated with

harmful bacteria, such as Salmonella enterica and Cronobacter sakazakii (WHO, 2007).

Currently these are the primary bacterial pathogens of concerns in this product.

7
 a. Salmonella

Salmonella spp. are Gram-negative, rod–shaped motile bacteria (non-motile

exceptions S. gallinarum and S. pullorum), and is non-spore forming (FDA, 2012).

Salmonella is a generic name applied to a group of approximately 2,000 biochemically

related serotypes responsible for millions of cases of foodborne illness worldwide. The

number of foodborne cases caused by salmonellosis annually is roughly 0.6 to 1.7

million in the United States (Scallan et al., 2011). Salmonellosis is a self-limiting

gastroenteritis which may be misdiagnosed as intestinal influenza by the patient or the

physician, and therefore it is grossly underreported (IFT, 2004). Salmonella are

recognized by two clinical indicators, enteric fever and foodborne illness syndrome (IFT,

2004). The enteric fever (a severe, life-threatening illness) is commonly referred to as

typhoid fever, it is primarily caused by Salmonella Typhi. Foodborne illness syndrome

is usually caused by Salmonella enterica.

The microorganisms responsible, in both cases, enter the body via the oral route.

Typically, Salmonella infection resulting from foodborne illness is commonly

characterized by a self-limiting acute gastroenteritis. The usual but not the only common

vehicle of Salmonella contamination is food or water. Salmonella infectious dose is as

few as 15 (IFT, 2004).

Salmonella can enter the food supply in multiple ways (IFT, 2004): 1) food

animals can harbor Salmonella, making meats, poultry, eggs, and milk often implicated

vehicles for Salmonella; 2) Salmonella, which is introduced into the environment

possibly through manure and litter, may survive and contaminate fruits and vegetables

on the farm; and 3) cross-contamination in the food service environment or the home,

8
often between raw poultry and ready-to-eat products, such as raw vegetables, can also

cause salmonellosis.

Cross contamination is an important concern with this pathogen, since it is one of

the ways for Salmonella to enter the food supply especially with food that does not

require a further kill step before consumption, such as some low-moisture product. A

significant food safety risk may occur when this transfer takes place where the product

is ready-to-eat with no additional Salmonella kill step in the process. This type of

product includes low moisture products that do not support Salmonella growth.

However, low cells counts of Salmonella in foods can cause illness, and the presence of

the organism in low moisture ready-to-eat foods must be prevented. A number of

outbreaks of salmonellosis have been associated with the consumption of ready-to-eat

and low-moisture products, including chocolate, powdered infant formula and more

recently peanut butter. Although Salmonella outbreaks are rare from low-moisture

products, they can impact large numbers of people (GMA, 2009).

b. Cronobacter sakazakii

Cronobacter sakazakii, formerly Enterobacter sakazakii, is a Gram-negative,

motile, rod-shaped, non-sporulating pathogenic bacterium that can cause foodborne

illness, primarily among infants and immunocompromised adults. The organism is able

to survive in low-moisture foods, such as powdered infant formula, for long periods

(FDA, 2012). For children infected with C. sakazakii, 50% are less than 1 week old and

75% are less than 1 month old.

9
 Cronobacter infections were reported in immuno-compromised and elderly adults

(FDA, 2012). Typically, contaminated powdered infant formula products with

Cronobacter are the common source for infections (FDA, 2012). Powdered infant

formula manufacturers worldwide are threatened by Cronobacter spp; therefore, it is

important to understand the factors that support the survival and the growth of this

pathogen. Cronobacter spp are tolerant to high temperature and low water activity

which make it difficult to be controlled. Cronobacter spp. stress tolerance appears to be

dependent on the strain and growth phases (Norberg et al., 2011). Cronobacter

contamination can occur after the heat treatment pasteurization step used for the

powdered milk (FDA, 2012). This indicates poor hygiene practices after the

pasteurization step is the source of Cronobacter spp. contamination (Norberg et al.,

2011).

4. Control of Salmonella and C. sakazakii in low-moisture foods

In low moisture foods, the barrier to grow vegetative pathogens, including

Salmonella spp is water activity ) (GMA, 2009). Water activity ( ), is defined by the

ratio of water vapor pressure of food to the vapor water pressure of pure water at a

specific atmosphere. Moreover, low moisture products are characterized as a low water

activity product which does not support the growth of Salmonella. Products which are

characterized as low moisture products are powdered milk products, chocolate, peanut

butter, infant formula and toasted cereal. The presence of Salmonella in low-moisture

products is a concern because low numbers of Salmonella in foods can survive and

cause illness. This is contrary to a common misconception that low numbers of

10
Salmonella are not a problem in low-moisture foods because these products do not

support Salmonella growth. Although these products do not support the growth of

Salmonella, all have been implicated in outbreaks of salmonellosis. After investigating

the outbreaks it was suggested that cross contamination plays a major role in the

contamination by Salmonella of these products (GMA, 2009).

C. Microbiological Environmental Sample Selection

An environmental sampling plan can be designed to optimize the detection and

control of undesirable microorganisms. The plan should include specific sample

information such as type, quantity, frequency, location, analysis, etc. Additionally,

strategies to design these sampling plans for food processing environments should

consider the following (Tompkin, 2004):

 Control the condition which can lead to creating biofilms. This approach may

require redesign of equipment to use appropriate material.

 Establish an effective environmental sampling program. The need of this type of

program is to detect any pathogenic contamination in a timely manner.

 Short term assessment of the results.

 Longer term review of the data quarterly or annually.

1. Zoning in a processing environment:

An environmental sampling program must include various areas throughout the

production process. Dividing the production process line into zones is the simplest way

to design an effective environmental sampling program. Based on the production facility

11
and process line specification, sampling sites from each zone can be determined. To

improve the control of Salmonella a zoning concept must be established (Marriott and

Gravani, 2006). The zone concept is based on four sampling areas and sample types

which can aid the identification of contamination spots:

Zone 1— includes the direct contact surfaces. These areas include equipment

utensils, and containers with direct contact with food products.

Zone 2— area with indirect contact with food in the facility such as equipment

parts or other surfaces that personnel may come in contact with near Zone 1.

This includes drains, utility pipes, and heating, ventilation or air conditioning

system equipment.

Zone 3— area with less contact to food in the facility unlike Zone 2. That includes

floors, walls, and other items in contact with floors, walls, cleaning equipment.

Zone 4— includes maintenance equipment and areas further away from

production such as hallways and entrances of the facilities. Some food

processors may group Zone 3 and Zone 4 samples together when designing a

three zone sampling plan.

2. Sample type and frequency

Each processing facility must select appropriate sample types and a frequency of

sample collection that provides them sufficient information to maintain or improve the

level of plant hygiene or reduce the presence of pathogenic bacteria. Samples that are

typically collected for microbiological analysis include raw products or ingredients,

equipment surfaces, processing water, walls, floors, drains and air. For each sample,

12
the time of collection, location sampled, sample type, sample quantity, and analysis

required must be specified.

The frequency of sample collection for specific sample types or locations can be

based on several factors including traffic patterns in the plant, production volume,

sanitation procedures and frequencies, previous history of sample analysis data, and

microbiological guidelines or action levels. The frequency of sample collection can vary

for different plant locations or surfaces.

3. Number of sample sites

For site qualification, the International Organization for Standardization (ISO)

standard 14644-1 describes a method to determine the number of sampling sites

(Sutton, 2010). The minimum number of sample sites should be determined by the

following equation (Sutton, 2010):

Where:

Is the minimum number of sampling locations (rounded up to a whole number)

A is the area of the clean room or zone in meters 2.

Alternatively, food processors may choose other methods to determine an appropriate

number of samples to collect for their sampling plan. Their choices may be based on

the number of processing lines in their plant, the number of hours or days per week that

they process products, regulatory guidelines, previous sampling history, and, of course,

cost and convenience.

13
 Protocols for the environmental sampling plan may not be designed on a

statistical basis. Thus, sampling plans are based on experience or knowledge of the site

to detect any failure on GHP. When evidence indicates increased risk of contamination

the number, timing, frequency and sampling sites may be increased (ICMSF, 2002).

4. Sample analyses

Sampling and analytical tests may be conducted for specific pathogens such as

Salmonella or Listeria monocytogenes. Alternatively they can target indicators of

pathogen presence such as tests for Listeria spp. to demonstrate the possibility that

Listeria monocytogenes is present or tests for Enterobacteriaceae that may indicate the

presence of Salmonella. Also, aerobic plate counts and ATP bioluminescence assays

are often used to determine areas that need additional cleaning and sanitation.

D. Sampling Plans in Food Manufacturing

An environmental sampling and monitoring program can include numerous

locations throughout the process line. The collected samples can also vary from spilled

product on the floor or equipment, vacuum cleaners, floors, walls, and other surfaces.

And, the type of collected sample may vary, as well, from solid to liquid samples.

Locations where samples must be collected from differ depend on their proximity to the

food products and process. The area with high exposure of the product will be sampled

more. The frequency of the samples collection may vary to ensure that the hygienic

requirement is met during the production cycle. Conducting this type of monitoring

program should include: sample type, sample location, sampling time and frequency,

14
sample quantity, and procedures for sample collection and analyses. After establishing

an environmental sampling monitoring program, it must be verified and validated from

the processer owner. The monitoring program must be evaluated periodically to ensure

the adequacy of it met the goal of sustain hygienic production environment.

Food manufacturers are required to follow guidelines and adopt control

measures to ensure that the food produced from their facility is safe. Compliance with

food safety regulations and adopting appropriate regulatory guidelines is the first step in

producing a safe food. Procedures and policies implemented on site in each food

manufacturing establishment differ from another. During production it is crucial to

comply with the control measure in place to ensure the safety of the food. Evaluating

those measures periodically and the risk to the finish product is important as well.

Producing food in a sanitary environment is an important factor as the risk of

cross contaminating the food will reduce. Implementing a food sampling monitoring

program is essential. Under those monitoring program any part of the process can be

included. Example: incoming material, online samples, equipment’s, tools, floor, walls

and the finish product.

As a normal practice, most of the food manufacturers, following the governmental

regulations or guidance, design their own monitoring sampling plan which is applicable

to their specific production line. Therefore, individual environmental monitoring sampling

plan is established by each processer with consideration of compliance with

government regulation.

Food manufacturers are required to comply with a regulatory authority in the

means of releasing a safe finish product to consumer. The type of evidence may vary

15
from process control documents, online inspection, a product or environmental sampling

plan, or microbial test results. Moreover, food manufacturers are required to prove the

adequacy of the production line. Adapting several control methods such as hazard

analysis critical control point (HACCP) programs, environmental zoning concept,

environment sampling monitoring program and good manufacturing practices (GMP)

would support the decision of whether they are capable of releasing a safe product.

These types of control measures in place influence the readiness of the food

manufacturing facility to produce safe food. Conducting a study on the posed hazards

from the production line within the food manufacturing facility is necessary. These study

questions must answer: What food am I producing? Where am I producing the food?

What are the hazards of concern to my food? And many other factors must be

considered, all which reflect on the control implemented to ensure the safety of food

produced by the manufacturer.

1. Sampling plan rationale

A sampling plan is an executable plan of action that addresses the sampling and

analytical requirements of a specific situation and adheres to the specific sampling

strategy. The sampling plan must specify the sampling approaches, methods, and

analyses, as well as the number, types, and locations of samples to be collected in a

given physical space. The sampling plan should account for the area under

consideration, the number of samples, and the collection locations needed for statistical

confidence as determined by directed and/or statistical sampling designs.

16
Directed sample collection utilizes an expert in the field to determine the suitability of the

plan to meet the experimental objectives. Statistical sampling utilizes a mathematical

framework to determine if the number and location of sample collection sites meets

specific characterization objectives (NIST, 2012).

A sampling plan can vary widely in its goal. The objectives of a sampling design

for data collected from the environment are (EPA, 2002):

 Support a decision whether the contamination level exceeds a threshold

unacceptable risk.

 Determine whether the characteristics of two populations differ by some

amount.

 Estimate the mean characteristic of population with the same interest.

 Identify locations having high level of contaminations.

 Characterize the extent of contamination at site.

 Monitor trend of the environment condition.

To ensure that resulting data are adequately representative of the target

population, a well-planned sampling design is made. Efficient use of time, money, and

human resources are critical considerations for sampling design process. Minimum

costs of resources should meet the needs of the study of a good sampling design (EPA,

2002). A number of samples and identifications of the particular samples are indicators

of a complete sampling design. Moreover, a complete sampling design will include an

explanation and justification for the number and the positions/timings of the samples

(EPA, 2002):

17
 Detecting pathogens in foods was thought to be strictly outside the scope of

environmental sampling. A food manufacturing facility can potentially become cross

contaminated from environmental pathogens so we will consider briefly this aspect of

pathogen sampling. A food processing facility is assumed to follow the principles of

good hygienic practice (GHP) as outlined by Codex (1997). Leading us to presume that

there are procedures in place to manage the risk of environmental pathogens by: 1)

minimizing cross contamination from raw materials to finished product; 2) having

equipment of suitable sanitary design; 3) appropriately maintaining and sanitizing

equipment; 4) removing waste; and 5) training personnel. Then, to ensure that those

procedures are working an environmental sampling program must be implemented

(Legan & Vandeven, 2003).

With regards to their statistical background and in relation to other risk

management approaches such as HACCP or Food Safety Objectives, microbiological

criteria and sampling plans are not fully understood. Microbiological test performed on

several sample units is a simple way to decide whether to accept or reject a food lot

(Dahms, 2003).

2. Sampling plans for monitoring product and process safety and quality

To ensure food quality and safety, two sampling plans are used: attributes

sampling plans and variables sampling plans. These types of microbiological testing are

used to make decisions concerning the safety or quality of foods (Dahms, 2003).

Attributes plans are used to evaluate qualitative data (presence or absence of analyte)

18
or quantitative data that have been grouped (e.g., <10 cfu/g, 10 to 100 cfu/g, >100

cfu/g). Non-grouped quantitative data can be evaluated with variables plans.

a. Attributes sampling plans: Attribute sampling, also known as proportional

sampling, allows one to measure the probability of discrete possible outcomes. The

presence or absence of a specific contaminant is an example of a discrete outcome that

could be measured with attribute sampling. There are two methods of attribute

sampling plans qualitative two-class attribution and quantitative three-class attribution

test (Dahms, 2003). For the qualitative test of presence or absence of the pathogens, a

two-class plan is defined by two numbers for decision making process. First, denote n,

which determines the samples units independently and randomly picked from a

production lot. Secondly, denote c, which is the maximum acceptable number of

samples yielding unsatisfactory result. In the case of quantitative grouped data applied

to a two-class plan, there is one microbiological limit denoted m, which separates

acceptable from defective quality.

Moreover, an Operating Characteristic (OC) curve is used to visualize the

performance of the sampling plan. The OC curve has two scales: a horizontal scale

showing the percentage of positive units in the lot being tested, and a vertical scale

giving the probability of acceptance.

b. Variables sampling plans: Variable sampling allows one to measure quantities.

For examples, variable sampling could be used to measure how many bacteria are

present at a given time, or temperatures of a processing area at defined time intervals.

When decisions are not based on qualitative analytical tests, quantitative analytical tests

19
are applied, working with data grouped according to a single microbiological logical limit

m. Three-class plans are used where the quality of food is divided into three attribute

classes. In two-class plans samples results based on quantitative analytical results

above a concentration m that in a three-class plan separate good quality from slightly

acceptable quality but a certain number donate as c, can be accepted. However,

sample test results exceeding the second microbiological limit M are rejected if any test

result of n sample unit is above M. Therefore, three-class plans OC curves result in

three dimensional graphs which are difficult to compare with two dimensional OC

curves.

3. Sampling for routine inspection:

To ensure that the operation remains under control by detecting any increase in

the risk of cross-contamination, routine environmental samples must be collected. A

normal risk of cross-contamination, when the operation is under control, first must be

determined by:

 Selection of one pathogen or indicator organism for the ongoing monitoring.

 Take samples at various stages of the process flow, to determine the

microbiological condition facility.

 Residues of food samples can be collect as monitoring samples.

Environmental sampling protocols are designed to pay most attention to those areas

known to pose the highest risk of product contamination and they are not statistically-

based. However, some statistical concepts are applicable. Assigning number of zones

within the processing plant is common practice, where different zones have different

20
levels of risk of contaminating the product. Zoning concept is defined in four zones as

mentioned previously. Based on experience of the sites the exact balance of samples

from the four zones is most likely to indicate that the operation is ‘out of control’ in terms

of good hygienic practice (Legan & Vandeven, 2003).

The sampling protocol and improved knowledge of the operation will over time

shift the selection sites. Knowing where to sample as well as when to sample are

important. The most critical time may be immediately after startup in some operations.

Special sampling can be implemented in response to known risk factors such as

construction the frequency of sampling may be increased if evidence indicates an

increased risk (Legan & Vandeven, 2003).

The Grocery Manufacturers Association (GMA, 2009) included an example of an

environmental monitoring program for production of low-moisture foods in their

guidance document on Salmonella control in these products. The GMA document

details a monitoring program with 4 sampling zones where samples are tested for

Salmonella primarily in zones 2, 3, and 4. The number of samples in each zone

decreases as you move from zone 2 to zone 3 and zone 4. Tests for indicator

organisms such as Aerobic Plate Count or Enterobacteriaceae are recommended for

zone 1 (product contact surfaces in primary control areas). They note that

Enterobacteriaceae is a useful indicator of process hygiene and may be monitored in

parallel as a hygiene indicator for verification of general sanitation effectiveness.

However, it cannot be a substitute for the direct monitoring of Salmonella because,

while high levels of Enterobacteriaceae suggest an increased risk for the presence of

21
Salmonella, low levels of Enterobacteriaceae do not guarantee the absence of the

pathogen (EFSA, 2007; Cordier, 2008).

Another example of sampling plan guidance was recently published by the US

Dept. of Agriculture, Food Safety and Inspection Service for use by their Enforcement,

Investigations and Analysis Officers (EIAOs) to follow when collecting product samples

during Intensified verification testing protocols for sampling of product, food contact

surfaces, and environmental surfaces for Listeria monocytogenes (lm) or Salmonella.

Effective in 2013, samples collected for the L. monocytogenes program will consist of

sampling units of 10 food contact surface, 5 environmental (non-food contact surface),

and 5 ready-to-eat (RTE) product samples. Additionally, when sampling for Salmonella,

EIAOs are to collect 5 food contact surface samples, 8 environmental samples, and 5

RTE product samples. The instructions provide some guidance on when to collect the

samples during a day, and suggest locations to sample, but they do not state the

quantity and frequency of sample collection at each sample site (USDA, 2013).

4. Investigational sampling:

Information suggesting that a problem already exists can lead to an investigation

response. To correct the problem the source must be known first. To investigate a

problem efficiently, a random sample must be obtained and knowledge must be applied

to: 1) microbiology; 2) process operations, 3) equipment design; 4) information gained

from visual inspection of the operation; and 5) sampling sites most likely to harbor the

organism(s) of concern. It is important not to jump to conclusions, even at the same

time as applying pre-existing knowledge, mainly when resources are limited and time is

22
constrained. Investigational sampling is very likely to be repeated to maintain a degree

of flexibility. In more detail the influences between the last ‘point of absence’ and the

first ‘point of detection’ can be examine. The investigation work over time can

sometimes be very lengthy towards identifying the source of contamination (Legan &

Vandeven, 2003).

5. Tightened inspection/ skip lot sampling

With an ongoing relationship with a supplier, or other ongoing sampling situation,

a level of confidence in the performance of that supplier over time can be developed.

Skipping sampling of some lots altogether and using the freed sampling resources

where they can be more beneficial that may lead us to relax the rate of sampling.

However, if a defective were detected the initial sampling rate will be reverted or even

more stringent, until the confidence of the suppler is developed again to an acceptable

level (Legan & Vandeven, 2003).

E. Development and Evaluation of Environmental Sampling Plans

The response to information that indicates a problem is a major limitation

throughout the food industry. This can be due to failure to organize the results in a

manner that facilitates review (Tompkin, 2004). For example:

1. Failure to recognize an evolving problem or its significance,

2. Simply filing results without review or,

3. Finally, an individual or group is not assigned responsibility or held

accountable for identifying and responding to a problem.

23
Time needed and the difficulty to detect the source of contamination is a limitation in

industry’s response. All samples should be analyzed individually rather than pooled,

and, samples should be collected more frequently and additional sites should be

included when striving to detect the source. Dates and times when positive results have

occurred should be easily traced to a location on the map showing the layout of

equipment’s in the room and sites. This can be demonstrated in the following order

(Tompkin, 2004):

1. Do the results reveal patterns with certain equipment showing more positives?

2. Where in the flow of food through the process do the first positives occur?

In general, the microorganisms flow downstream from the source of contamination with

the food. Identifying the source and the pathways of contamination can be determined

by fingerprinting isolates (Tompkin, 2004).

Frequently, food processors react to unacceptable test results through additional

sampling or sanitation procedures. However, continual, long-term evaluation of

environmental sampling plans and test results should be performed to determine if there

are trends in microbial detection. The evaluation of the sampling plan and the test data

over extended times may lead to changes in the number of samples collected, test

sample frequency, location and analysis performed, or in the plant's corrective actions.

A thorough evaluation of the data can lead to increased sampling for potential problem

areas and decreased sampling frequencies for areas that have generally negative test

results.

24
 MATERIALS AND METHODS

An interactive spreadsheet tool for designing sampling monitoring plans, that

suggest one of 9 basic sampling plans based on product/ process risk (3 levels) and

production volume (3 levels), was developed using Microsoft Excel 2010 (Microsoft

Corp., Redmond, WA). This study used an approach requiring that an initial

microbiological environmental sampling plan be based on the answers to a series of

questions related to product hazards, processing risks and controls, and knowledge of

appropriate microbiological sampling and testing protocols. Furthermore, the initial

sampling plan can be related to the volume of product and size of the processing

facility.

These sampling plans will provide the user with the total number of samples to

collect each month for a pathogen (qualitative) and an indicator organism (quantitative).

Furthermore, these sample quantities are distributed between three environmental

sampling zones. Additional sampling guidance is provided.

The outputs of the initial sampling plan design spreadsheets were linked to

another spreadsheet that can be used to record and summarize environmental

sampling test results. Together these were customized to create an interactive tool that

will suggest modifications to the current sampling plan based on the cumulative test

results obtained over one month and three months. The development of these

sampling plan design tools are described below.

25
A. Development of an Interactive Tool to Establish an Environmental Sampling

Plan for an Infant Formula/ Infant Cereal Processor

1. Initial sampling plan based on product and process food safety risk

To setup a microbiological environmental sampling plan or to validate an existing

sampling plan several factors must be considered that require some knowledge of how

the product is produced, the microbial hazards that could be found in the raw

ingredients or introduced in the process, the food safety controls used in the processing

plant, the number and volume of products produced over time, and the appropriate

qualitative or quantitative microbiological tests that are needed. Additionally, the

sampling guidance does not assume that the user has conducted microbiological

sampling and testing in the past or has knowledge of best practices for sampling or has

knowledge of the previous level of environmental sampling for a processing plant.

A typical food processing plant has numerous locations or zones where an

environmental sample could be collected. Each sampling zone may have a specific

hygiene requirement and unacceptable test results from one or more samples within a

zone may require a different response from the processor. Frequently, food processors

categorize environmental samples as originating from one of these three zones:

• Zone 1 – includes in-process product and surfaces that can contact the product;

• Zone 2 – encompasses the areas directly adjacent to Zone 1 and includes all

non-food contact or indirect contact areas in the processing plant;

• Zone 3 – sampled areas that are usually environmental (floors, ceilings walls)

and typically not in the food processing rooms.

26
 In the initial setup of the sampling plan at least one sample must be collected from

each assigned zone location spread throughout the processing line.

2. Questionnaire to determine relative product/process risk:

A questionnaire was designed to elicit responses on food safety procedures,

processing conditions, and sampling and testing protocols for a processor of powdered

infant formula or infant cereal. These are low-moisture foods that may be considered

unsafe due to possible contamination by the microbial pathogens Salmonella spp. and

Cronobacter sakazakii.

An initial list of 80 questions was sorted and ranked by their importance for

affecting the relative food safety of the product and process. The final list of 22

questions are discussed below and presented in Figure 1. The process of developing

the sampling monitoring plan was based on three sets of questions:

Step One: Food safety procedures

The questions listed below ask if the user practices or is aware of some basic

food safety related operations including Good Manufacturing Practices,

monitoring of ingredients and employee training.

 Do you follow Good Manufacturing Practices (GMPs)?

 Do you have an environmental monitoring sampling plan (including sample

locations, frequencies, types, sizes)?

 Do you know the source of your raw materials and ingredients?

 Do you maintain a certificate of analysis for your materials?

27
 Do you require specifications for all ingredients?

 Did / Will your staff receive food safety and GMP training this year?

Step Two: Processing hazards

The questions listed below ask about procedures that may increase or decrease

the relative microbial safety of the product. Also, the questions help ascertain if

the user has knowledge of the potential hazards and food safety controls in their

process and if they can provide the required measures to control the processing

environment. As a control measure, the staff food safety awareness is vital to

handle this type of food. The sources or frequency of contamination can be

lowered or eliminated by following good manufacturing and hygienic practices.

 What is the physical condition of your process facility?

 Does your processing room have positive pressure?

 Do you use a HEPA filter in you air unit?

 What type of barrier between zones do you have in the process line?

 Are you knowledgeable of the type of hazards to your product?

 What is the risk of pathogen contamination to your product?

 Does the risk of your product to the consumer change after production?

 Is any source of pathogen contamination eliminated from the process line?

Step Three: Microbiological Sampling and Testing

The questions listed below are used to verify the user’s awareness of the zoning

concept and current sampling plan, if any, for the environment and product.

28
 Microbial sampling and testing is commonly used to validate and verify effective

hazard control measures in place.

 Do you know the (pathogenic) microorganism of concern?

 Do you test product samples for pathogens?

 Do you test for indicator or non-pathogenic organisms?

 Is your staff trained to collect environment samples?

 Do you know how many environmental samples you take per lot or per day?

 Do you have a list of locations to collect environment samples from?

 How many environmental sampling zones do you have?

 Do you know the microbiological test specifications of your analyses?

3. Questionnaire to determine relative production volume/ plant size:

Five additional questions were added to the questionnaire described above, to

elicit responses for categorizing the production volume of a powdered infant formula or

infant cereal processor. These questions asked about the daily operating hours of the

plant, number of employees, annual production volume and annual sales volume.

Generally, food processors will collect more environmental samples in larger plants

since there are more locations to sample and more places that pose a risk for product

cross-contamination. These additional questions are listed below and in Figure 2.

 How many work shifts per day for production?

 How many production lines do you have in your facility?

 How many employees you have?

 What is your production volume per year (units or pounds)?

29
  What is your sales volume per year?

4. Question response scoring

For each question, four choices were provided. In many cases, the four choices

include the responses of 1) “yes”; 2) “no”; 3) “usually”, “sometimes” or “not sure”; and 4)

“don’t know” or “unknown”. For each question, the responses were assigned a value

ranging from 1 to 5. A value of 5 reflected that the response could be related to

improved food safety. A lower value implies that these responses are detrimental to

food safety or do not enhance food safety. The response of “don’t know” or “unknown”

was always scored as “1”. For the questions related to production volume and size of

operations, the four choices for each question carried a number or a range of numbers.

A value of 4 or 5 reflected that the selected response represented a larger size or

volume plant, and an answer that reflected a relatively small volume plant resulted in a

score of 1 or 2 (Figure 1 and 2).

Each of the concepts represented by the 27 questions was ranked as having a

relatively high to low food safety impact or significance to the sampling plan. Multiplier

factors were used for each ranked question depending on its importance to form the

sampling plan. The multipliers were 10 (low), 20 (medium) and 30 (high). The higher

the multiplier meant the question was relatively more important for food safety impact.

To determine the relative risk level of the product and process, the scores for

each of the first 22 question responses can be multiplied by their food safety impact

score (10, 20, or 30) and summed. The total score possible ranges from 420 to 2100.

30
Twenty combinations of question responses were used to determine appropriate total

score ranges that would correspond to low, medium and high process risk (Table 1).

To determine the relative production volume of the processing plant, the scores

for each of the final 5 question responses can be multiplied by their food safety impact

score (20 or 30) and summed. The total score possible ranges from 220 to 600. Twenty

combinations of question responses were used to determine the appropriate total score

ranges that would correspond to low, medium and high production volume (Table 1).

5. Sampling plans based on process risk and production volume

Nine basic sampling plans were created based on product/ process risk levels

(low, medium or high) and production volume levels (low, medium or high). Each of

these sampling plans will provide the user with the total number of samples to collect

each month for a qualitative pathogen test (Salmonella, for example) and a quantitative

indicator organism (Enterobacteriaceae, for example). For the infant formula/ infant

cereal product and process used in this example, each environmental sample is tested

for both a pathogen (Salmonella) and an indicator organism (Enterobacteriaceae (EB)

count). Furthermore, these sample quantities are distributed between three

environmental sampling zones (Table 2). The sampling plans take into consideration

two main factors- sampling frequency and zone location. The sampling frequency

represents how often samples must be taken from each zone. Each zone should be

tested for any pathogen contamination, but the number of samples from each zone may

vary.

31
 To illustrate further, the number of sample collected for the process line is

determine by the production volume, employee number and number of processing line.

The bigger the production volume the more samples would be collected. Another factor

that might increase the sampling number includes how many shifts would the operation

run per day. Due to the increase of the operational shifts per day, more samples must

be attained to validate the collected data over a specific period of time (Figure 2).

B. Development of a Spreadsheet Tool to Record and Summarize Environmental

Sample Test Results for an Infant Formula/ Infant Cereal Processor

1. Recording environmental sample test results

A spreadsheet template for recording and evaluating environmental sampling

data was developed based on the format described by Eifert and Arritt (2002). This

template provides a format for recording environmental sample identification information

including time of collection (day, date, shift), plant area location, sample location,

analytical test (qualitative or quantitative) and test result. For the current project, a data

set of environmental sample test results was constructed for test purposes and

analyzed using the "PivotTable" feature in Microsoft® Excel 2010 (Microsoft Corp.,

Redmond, WA). A PivotTable is interactive tables that quickly summarizes, or cross-

tabulates, large amounts of data, including user-selected subsets of the data. The user

can rotate the rows and columns to see different summaries of the source data, filter the

data by displaying different pages, display the details for areas of interest, and

ultimately chart the PivotTable data.

The data set constructed for this example contained 360 line entries with test

results for quantitative (Enterobacteriaceae) tests and qualitative (Salmonella) tests. A

32
portion of the data table, shown in Table 3, includes the following "Fields" and

accompanying data ranges.

• Date sample taken:

• Date sample analysis on:

• Environmental Zone (1, 2, or 3)

• Sample site (within a zone) code or abbreviation

• EB count test result (CFU/mL)

• Salmonella test result: “1” = positive, “0” = negative, blank = no test

• EB count test decision: sample fails when counts exceed 10/mL from zone 1,

100/mL from zone 2, and 500/mL from zone 3

While our example data set required data input directly into the cells of the

spreadsheet, the environmental sampling tool was designed so that users will enter

sample information into a separate dialog box for each sample. That information will

then be transferred to the cumulative record of sample information. Examples: Date of

analysis used to validate the compliance with any policy implemented in the

manufacturer lab. The date of analysis can also be used to verify if the sample had

sufficient time to be analyzed. The “Date sample taken” inputs can verify if the sampling

plan frequency is sufficient to be conducted in a timely manner.

2. Sample Test Result Summaries

Summaries of the sample description and test result data can be reported with

Pivot Tables or Pivot Charts to show trends. Numerous combinations of data variables

and table and chart formats are possible. Moreover, the collected data can be

33
represented in trends to illustrate the need to increase or decrease the control level on

the processing environment. Pivot Tables of the test result summaries can be used to

present test results of a specific zone, period of time or pathogens of concern and

indicator microbe. Evaluating the test result summaries can enhance the decisions to be

made to improve the controls implemented on site. Correlation between indicator and

pathogen analytical test results, over time, obtained from processing environments can

be investigated for trends, but comparison for individual samples is usually not

recommended. Comparing test results between or within zones can illustrate the

source of any contamination. The sampling tool will be designed to display a minimum

number of data summaries to reduce confusion. To create additional data summaries,

users will be able to export the sample information data for further manipulations.

C. Sampling Plan Modification based on Monthly Evaluations of Test Results

Evaluating the data collection over a given time period can support the decision

to modify the initial sampling plans. Test result evaluation can determine the need to

increase or decrease the sampling number obtained from the process line environment.

The sampling tool will be designed to evaluate test results over a specific time period.

Whit this function the program will then inform the user that the sampling plan may need

to be changed in the next time period (the following month, for example). The guidelines

included in the sampling tool will specify the general changes in the sampling required

that will cause an increase or decrease in the number of samples collected in each zone

in the following month. For both pathogenic and indicator test results over time, four

specific guidelines will be used to determine the need to change the initial sampling

34
plan. If one or more of the specified guidelines were compromised the sampling plan will

change as per the described guidelines.

In the guidelines, two time periods were specified, one- and three- months of

data collection to direct the change in the initial sampling plan. After one month of data

collection if the initial sampling plan test result complied with the given guideline the

same sampling plan shall continue to the next month cycle. Unless one or more of the

guidelines are compromised the initial sampling plan shall shift to a higher strict

sampling plan, for a specified time described in the guideline. If the test results obtained

from the environmental samples are below the limit for unacceptable results, then the

sampling will return to the lower sampling plan. If the test result obtained from the

environment samples were positive the higher sampling plan shall continue until the

process environment comes under control. If the data collected from the processing

environment test result were negative for a consecutive three month, the sampling plan

shall decrease to 10% of the total initial sampling plan size. The new reduced sampling

plan will be evaluated for one month. If the new decreased sampling plan leads to

results that are below the fail limit over a consecutive three month cycle, then the

sample plan shall decrease for an additional10%. An overall reduction of 20% from the

initial sampling plan is the maximum sampling plan decrease.

The number of samples required for each of the 9 sampling plans at sampling

state 0 (initial plan) and sampling plan state 1, 2, 3, and 4 are shown in Tables 4, 5, 6,

and 7, respectively. The sampling plan tool will alert the user when one of these four

situations occurs that will result in a change in the monthly sample plan by zone. For

sampling state 1 and 2, the number of samples in the next plan will be higher if the

35
number of “fail” samples in a month exceeds the limit for Salmonella or

Enterobacteriaceae, respectively. For sample state 3, exceeding the fail limit 3 months

in a row will result in this new sampling plan. If the number of samples that fail in a

month is below the limit, for 3 consecutive months, then the number of samples will be

reduced by 10% the following month for sample state 4.

36
 RESULTS

The “Sampling Test Improvement Workbook” developed for this thesis research

project is available as an electronic file (*.xlsm) in Microsoft Excel. The inputs and

features of this software program are further described below. Each input can be used

to improve the sampling plan tool. The inputs integrate with each other, to provide data

summaries and sampling plan recommendations. The data can be presented as needed

to support sampling decisions made to improve or maintain the hygienic condition in the

process line environment.

Selected screenshots of the environmental sampling tool software program are

included in Figures 3 – 7. The parameters of the current sample plan, with the number

of samples to collect at each site, are shown in Figure 3. An example of a list of sample

sites and zones is shown in Figure 4. Sample location descriptions can be abbreviated

with a two-character code to facilitate data entry. An example of two Pivot Table

summaries for Enterobacteriaceae “EB” or Salmonella “S” test results by zone and site

is in Figure 5. Figure 6 displays the sample date, site and test result window for the

software user. Finally, in Figure 7, the initial sample plan software interface is

displayed.

After defining the inputs requested by the sampling plan program, an initial

sampling plan will be selected which can be modified based on sample test results.

Unsatisfactory test results may change the sampling plan state and sample number.

The increase and decrease of the sampling plan shall change upon the test result

entered to the software program. The minimum number of samples is one per

processing location.
37
 For the environmental sampling monitoring tool a set of questions was developed

to focus on the product and process food safety risks. The level of hygienic control in

the process line can be determined from the risk of the product, product exposure and

strictness of the sampling plan. Another set of questions was used to estimate the

relative production volume for the starting point of the sampling plan. The bigger the

volume the more control must be implemented on the level of hygienic control.

For each question and set of responses, point scores were assigned to guide the

initial sampling plan as per the process needs. A higher score determines the food

manufacturers’ awareness of the process requirements and controls implemented on

site. The higher the score the less strict and fewer samples the sampling plan will be,

with consideration of the sample location which must be sampled at least once per

week. The formation of nine sampling plans is included in the program software. Upon

the answered questions and scores attained, a specific sampling plan would be

assigned to each process facility. A recommendation is given to start from the higher

level of the hygienic control to a lowest level of control. Samples to collect each month

from any group would be categorized for each zone 1, 2, 3 in order of 50, 30, and 20%

of the total.

For each sample collected from an individual location both pathogen (qualitative)

and indicator (quantitative) tests are recommended. The collected data result from both

tests shall be analyzed for further assessment of the food manufacturer process line

hygienic condition and decision shall be made to increase or decrease the sampling

size as per the advised sampling plan.

38
 A specific guideline for each state of sample collection is described in the initial

sampling monitoring plan. Guidelines illustrate the mechanism of the sample collection

size and location. Moreover, unsatisfactory test results from any collected sample shall

trigger a change in the sampling number, location and frequency. Guidelines provide

the limits and specification to change the sampling plan. Guidelines are subject to

change as needed by the food manufacturer. The rule of increase and decrease the

sampling plan size is described in the guidelines. Limits of accept or reject any test

result is described as well in the guidelines. The start and end of the sampling plan

cycle is written in the guidelines as well.

Each sampling plan has three hygienic levels of control which have been

designated as a “State”. Any change to the sampling plan shall be between the same

set of hygienic level of control. The higher the state the more strict the sampling

requirement would be (higher number of samples to collect). A guideline is provided to

describe the change between sampling states.

39
 DISCUSSION

Development of an appropriate environmental sampling plan is essential to

determine the readiness of the production facility to produce a potentially hazardous

food product. This environmental sampling tool provides a method of creating a

sampling plan with only basic knowledge of the product, process, hazards, and volume

of the production. This basic knowledge can be driven from experience or process

needs. The data outcomes from the initial sampling plan can be analyzed and decision

can be made from it. The initial sampling plan must comply with the process needs as

well as the regulatory authority.

The initial question set provides a way to initiate the process of determining the

appropriate number of samples to collect for initiating a new sampling plan, as well as to

clarify what are the main food safety concerns in the process line. These questions and

response choices were designed so that someone with incomplete knowledge of the

manufacturing operation could design a sampling plan, and so that a facility could

consider using this tool even if they already were collecting microbiological

environmental samples.

Questions can be added that can elicit specific information about the product and

process or can be modified to meet the manufacturer needs and concerns. The

questions and question responses can be re-weighted as needed by the manufacturer

or alternate responses could be created. Moreover, the questions can be presented in

an alternate order to meet the manufacturer need starting from the end to beginning of

the process line. The question set can also guide the manufacturer for how many

40
samples they have to collect from the process line after specifying the number of

locations in their process. Furthermore, these questions could be used to determine if

procedures have been established that are specific to a Hazard Analysis Critical Control

Program (HACCP). Having such control procedures in place is a significant way to

reduce or eliminate the food safety hazard that threatens your product. Adapting

HACCP is important to ensure that the product is produced under well-established

measures. Taking into consideration the pathogen of concern and the detection method,

would assure the awareness.

General questions on the production volume, sales volume, number of

employees, and the length of daily operations, were used to determine three levels of

total sampling plan size needed to be collected from the process line environment.

Sampling frequency can be driven from the question set as for how many shifts the

operation is running. As stated previously, all the questions used can be customized as

per the manufacturer’s needs and compliance with the regulatory authority. Having a

schedule sampling plan is very important to avoid any gap in the sampling plan. A

proactive sampling plan is important to ensure that the food is delivered under a safe

environment.

The environmental sampling tool utilizes the zoning concept to define and

organize sampling sites in a process facility. Complying with the zoning concept is

significant, to ensure that each individual zone has been tested for any sign of

contamination. Testing each zone and evaluating the result will assess the hygienic

condition of the process line. The proportion of environmental samples to collect in

each of the three environmental zones, suggested by the program, emphasizes a higher

41
number of samples in zone 1, and the fewest in zone 3. Description of each sample

location is written in the sampling plan to refine the sampling location. The sample

proportion should be verified by a food manufacturer. The estimated minimum number

of samples taken from the process line must be at least one sample from each zone

location for a specific time period. Users of this program may want to alter the starting

percentages, but those changes would require access to macros and Visual Basic code

used to develop the sampling tool.

The number of environmental samples which must be collected from the

processing area is very crucial. Not only must the processor ensure that the food

product is produced under hygienic condition, they may also need to comply with a

regulatory authority which could require that a minimum number of environmental

samples are collected and tested. Food manufacturers may want to pool or combine

samples within a zone or from a site prior to conducting a microbial analysis. This

procedure can reduce the number of samples taken from the process line, but it must

be verified and validated by the food manufacturer. Traceability of each sampled

location is a must, to ensure the accuracy of the environment sampling monitoring plan.

For data entry, a guideline is provided to support the decision of any changes to

the sampling plan. Guidance is given to determine when the sampling plan must be

changed and when the new sampling plan shall start, as well as if any unsatisfactory

result is entered what corrective action should be made. In the sampling plan a specific

sample number derivative from the number of sampling location.

Test result data can be easily summarized to determine if actions are needed, if

the plan needs modification, or to evaluate trends. Sample test data records and test

42
result summaries could be modified to include: sample results on a weekly basis, mean

quantitative test results, sample size or area specifications, and randomized selection of

sample sites within a zone, for examples.

Trends observed in the sample test results can indicate the level of hygiene of

the process line or facility. This information can illustrate the need to increase or

decrease the control on the processing environment by adjusting the number of

samples collected from the process line. Additional measures can be taken to improve

and verify the hygienic condition of the process line environment on an immediate or

long-term basis as needed. The availability to the data collected over a year or more

can support the decision made to improve or change the sampling plan or other food

safety programs.

The microbiological environmental sampling plan tool can be modified for other

food products besides low-moisture foods or milk-based products. Adaptation to other

food products and processes would require some knowledge of the food process, the

finished product, the intended user of the product, the microorganisms of concern, the

recommended tests for these microorganisms, the likelihood of product or process

contamination, and regulatory testing requirements. While this approach for creating

and modifying an environmental sampling plan can improve the hygienic condition of

the process line environment and enhance food safety, a food processor may still need

to conduct appropriate microbial testing on finished products.

43
 SUMMARY

Complying with the GHP in process lines is essential to prove that the food is

produce under a sanitary condition. The hygienic condition of the processing

environment is conceder a critical factor to food safety. Collecting samples from the

processing environment indicate the hygienic condition of the process line. Developing

a suitable sampling monitoring plan is crucial to control the source of contamination.

The developed sampling plan can support the decision made to increase or

decrease the environment sampling number. Evaluating the data collected for the test

result obtained from pathogenic and indicator analysis can illustrate the need to change

or modify the initial sampling plan. Features included in the software program to assess

the need to change that sampling plan accordingly. Pivot tables and trend analysis

represent the data collected of a specific need. Improving and controlling the hygienic

condition of a process should be a concern of a food manufacturer. Additionally, using

such a tool can illustrate the need to take major corrective actions, if needed.

44
 Rate Rate Rate Rate
Category Q. # Questions Response Response Response Response
score score score score

score

multiplier
multiplier

maximum
Maximum

Importance
Importance
Total score

Product and Product Risk

Food Safety
1 Do you follow Good Manufacturing Practices (GMPs)? H 30 Yes 5 No 1 Usually 3 Don't Know 1 5 30 150
Procedures
Food Safety Do you have an environmental monitoring sampling plan
2 M 20 Yes 5 No 1 Partially 3 Don't Know 1 5 20 100
Procedures (including sample locations, frequencies, types, sizes)?
Food Safety Do you know the source of your raw materials and
3 L 10 Yes 5 No 2 Sometimes 2 Don't Know 1 5 10 50
Procedures ingredients?
Food Safety Do you maintain a certificate of analysis for your
4 L 10 Yes 5 No 1 Sometimes 2 Don't Know 1 5 10 50
Procedures materials?
Food Safety
5 Do you require specifications for all ingredients? M 20 Yes 5 No 1 Usually 2 sometimes 1 5 20 100
Procedures
Food Safety Did / Will your staff receive food safety and GMP training some only when
6 M 20 Yes 5 No 1 3 3 5 20 100
Procedures this year? employees hired

Process 1 What is the physical condition of your process facility? L 10 Very Good 5 Good 3 Fair 1 unknown 1 5 10 50

Process 2 Does your processing room have positive pressure? M 20 Yes 5 No 2 Usually 3 unknown 1 5 20 100

Process 3 * Do you use a HEPA filter in you air unit? L 10 Yes 5 No 1 sometimes 2 unknown 1 5 10 50
procedures, processing, sampling /testing).

What type of barrier between zones do you have in the

Process 4 L 10 Physical 4 Virtual 3 both 5 Don't Know 1 5 10 50
process line?
Are you knowledgeable of the type of hazards to your
Process 5 H 30 Yes 5 No 1 probably 3 Don't Know 1 5 30 150
product?
What is the risk of pathogen contamination to your
Process 6 H 30 High 1 Medium 3 Low 5 not sure 1 5 30 150
product?
Does the risk of your product to the consumer change
Process 7 L 10 increased 1 no change 3 reduced 5 not sure 1 5 10 50
after production?
Is any source of pathogen contamination eliminated from
Process 8 H 30 Yes 5 No 1 Usually 3 not sure 1 5 30 150
the process line?
Sampling/ * Do you know the (pathogenic) microorganism of
1 M 20 Yes 5 No 2 not sure 3 Don't Know 1 5 20 100
Testing concern?
Sampling/
2 Do you test product samples for pathogens? H 30 Yes 5 No 1 sometimes 1 Don't Know 1 5 30 150
Testing
Sampling/
3 Do you test for indicator or non-pathogenic organisms? M 20 Yes 5 No 1 Sometimes 3 Don't Know 1 5 20 100
Testing
Sampling/
4 Is your staff trained to collect environment samples? L 10 Yes 5 No 2 Somewhat 3 Don't Know 1 5 10 50
Testing
Sampling/ Do you know how many environmental samples do you
5 M 20 Yes 5 No 2 not sure 2 Don't Know 1 5 20 100
Testing take per lot or per day?
Sampling/ Do you have a list of locations to collect environment
6 H 30 Yes 5 No 1 not sure 2 Don't Know 1 5 30 150
Testing samples from?
Sampling/
7 How many environmental sampling zones do you have? M 20 one 1 two 3 three or four 5 Don't Know 1 5 20 100
Testing
Sampling/ Do you know the microbiological test specifications of
8 L 10 Yes 5 No 1 Somewhat 3 Don't Know 1 5 10 50
Testing your analyses?
Figure 1. Questions related to product and product risk determination (food safety

Total: 110 420 2100

45
 Category Q. # Questions Response
Rate
Response
Rate
Response
Rate
Response
Rate .
score score score score
score

multiplier
multiplier

maximum
Maximum

Importance
Importance
Total score

Production Volume

Volume of
1 How many work shifts per day for production? M 20 one 2 two 4 one or two 3 three 5 5 20 100
Production
Volume of more than
2 How many production line do you have in your facility? H 30 one 2 two 3 5 not sure 4 5 30 150
Production two
Volume of
3 How many employees you have? M 20 >500 5 100-500 3 10-100 2 <10 1 5 20 100
Production
Volume of What is your production volume per year (units or 100,000 - 1
4 H 30 >1 million 5 3 <100,000 2 not sure 3 5 30 150
Production pounds)? million
Volume of $ 100,000 -
5 What is your sales volume per year? M 20 >$ 1 million 5 3 < $100,000 2 not sure 3 5 20 100
Production 1 million

Total: 25 120 600

Figure 2. Questions related to relative production volume.

46
Figure 3. Current sampling plan view describes the sample plan category code, risk
level, current guideline state, production volume level, data entry and data report view.

Cycle Start: 4/20/2013

Current Plan Date: 4/26/2013

Sampling Plan: 2C medium Risk: high

Current State: 0

Enter new data View Reports

Zone Site Taken: Goal: Remaining: Completed:

1 1a 2 9 7
1b 2 9 7
1c 2 9 7
1d 2 9 7
1e 3 9 6
1f 1 9 8
1g 1 9 8
1h 1 9 8
1i 1 9 8
1j 1 9 8
2 2a 2 7 5
2b 2 7 5
2c 1 7 6
2d 1 7 6
2e 1 7 6
2f 1 7 6
2g 1 7 6
2h 0 7 7
3 3a 2 8 6
3b 1 8 7
3c 1 8 7
3d 1 8 7
3e 1 8 7
Grand Total 31 186 155

47
 Site Zone Description
1a 1 Conveyor Belt 1
These are your sampling sites. Pleaseremember
1b 1 Conveyor Belt 2
to ALWAYSassign a site to zone "1", "2", or "3".
1c 1 Conveyor Belt 3
1d 1 Processing Machine 1 Once you have finished editing sites, press the
1e 1 Processing Machine 2 button below.
1f 1 Main area floor (left)
1g 1 Main area floor (right)
1h 1 Main area ceiling
1i 1 Main area west wall
1j 1 Main area east wall
2a
Edits
2 Hallway 1
2b 2 hallway 2
Complete
2c 2 hallway 3
2d 2 Facility North wall
2e 2 Facility South wall
2f 2 General office 1
2g 2 General Office 2
3a 3 Facility Entrance
3b 3 Emergency Entrance 1
3c 3 Emergency entrance 2
3d 3 Shipping Dock 1
3e 3 Shipping Dock 2
2h 2 Random Area 1
Figure 4. Sample site view describes site code, zone, sample description or location.

48
 Current Cycle Results
Current 1 Current 1
EB-Test Results S-Test Results
Total # EB_pass Total # S_Test
Main
Zone Site Fail Pass Total Samples Zone Site Pass Total Samples Menu
1 1a 2 2 1 1a 2 2
1b 1 1 2 1b 2 2
1c 2 2 1c 2 2
1d 1 1 2 1d 2 2
Current
1e 3 3 1e 3 3 Plan
1f 1 1 1f 1 1
1g 1 1 1g 1 1
1h 1 1 1h 1 1
1i 1 1 1i 1 1
1j 1 1 1j 1 1
1 Total 2 14 16 1 Total 16 16
2 2a 2 2 2 2a 2 2
2b 2 2 2b 2 2
2c 1 1 2c 1 1
2d 1 1 2d 1 1
2e 1 1 2e 1 1
2f 1 1 2f 1 1
2g 1 1 2g 1 1
2 Total 9 9 2 Total 9 9
3 3a 2 2 3 3a 2 2
3b 1 1 3b 1 1
3c 1 1 3c 1 1
3d 1 1 3d 1 1
3e 1 1 3e 1 1
3 Total 6 6 3 Total 6 6
Total Samples 2 29 31 Total Samples 31 31
Figure 5. Summary report pivot table for both enterobacteriaceae “EB” count and
Salmonella “S” test result, including zone, total number of samples and site location.

49
Figure 6. Sample data entry window includes (date sample taken, date sample
analyzed, sample zone, sample site code, and enterobacteriaceae “EB” count and
Salmonella “S” test result.

50
Figure 7. Initial sample plan software interface.

51
Table 1. Summary of question response scores required for each of the 9 possible
starting environmental sampling plans.

Product/
Volume
Process Risk Initial
Question Plant Product/
Question Sample
Response Volume process risk
Response Plan Code
Scores Sum
Scores Sum

<300 low >1500 low 1A

<300 low 1000-1499 medium 1B
<300 low <1000 high 1C

300-450 medium >1500 low 2A

300-450 medium 1000-1499 medium 2B
300-450 medium <1000 high 2C

>450 high >1500 low 3A

>450 high 1000-1499 medium 3B

>450 high <1000 high 3C

52
Table 2. Summary of total environmental samples required, and maximum number of
unacceptable (failed) samples, for each of the 9 possible starting environmental
sampling plans.

Environmental samples / month Sample fail limit

Initial
Plant Product/ per month
Sample
Volume process risk
Plan zone 1 zone 2 zone 3 "S" "EB"
Total (50%) (30%) (20%)
1A low low 30 15 9 6 1 10

1B low medium 60 30 18 12 1 20

1C low high 90 45 27 18 1 30

2A medium low 60 30 18 12 1 30

2B medium medium 120 60 36 24 1 50

2C medium high 180 90 54 36 1 70

3A high low 90 45 27 18 1 40

3B high medium 180 90 54 36 1 70

3C high high 270 135 81 54 1 120

All samples can be tested for both Salmonella “S” (qualitative) and Enterobacteriaceae
“EB” (quantitative)

Monthly sample total should be divided so that an equivalent number is collected each
week.

Number of sample sites per zone per week should be at least 3 unless fewer samples
are required

53
Table 3. Data collection report view includes date sample taken, date sample analyzed,
sampling zone, sample site code, enterobacteriaceae “EB” test result, Salmonella “S”
test result, and sample test result validation (Pass/Fail).

DateTaken DateAnalyzed Zone Site Code EB_Test S_Test EB_pass

3/1/2013 3/2/2013 1 1a 0 0 Pass
3/2/2013 3/2/2013 1 1a 0 0 Pass
3/3/2013 3/4/2013 1 1b 20 0 Fail
3/3/2013 3/4/2013 1 1b 0 0 Pass
3/3/2013 3/4/2013 1 1c 12 0 Fail
3/3/2013 3/4/2013 1 1c 0 0 Pass
3/3/2013 3/4/2013 1 1d 4 0 Pass
3/3/2013 3/4/2013 1 1d 12 0 Fail
3/4/2013 3/4/2013 1 1e 33 0 Fail
3/4/2013 3/4/2013 1 1e 100 0 Fail
3/4/2013 3/4/2013 1 1f 0 0 Pass
3/4/2013 3/4/2013 1 1g 0 0 Pass
3/4/2013 3/4/2013 1 1h 0 0 Pass
3/4/2013 3/4/2013 1 1i 120 0 Fail
3/4/2013 3/4/2013 1 1j 0 0 Pass
3/5/2013 3/6/2013 2 2a 0 0 Pass
3/5/2013 3/6/2013 2 2a 0 0 Pass
3/5/2013 3/6/2013 2 2b 30 0 Pass
3/5/2013 3/6/2013 2 2b 0 0 Pass
3/5/2013 3/6/2013 2 2c 90 0 Pass
3/6/2013 3/6/2013 2 2d 0 0 Pass
3/6/2013 3/6/2013 2 2e 0 0 Pass

54
Table 4. Number of samples required for each of 9 sampling plans at sampling state 0
(initial) and state 1 with guideline for adjusting sample plan totals by zone (“if # of failed
Salmonella samples > 1 sample per zone (1, 2 or both 1.2), then shift to higher risk
sampling plan (low --> High or med. --> high) for 1 months, untill you receive a negative
result for 2 weeks”).

result for 2 week.

receive a negative

27
27
27

36
36
36

54
54
54
months, Till you z3
med. --> high) for 1
(low --> High or
State 1

40
40
40

54
54
54

81
81
81
z2

risk sampling plan

then shift to higher
(1, 2 or both 1.2),
sample per zone

135
135
135
52
52
52

90
90
90
z1

samples > 1
if # of failed
month
Total

sample fail limit per

1
1
1

1
1
1

1
1
Total Salmonella 1
zone 1 zone 2 zone 3
Salmonella samples per month

(20%)

12
18

12
24
36

18
36
54
6
(30%)

18
27

18
36
54

27
54
81
9
(50%)

135
15
30
45

30
60
90

45
90
Total

120
180

180
270
30
60
90

60

90
process risk
Product/

medium

medium

medium
high

high

high
low

low

low
medium
medium
medium
Volume
Plant

high
high
high
low
low
low
Sample
Initial

Plan

1B
1C

2A
2B
2C

3A
3B
3C
1A

55
 .

If # failed samples
in a month > than
limit, then increase
EB samples by 2X
each zone for 1
month till you
receive under norm
results

Total EB sample fail

limit per month
from all zones
Initial EB samples per month State 2
Plant Product/
Sample zone 1 zone 2 zone 3
Volume process risk
Plan Total (50%) (30%) (20%) Total z1 z2 z3
1A low low 30 15 9 6 10 30 18 12
1B low medium 60 30 18 12 20 60 36 24
month until you receive under norm results”).

1C low high 90 45 27 18 30 90 54 36

2A medium low 60 30 18 12 30 60 36 24
2B medium medium 120 60 36 24 50 120 72 48
2C medium high 180 90 54 36 70 180 108 72

3A high low 90 45 27 18 40 90 54 36
3B high medium 180 90 54 36 70 180 108 72
3C high high 270 135 81 54 120 270 162 108
samples in a month > than limit, then increase EB samples by 2X each zone for 1
(initial) and state 2 with guideline for adjusting sample plan totals by zone (“If # failed
Table 5. Number of samples required for each of 9 sampling plans at sampling state 0

56
 if # of failed
samples > limit for
3 consecutive
months, then shift
to higher risk
sampling plan (low -
-> med., or med. --
> high) for 1
months till you
receive under norm
results

Total EB sample fail

limit per month
from all zones
Initial EB samples per month State 3
Plant Product/
Sample zone 1 zone 2 zone 3
Volume process risk
Plan Total (50%) (30%) (20%) Total z1 z2 z3
1A low low 30 15 9 6 10
1B low medium 60 30 18 12 20 30 18 12
1C low high 90 45 27 18 30 45 27 18

2A medium low 60 30 18 12 30
2B medium medium 120 60 36 24 50 60 36 24
2C medium high 180 90 54 36 70 90 54 36

3A high low 90 45 27 18 40
3B high medium 180 90 54 36 70 90 54 36
3C high high 270 135 81 54 120 135 81 54
med., or med. --> high) for 1 months till you receive under norm results”).
Table 6. Number of samples required for each of 9 sampling plans at sampling state 0
(initial) and state 3 with guideline for adjusting sample plan totals by zone (“if # of failed
samples > limit for 3 consecutive months, then shift to higher risk sampling plan (low -->

57
 if # of failed
samples < limit for
3 consecutive
months, then you
may reduce total #
of sample analyses
by 10% (by pooling
samples) if you
have a good history
of your results

Total EB sample fail

limit per month
from all zones
Initial EB samples per month State 4
Plant Product/
Sample zone 1 zone 2 zone 3
Volume process risk
Plan Total (50%) (30%) (20%) Total z1 z2 z3
1A low low 30 15 9 6 10 14 8 5
1B low medium 60 30 18 12 20 27 16 11
1C low high 90 45 27 18 30 41 24 16

2A medium low 60 30 18 12 30 27 16 11
2B medium medium 120 60 36 24 50 54 32 22
2C medium high 180 90 54 36 70 81 49 32

3A high low 90 45 27 18 40 41 24 16
3B high medium 180 90 54 36 70 81 49 32
3C high high 270 135 81 54 120 122 73 49
samples < limit for 3 consecutive months, then you may reduce total # of sample
analyses by 10% (by pooling samples) if you have a good history of your results”).
Table 7. Number of samples required for each of 9 sampling plans at sampling state 0
(initial) and state 4 with guideline for adjusting sample plan totals by zone (“if # of failed

58
REFERENCES

Cordier, J.-L. 2008. Production of powdered infant formula and microbiological control

measures, pp. 145-185. In J. M. Farber, and S. Forsythe (eds.). Enterobacter

sakazakii. ASM Press. Washington, DC.

Dahms, S. 2003. Microbiological sampling plans –Statistical aspects. In Mitt.

Lebensm. Hyg. 95 (1): 32-44.

EFSA (European Food Safety Authority). 2007. Opinion of the Scientific Panel on

Biological Hazards (BIOHAZ) on the request for review of the opinion on

microbiological risks in infant formulae and follow-on formulae with regard to

Enterobacteriaceae as indicators. Adopted on January 24, 2007. The EFSA

Journal. 444:1-14.

Eifert, J.D. and Arritt, F.M. 2002. Evaluating environmental sampling data and

sampling plans. Dairy, Food and Environ. Sanit. 22(5): 333-339.

Environmental Protection Agency (EPA). December, 2002. Guidance on Choosing a

Sampling Design for Environmental Data Collection. Retrieved, May 17 2011,

from: www.epa.gov/QUALITY/qs-docs/g5s-final.pdf

Farmer, J. J., M. A. Asbury, F. W. Hickman, and D. J. Brenner. 1980. Enterobacter

sakazakii—a new species of Enterobacteriaceae isolated from clinical

specimens. Int. J. Syst. Bacteriol. 30:569–584.

Grocery Manufacturers Association (GMA). 2009. Control of Salmonella In low-

moisture foods, February 4, 2009 (Minor corrections March 16, 2009).

Washington, DC: Grocery Manufacturers Association.

59
Institute of Food Technologists (IFT). 2004. Scientific Status Summary: Bacteria

Associated with Foodborne Diseases. Retrieved, April 03 2012, from

http://www.ift.org/knowledge-center/read-ift-publications/science-

reports/scientific-status- summaries/bacteria-associated-with-foodborne-

diseases.aspx

International Commission on Microbiological Specifications for Foods (ICMSF). 2011.

Milk and Dairy products, Ch. 23 In: Microorganisms in Foods 8: Use of Data for

Assessing Process Control and Product Acceptance. Springer: New York City,

New York.

International Commission on Microbiological Specifications for Foods (ICMSF). 2002.

Sampling to Assess Control of the Environment, Ch. 11, In: Microorganisms in

Foods 7: Microbiological testing in food safety management. Kluwer Academic

Plenum Publishers: New York City, New York.

International Commission on Microbiological Specifications for Foods (ICMSF). 1986.

Microorganisms in Foods 2: Sampling for Microbiological Analysis: Principles and

Specific Applications. University of Toronto Press: Toronto.

Jay LS, Davos D, Dundas M, Frankish E, Lightfoot D. 2003. Salmonella. In: AD

Hocking (eds). Foodborne Microorganisms of Public Health Significance. Sixth

edition. Waterloo, NSW: Australian Institute of Food Science and Technology

Incorporated, NSW Branch, Food Biology Group.

Legan, D., & Vandeven, M. H. 2003. Sampling techniques. In T. McMeekin

(Ed.), Detecting Pathogens in Food (pp. 20-51). Cambridge, London: Woodhead

Publishing Limited

60
Marriott, N. G. and Gravani, R. B. 2006. Principles of Food Sanitation: Food

Contamination Sources (pp. 78). New York, New York: Springer Science +

Business Media, Inc.

New Zealand Food Safety Authority (NZFSA), October, 2010. Institute of

Environmental Science and Research Limited (ESR): Risk profile: Salmonella

(non typhoidal) in cereal grains.

NIST. 2012. Challenges in Microbial Sampling in the Indoor Environment – Workshop

Summary Report. National Institute of Standards and Technology (NIST)

Technical Note 1737. Retrieved, Jan 18 2012, from: www.nist.gov/

Norberg, S., Stanton, C., Ross, R. P., Hill, C., Fitzgerald, G., & Cotter, P. 2011.

Cronobacter spp. in powdered infant formula. Journal of Food Protection, 75(3),

607-620.

Scallan, E., Hoekstra, R.M., Angulo, F.J., Tauxe, R.V., Widdowson, M.A., Roy, S.L.,

Jones, J.L., Griffin, P.M. 2011. Foodborne illness acquired in the United States--

major pathogens. Emerg. Infect. Dis. 17: 7-15.

Sutton, S. 2010. The environmental monitoring program in a GMP environment.

Journal of GXP Compliance 14(3): 22-30.

Tompkin, B. 2004. Environmental sampling – A tool to verify the effectiveness of

preventive hygiene measures. In Mitt. Lebensm. Hyg. 95 (1): 45–51.

U.S. Food and Drug Administration (FDA). 2012 Bad Bug Book (second edition):

Foodborne Pathogenic Microorganisms and Natural Toxins Handbook.

Salmonella spp. Retrieved, April 26 2012, from:

61
 http://www.fda.gov/downloads/Food/FoodSafety/FoodborneIllness/FoodborneIlln

essFoodbornePathogensNaturalToxins/BadBugBook/UCM297627.pdf

USDA FSIS. 2013. Intensified verification testing (ivt) protocol for sampling of product,

food contact surfaces, and environmental surfaces for Listeria monocytogenes

(lm) or Salmonella spp. U.S. Department of Agriculture, Food Safety and

Inspection Service, Directive 10300.1. Retrieved, Feb10 2012, from:

www.fsis.usda.gov/OPPDE/rdad/FSISDirectives/10300.1.pdf

World Health Organization (WHO). 2007. Safe preparation, storage and handling of

powdered infant formula guidelines. Switzerland: World Health Organization.

Retrieved, Mar 17 2011, from:

http://www.who.int/entity/foodsafety/publications/micro/pif_guidelines.pdf

62
 Sampling Test Improvement Workbook

Current Initial
Plan Setup

Edit
Data Sampling
Entry Locations

View Guide
Reports (HowTo)

Export
Data
Developed by Hassan Masri and Ivan Volonsevich under guidance of Joseph D. Eifert
Virginia Tech
FS Procedures Questions
Do you follow Good Manufacturing Practices
(GMPs)?
Thanks for
Yes
trying out the
tool!
No
To begin, please
answer the
Usually questions listed
here. Just scroll
Don't Know down until you
answer all of the
questions, and click
"next" to see the
Do you have an environmental monitoring sampling next category of
plan (including sample locations, frequencies, types,
sizes)?
Yes

No

Partially

Don't Know

Do you know the source of your raw materials and

ingredients?

Yes

No

Sometimes

Don't Know
Do you maintain a certificate of analysis for your
materials?

Yes

No

Sometimes

Don't Know

Do you require specifications for all ingredients?

Yes

No

Usually

sometimes

Did / Will your staff receive food safety and GMP

training this year?

Yes

No

some employees
Next
only when hired
Process Questions
What is the physical condition of your process
facility?

Very Good

Good

Fair

unknown

Does your processing room have positive pressure?

Yes

No

Usually

unknown

* Do you use a HEPA filter in you air unit?

Yes

No

sometimes

unknown
What type of barrier between zones do you have in
the process line?

Physical

Virtual

both

Don't Know

Are you knowledgeable of the type of hazards to

your product?

Yes

No

probably

Don't Know

What is the risk of pathogen contamination to your

product?

High

Medium

Low

not sure
 Does the risk of your product to the consumer
change after production?

increased

no change

reduced

not sure

Is any source of pathogen contamination eliminated

from the process line?

Yes

No

Usually

Next
not sure
Do you have a list of locations to collect environment
samples from?

Yes

No

not sure

Don't Know

How many environmental sampling zones do you

have?

one

two

three or four

Don't Know

Do you know the microbiological test specifications

of your analyses?

Yes

No

Somewhat

Don't Know
Next
Sampling and Testing Questions
* Do you know the (pathogenic) microorganism of
concern?
Yes

No

not sure

Don't Know

* Do you know the (pathogenic) microorganism of

concern?

Yes

No

not sure

Don't Know

Do you test product samples for pathogens?

Yes

No

sometimes

Don't Know

Do you test for indicator or non-pathogenic

organisms?

Yes

No

Sometimes

Don't Know
Is your staff trained to collect environment samples?

Yes

No

Somewhat

Don't Know

Do you know how many environmental samples do

you take per lot or per day?

Yes

No

not sure

Don't Know
Production Volume Questions

How many work shifts per day for production?

one

two

one or two

three

How many production line do you have in your

facility?

one

two

more than two

not sure

How many employees you have?

>500

100-500

10-100

<10
What is your production volume per year (units or
pounds)?

>1 million

100,000 - 1 million

<100,000

not sure

What is your sales volume per year?

>$ 1 million

$ 100,000 - 1 million

< $100,000

not sure Finishe

Based on our analysis of risk:
your answers, we have
concluded that this is the
risk level representative
high
of your process.

volume:
Also, we have concluded
that this is representative
of your volume level.
medium

Plan: The number

Based on these two represents
factors, this is the
suggested initial sampling
plan.
2C volume. 1 is the
lowest, 3 is
highest. The
letter represents

Edit
Now you should enter your Sampling
Locations
Site Zone Description
1a 1 Conveyor Belt 1
These are your
1b 1 Conveyor Belt 2
sampling sites.
1c 1 Conveyor Belt 3
Please remember
1d 1 Processing Machine 1
to ALWAYS assign
1e 1 Processing Machine 2 a site to zone "1",
1f 1 Main area floor (left) "2", or "3".
1g 1 Main area floor (right)
1h 1 Main area ceiling Once you have
1i 1 Main area west wall
1j 1 Main area east wall Edits
2a 2 Hallway 1 Complete
2b 2 hallway 2
2c 2 hallway 3
2d 2 Facility North wall
2e 2 Facility South wall
2f 2 General office 1
2g 2 General Office 2
3a 3 Facility Entrance
3b 3 Emergency Entrance 1
3c 3 Emergency entrance 2
3d 3 Shipping Dock 1
3e 3 Shipping Dock 2
2h 2 Random Area 1
 Cycle Start: 4/20/2013
Current Plan Date: 5/9/2013

Sampling Plan: 2C medium Risk: high

Current State: 0

Enter new data View Reports

Zone Site Taken: Goal: Remaining: Completed:

1 1a 6 9 3
1b 6 9 3
1c 6 9 3
1d 6 9 3
1e 7 9 2
1f 3 9 6
1g 3 9 6
1h 3 9 6
1i 3 9 6
1j 3 9 6
2 2a 6 7 1
2b 6 7 1
2c 3 7 4
2d 3 7 4
2e 3 7 4
2f 3 7 4
2g 2 7 5
2h 1 7 6
3 3a 6 8 2
3b 3 8 5
3c 3 8 5
3d 3 8 5
3e 3 8 5
Grand Total 91 186 95
DateTakenDateAnalyzed
Zone Site EB_Test S_Test EB_pass Current
3/1/2013 3/2/2013 1 1a 0 0 Pass 0
3/2/2013 3/2/2013 1 1a 0 0 Pass 0
3/3/2013 3/4/2013 1 1b 20 0 Fail 0
3/3/2013 3/4/2013 1 1b 0 0 Pass 0
3/3/2013 3/4/2013 1 1c 12 0 Fail 0
3/3/2013 3/4/2013 1 1c 0 0 Pass 0
3/3/2013 3/4/2013 1 1d 4 0 Pass 0
3/3/2013 3/4/2013 1 1d 12 0 Fail 0
3/4/2013 3/4/2013 1 1e 33 0 Fail 0
3/4/2013 3/4/2013 1 1e 100 0 Fail 0
3/4/2013 3/4/2013 1 1f 0 0 Pass 0
3/4/2013 3/4/2013 1 1g 0 0 Pass 0
3/4/2013 3/4/2013 1 1h 0 0 Pass 0
3/4/2013 3/4/2013 1 1i 120 0 Fail 0
3/4/2013 3/4/2013 1 1j 0 0 Pass 0
3/5/2013 3/6/2013 2 2a 0 0 Pass 0
3/5/2013 3/6/2013 2 2a 0 0 Pass 0
3/5/2013 3/6/2013 2 2b 30 0 Pass 0
3/5/2013 3/6/2013 2 2b 0 0 Pass 0
3/5/2013 3/6/2013 2 2c 90 0 Pass 0
3/6/2013 3/6/2013 2 2d 0 0 Pass 0
3/6/2013 3/6/2013 2 2e 0 0 Pass 0
3/6/2013 3/6/2013 2 2f 0 0 Pass 0
3/7/2013 3/7/2013 2 2g 190 0 Fail 0
3/7/2013 3/7/2013 3 3a 0 0 Pass 0
3/7/2013 3/7/2013 3 3a 0 0 Pass 0
3/7/2013 3/7/2013 3 3b 0 0 Pass 0
3/7/2013 3/7/2013 3 3c 355 0 Pass 0
3/7/2013 3/7/2013 3 3d 0 0 Pass 0
3/7/2013 3/7/2013 3 3e 280 0 Pass 0
3/9/2013 3/9/2013 1 1a 0 0 Pass 0
3/9/2013 3/9/2013 1 1a 0 0 Pass 0
3/9/2013 3/9/2013 1 1b 20 0 Fail 0
3/9/2013 3/9/2013 1 1b 0 0 Pass 0
3/9/2013 3/9/2013 1 1c 44 0 Fail 0
3/9/2013 3/9/2013 1 1c 0 0 Pass 0
3/9/2013 3/9/2013 1 1d 4 0 Pass 0
3/11/2013 3/11/2013 1 1d 12 0 Fail 0
3/11/2013 3/11/2013 1 1e 0 0 Pass 0
3/11/2013 3/11/2013 1 1e 66 0 Fail 0
3/11/2013 3/11/2013 1 1f 0 0 Pass 0
3/11/2013 3/11/2013 1 1g 54 0 Fail 0
3/11/2013 3/11/2013 1 1h 0 0 Pass 0
3/11/2013 3/11/2013 1 1i 8 0 Pass 0
3/11/2013 3/11/2013 1 1j 0 0 Pass 0
3/11/2013 3/11/2013 2 2a 34 0 Pass 0
3/11/2013 3/11/2013 2 2a 0 0 Pass 0
3/11/2013 3/11/2013 2 2b 30 0 Pass 0
3/13/2013 3/13/2013 2 2b 0 0 Pass 0
3/13/2013 3/13/2013 2 2c 66 0 Pass 0
3/13/2013 3/13/2013 2 2d 0 0 Pass 0
3/13/2013 3/13/2013 2 2e 0 0 Pass 0
3/13/2013 3/13/2013 2 2f 0 0 Pass 0
3/13/2013 3/13/2013 2 2h 50 0 Pass 0
3/13/2013 3/13/2013 3 3a 0 0 Pass 0
3/13/2013 3/13/2013 3 3a 0 0 Pass 0
3/13/2013 3/13/2013 3 3b 777 0 Fail 0
3/13/2013 3/13/2013 3 3c 90 0 Pass 0
3/13/2013 3/13/2013 3 3d 300 0 Pass 0
3/13/2013 3/13/2013 3 3e 28 0 Pass 0
3/16/2013 3/19/2013 1 1a 0 0 Pass 0
3/16/2013 3/19/2013 1 1a 0 0 Pass 0
3/16/2013 3/19/2013 1 1b 20 0 Fail 0
3/16/2013 3/19/2013 1 1b 0 0 Pass 0
3/16/2013 3/19/2013 1 1c 0 0 Pass 0
3/16/2013 3/19/2013 1 1c 0 0 Pass 0
3/16/2013 3/19/2013 1 1d 100 0 Fail 0
3/16/2013 3/19/2013 1 1d 12 0 Fail 0
3/16/2013 3/19/2013 1 1e 0 0 Pass 0
3/19/2013 3/19/2013 1 1e 0 0 Pass 0
3/19/2013 3/19/2013 1 1f 0 0 Pass 0
3/19/2013 3/19/2013 1 1g 0 0 Pass 0
3/19/2013 3/19/2013 1 1h 300 0 Fail 0
3/19/2013 3/19/2013 1 1i 8 0 Pass 0
3/19/2013 3/19/2013 1 1j 0 0 Pass 0
3/19/2013 3/19/2013 2 2a 0 0 Pass 0
3/19/2013 3/19/2013 2 2a 39 0 Pass 0
3/19/2013 3/19/2013 2 2b 30 0 Pass 0
3/21/2013 3/21/2013 2 2b 0 0 Pass 0
3/21/2013 3/21/2013 2 2c 444 0 Fail 0
3/21/2013 3/21/2013 2 2d 0 0 Pass 0
3/21/2013 3/21/2013 2 2e 0 0 Pass 0
3/21/2013 3/21/2013 2 2f 0 0 Pass 0
3/21/2013 3/21/2013 2 2g 654 0 Fail 0
3/21/2013 3/21/2013 3 3a 0 0 Pass 0
3/21/2013 3/21/2013 3 3a 0 0 Pass 0
3/21/2013 3/21/2013 3 3b 0 0 Pass 0
3/21/2013 3/21/2013 3 3c 452 0 Pass 0
3/21/2013 3/21/2013 3 3d 660 0 Fail 0
3/21/2013 3/21/2013 3 3e 28 0 Pass 0
3/23/2013 3/23/2013 1 1a 0 0 Pass 0
3/23/2013 3/23/2013 1 1a 0 0 Pass 0
3/23/2013 3/23/2013 1 1b 233 0 Fail 0
3/23/2013 3/23/2013 1 1b 0 0 Pass 0
3/23/2013 3/23/2013 1 1c 32 0 Fail 0
3/23/2013 3/23/2013 1 1c 0 0 Pass 0
3/23/2013 3/23/2013 1 1d 4 0 Pass 0
3/23/2013 3/23/2013 1 1d 12 0 Fail 0
3/23/2013 3/23/2013 1 1e 0 0 Pass 0
3/26/2013 3/28/2013 1 1e 0 0 Pass 0
3/26/2013 3/28/2013 1 1f 0 0 Pass 0
3/26/2013 3/28/2013 1 1g 0 0 Pass 0
3/26/2013 3/28/2013 1 1h 0 0 Pass 0
3/26/2013 3/28/2013 1 1i 8 0 Pass 0
3/26/2013 3/28/2013 1 1j 753 0 Fail 0
3/26/2013 3/28/2013 2 2a 0 0 Pass 0
3/26/2013 3/28/2013 2 2a 0 0 Pass 0
3/26/2013 3/28/2013 2 2b 30 0 Pass 0
3/26/2013 3/28/2013 2 2b 0 0 Pass 0
3/26/2013 3/28/2013 2 2c 66 0 Pass 0
3/28/2013 3/28/2013 2 2d 0 0 Pass 0
3/28/2013 3/28/2013 2 2e 529 0 Fail 0
3/28/2013 3/28/2013 2 2f 0 0 Pass 0
3/28/2013 3/28/2013 2 2h 240 0 Fail 0
3/28/2013 3/28/2013 3 3a 0 0 Pass 0
3/28/2013 3/28/2013 3 3a 0 0 Pass 0
3/28/2013 3/28/2013 3 3b 483 0 Pass 0
3/28/2013 3/28/2013 3 3c 90 0 Pass 0
3/28/2013 3/28/2013 3 3d 0 0 Pass 0
3/28/2013 3/28/2013 3 3e 28 0 Pass 0
4/1/2013 4/3/2013 1 1a 369 0 Fail 0
4/1/2013 4/3/2013 1 1a 0 0 Pass 0
4/1/2013 4/3/2013 1 1b 0 0 Pass 0
4/1/2013 4/3/2013 1 1b 0 0 Pass 0
4/1/2013 4/3/2013 1 1c 0 0 Pass 0
4/1/2013 4/3/2013 1 1c 351 0 Fail 0
4/1/2013 4/3/2013 1 1d 4 0 Pass 0
4/1/2013 4/3/2013 1 1d 0 0 Pass 0
4/1/2013 4/3/2013 1 1e 0 0 Pass 0
4/1/2013 4/3/2013 1 1e 0 0 Pass 0
4/1/2013 4/3/2013 1 1f 0 0 Pass 0
4/1/2013 4/3/2013 1 1g 0 0 Pass 0
4/6/2013 4/7/2013 1 1h 0 0 Pass 0
4/6/2013 4/7/2013 1 1i 8 0 Pass 0
4/6/2013 4/7/2013 1 1j 0 0 Pass 0
4/6/2013 4/7/2013 2 2a 0 0 Pass 0
4/6/2013 4/7/2013 2 2a 0 0 Pass 0
4/6/2013 4/7/2013 2 2b 300 0 Fail 0
4/6/2013 4/7/2013 2 2b 0 0 Pass 0
4/6/2013 4/7/2013 2 2c 50 0 Pass 0
 4/6/2013 4/7/2013 2 2d 0 0 Pass 0
4/6/2013 4/7/2013 2 2e 0 0 Pass 0
4/6/2013 4/7/2013 2 2f 0 0 Pass 0
4/6/2013 4/7/2013 2 2g 10 0 Pass 0
4/6/2013 4/7/2013 3 3a 0 0 Pass 0
4/6/2013 4/7/2013 3 3a 0 0 Pass 0
4/6/2013 4/7/2013 3 3b 0 0 Pass 0
4/6/2013 4/7/2013 3 3c 90 0 Pass 0
4/6/2013 4/7/2013 3 3d 0 0 Pass 0
4/6/2013 4/7/2013 3 3e 80 0 Pass 0
4/10/2013 4/12/2013 1 1a 0 0 Pass 0
4/10/2013 4/12/2013 1 1a 0 0 Pass 0
4/10/2013 4/12/2013 1 1b 0 0 Pass 0
4/10/2013 4/12/2013 1 1b 0 0 Pass 0
4/10/2013 4/12/2013 1 1c 0 0 Pass 0
4/10/2013 4/12/2013 1 1c 0 0 Pass 0
4/10/2013 4/12/2013 1 1d 4 0 Pass 0
4/10/2013 4/12/2013 1 1d 0 0 Pass 0
4/10/2013 4/12/2013 1 1e 0 0 Pass 0
4/10/2013 4/12/2013 1 1e 0 0 Pass 0
4/10/2013 4/12/2013 1 1f 0 0 Pass 0
4/10/2013 4/12/2013 1 1g 0 0 Pass 0
4/10/2013 4/12/2013 1 1h 0 0 Pass 0
4/10/2013 4/12/2013 1 1i 8 0 Pass 0
4/12/2013 4/12/2013 1 1j 0 1 Pass 0
4/12/2013 4/12/2013 2 2a 0 0 Pass 0
4/12/2013 4/12/2013 2 2a 0 0 Pass 0
4/12/2013 4/12/2013 2 2b 30 0 Pass 0
4/12/2013 4/12/2013 2 2b 0 0 Pass 0
4/12/2013 4/12/2013 2 2c 66 0 Pass 0
4/12/2013 4/12/2013 2 2d 0 0 Pass 0
4/12/2013 4/12/2013 2 2e 0 0 Pass 0
4/12/2013 4/12/2013 2 2f 0 0 Pass 0
4/12/2013 4/12/2013 2 2h 50 0 Pass 0
4/12/2013 4/12/2013 3 3a 0 0 Pass 0
4/12/2013 4/12/2013 3 3a 0 0 Pass 0
4/15/2013 4/15/2013 3 3b 0 1 Pass 0
4/15/2013 4/15/2013 3 3c 90 0 Pass 0
4/15/2013 4/15/2013 3 3d 300 0 Pass 0
4/15/2013 4/15/2013 3 3e 28 0 Pass 0
4/20/2013 4/20/2013 1 1a 0 0 Pass 1
4/20/2013 4/20/2013 1 1a 0 0 Pass 1
4/20/2013 4/20/2013 1 1b 20 0 Fail 1
4/20/2013 4/20/2013 1 1b 0 0 Pass 1
4/20/2013 4/20/2013 1 1c 0 0 Pass 1
4/20/2013 4/20/2013 1 1c 0 0 Pass 1
4/20/2013 4/20/2013 1 1d 4 0 Pass 1
4/20/2013 4/20/2013 1 1d 12 0 Fail 1
4/20/2013 4/20/2013 1 1e 0 0 Pass 1
4/20/2013 4/20/2013 1 1e 0 0 Pass 1
4/20/2013 4/20/2013 1 1f 0 0 Pass 1
4/20/2013 4/20/2013 1 1g 0 0 Pass 1
4/20/2013 4/20/2013 1 1h 0 0 Pass 1
4/20/2013 4/20/2013 1 1i 8 0 Pass 1
4/20/2013 4/20/2013 1 1j 0 0 Pass 1
4/20/2013 4/20/2013 2 2a 0 0 Pass 1
4/24/2013 4/24/2013 2 2a 0 0 Pass 1
4/24/2013 4/24/2013 2 2b 30 0 Pass 1
4/24/2013 4/24/2013 2 2b 0 0 Pass 1
4/24/2013 4/24/2013 2 2c 66 0 Pass 1
4/24/2013 4/24/2013 2 2d 0 0 Pass 1
4/24/2013 4/24/2013 2 2e 0 0 Pass 1
4/24/2013 4/24/2013 2 2f 0 0 Pass 1
4/24/2013 4/24/2013 2 2g 80 0 Pass 1
4/24/2013 4/24/2013 3 3a 0 0 Pass 1
4/24/2013 4/24/2013 3 3a 0 0 Pass 1
4/24/2013 4/24/2013 3 3b 0 0 Pass 1
4/24/2013 4/24/2013 3 3c 90 0 Pass 1
4/24/2013 4/24/2013 3 3d 410 0 Pass 1
4/24/2013 4/24/2013 3 3e 28 0 Pass 1
4/28/2013 4/29/2013 1 1a 0 0 Pass 1
4/28/2013 4/29/2013 1 1a 0 0 Pass 1
4/28/2013 4/29/2013 1 1b 20 0 Fail 1
4/28/2013 4/29/2013 1 1b 0 0 Pass 1
4/28/2013 4/29/2013 1 1c 0 0 Pass 1
4/28/2013 4/29/2013 1 1c 0 0 Pass 1
4/28/2013 4/29/2013 1 1d 4 0 Pass 1
4/28/2013 4/29/2013 1 1d 12 0 Fail 1
4/28/2013 4/29/2013 1 1e 0 0 Pass 1
4/28/2013 4/29/2013 1 1e 0 0 Pass 1
4/28/2013 4/29/2013 1 1f 0 0 Pass 1
4/28/2013 4/29/2013 1 1g 0 0 Pass 1
4/28/2013 4/29/2013 1 1h 0 0 Pass 1
4/28/2013 4/29/2013 1 1i 8 0 Pass 1
4/28/2013 4/29/2013 1 1j 0 0 Pass 1
4/28/2013 4/29/2013 2 2a 0 0 Pass 1
4/28/2013 4/29/2013 2 2a 0 0 Pass 1
4/28/2013 4/29/2013 2 2b 30 0 Pass 1
4/28/2013 4/29/2013 2 2b 0 0 Pass 1
4/28/2013 4/29/2013 2 2c 66 0 Pass 1
4/28/2013 4/29/2013 2 2d 0 0 Pass 1
4/28/2013 4/29/2013 2 2e 0 0 Pass 1
4/28/2013 4/29/2013 2 2f 0 0 Pass 1
4/28/2013 4/29/2013 2 2h 240 0 Fail 1
4/28/2013 4/29/2013 3 3a 0 0 Pass 1
4/28/2013 4/29/2013 3 3a 0 0 Pass 1
4/28/2013 4/29/2013 3 3b 0 0 Pass 1
4/28/2013 4/29/2013 3 3c 90 0 Pass 1
4/28/2013 4/29/2013 3 3d 770 0 Fail 1
4/28/2013 4/29/2013 3 3e 28 0 Pass 1
5/2/2013 5/3/2013 1 1a 369 0 Fail 1
5/2/2013 5/3/2013 1 1a 0 0 Pass 1
5/2/2013 5/3/2013 1 1b 0 0 Pass 1
5/2/2013 5/3/2013 1 1b 0 0 Pass 1
5/2/2013 5/3/2013 1 1c 0 0 Pass 1
5/2/2013 5/3/2013 1 1c 351 0 Fail 1
5/2/2013 5/3/2013 1 1d 4 0 Pass 1
5/2/2013 5/3/2013 1 1d 0 0 Pass 1
5/2/2013 5/3/2013 1 1e 0 0 Pass 1
5/2/2013 5/3/2013 1 1e 0 0 Pass 1
5/2/2013 5/3/2013 1 1f 0 0 Pass 1
5/2/2013 5/3/2013 1 1g 0 0 Pass 1
5/6/2013 5/7/2013 1 1h 0 0 Pass 1
5/6/2013 5/7/2013 1 1i 8 0 Pass 1
5/6/2013 5/7/2013 1 1j 0 0 Pass 1
5/6/2013 5/7/2013 2 2a 0 0 Pass 1
5/6/2013 5/7/2013 2 2a 0 0 Pass 1
5/6/2013 5/7/2013 2 2b 300 0 Fail 1
5/6/2013 5/7/2013 2 2b 0 0 Pass 1
5/6/2013 5/7/2013 2 2c 50 0 Pass 1
5/6/2013 5/7/2013 2 2d 0 0 Pass 1
5/6/2013 5/7/2013 2 2e 0 0 Pass 1
5/6/2013 5/7/2013 2 2f 0 0 Pass 1
5/6/2013 5/7/2013 2 2g 10 0 Pass 1
5/6/2013 5/7/2013 3 3a 0 0 Pass 1
5/6/2013 5/7/2013 3 3a 0 0 Pass 1
5/6/2013 5/7/2013 3 3b 0 0 Pass 1
5/6/2013 5/7/2013 3 3c 90 0 Pass 1
5/6/2013 5/7/2013 3 3d 0 0 Pass 1
5/6/2013 5/7/2013 3 3e 80 0 Pass 1
5/10/2013 5/12/2013 1 1a 0 0 Pass 0
5/10/2013 5/12/2013 1 1a 0 0 Pass 0
5/10/2013 5/12/2013 1 1b 0 0 Pass 0
5/10/2013 5/12/2013 1 1b 0 0 Pass 0
5/10/2013 5/12/2013 1 1c 0 0 Pass 0
5/10/2013 5/12/2013 1 1c 0 0 Pass 0
5/10/2013 5/12/2013 1 1d 4 0 Pass 0
5/10/2013 5/12/2013 1 1d 0 0 Pass 0
5/10/2013 5/12/2013 1 1e 0 0 Pass 0
5/10/2013 5/12/2013 1 1e 0 0 Pass 0
5/10/2013 5/12/2013 1 1f 0 0 Pass 0
5/10/2013 5/12/2013 1 1g 0 0 Pass 0
5/10/2013 5/12/2013 1 1h 0 0 Pass 0
5/10/2013 5/12/2013 1 1i 8 0 Pass 0
5/10/2013 5/12/2013 1 1j 0 0 Pass 0
5/10/2013 5/12/2013 2 2a 0 0 Pass 0
5/10/2013 5/12/2013 2 2a 0 0 Pass 0
5/12/2013 5/12/2013 2 2b 30 0 Pass 0
5/12/2013 5/12/2013 2 2b 0 0 Pass 0
5/12/2013 5/12/2013 2 2c 66 0 Pass 0
5/12/2013 5/12/2013 2 2d 0 0 Pass 0
5/12/2013 5/12/2013 2 2e 0 0 Pass 0
5/12/2013 5/12/2013 2 2f 0 0 Pass 0
5/12/2013 5/12/2013 2 2h 50 0 Pass 0
5/12/2013 5/12/2013 3 3a 0 0 Pass 0
5/12/2013 5/12/2013 3 3a 0 0 Pass 0
5/15/2013 5/15/2013 3 3b 0 0 Pass 0
5/15/2013 5/15/2013 3 3c 90 0 Pass 0
5/15/2013 5/15/2013 3 3d 300 0 Pass 0
5/15/2013 5/15/2013 3 3e 28 0 Pass 0
5/20/2013 5/20/2013 1 1a 0 0 Pass 0
5/20/2013 5/20/2013 1 1a 0 0 Pass 0
5/20/2013 5/20/2013 1 1b 20 0 Fail 0
5/20/2013 5/20/2013 1 1b 0 0 Pass 0
5/20/2013 5/20/2013 1 1c 0 0 Pass 0
5/20/2013 5/20/2013 1 1c 0 0 Pass 0
5/20/2013 5/20/2013 1 1d 4 0 Pass 0
5/20/2013 5/20/2013 1 1d 12 0 Fail 0
5/20/2013 5/20/2013 1 1e 0 0 Pass 0
5/20/2013 5/20/2013 1 1e 0 0 Pass 0
5/20/2013 5/20/2013 1 1f 0 0 Pass 0
5/20/2013 5/20/2013 1 1g 0 0 Pass 0
5/20/2013 5/20/2013 1 1h 0 0 Pass 0
5/20/2013 5/20/2013 1 1i 8 0 Pass 0
5/20/2013 5/20/2013 1 1j 0 0 Pass 0
5/20/2013 5/20/2013 2 2a 0 0 Pass 0
5/20/2013 5/20/2013 2 2a 0 0 Pass 0
5/24/2013 5/24/2013 2 2b 30 0 Pass 0
5/24/2013 5/24/2013 2 2b 0 0 Pass 0
5/24/2013 5/24/2013 2 2c 66 0 Pass 0
5/24/2013 5/24/2013 2 2d 0 0 Pass 0
5/24/2013 5/24/2013 2 2e 0 0 Pass 0
5/24/2013 5/24/2013 2 2f 0 0 Pass 0
5/24/2013 5/24/2013 2 2g 80 0 Pass 0
5/24/2013 5/24/2013 3 3a 0 0 Pass 0
5/24/2013 5/24/2013 3 3a 0 0 Pass 0
5/24/2013 5/24/2013 3 3b 0 0 Pass 0
5/24/2013 5/24/2013 3 3c 90 0 Pass 0
5/24/2013 5/24/2013 3 3d 410 0 Pass 0
5/24/2013 5/24/2013 3 3e 28 0 Pass 0
5/24/2013 5/24/2013 1 1a 0 0 Pass 0
5/24/2013 5/24/2013 1 1a 0 0 Pass 0
5/24/2013 5/24/2013 1 1b 20 0 Fail 0
5/24/2013 5/24/2013 1 1b 0 0 Pass 0
5/24/2013 5/24/2013 1 1c 0 0 Pass 0
5/24/2013 5/24/2013 1 1c 0 0 Pass 0
5/24/2013 5/24/2013 1 1d 4 0 Pass 0
5/24/2013 5/24/2013 1 1d 12 0 Fail 0
5/24/2013 5/24/2013 1 1e 0 0 Pass 0
5/24/2013 5/24/2013 1 1e 0 0 Pass 0
5/24/2013 5/24/2013 1 1f 0 0 Pass 0
5/24/2013 5/24/2013 1 1g 0 0 Pass 0
5/24/2013 5/24/2013 1 1h 0 0 Pass 0
5/24/2013 5/24/2013 1 1i 8 0 Pass 0
5/28/2013 5/29/2013 1 1j 0 0 Pass 0
5/28/2013 5/29/2013 2 2a 0 1 Pass 0
5/28/2013 5/29/2013 2 2a 0 0 Pass 0
5/28/2013 5/29/2013 2 2b 30 0 Pass 0
5/28/2013 5/29/2013 2 2b 0 0 Pass 0
5/28/2013 5/29/2013 2 2c 66 0 Pass 0
5/28/2013 5/29/2013 2 2d 0 0 Pass 0
5/28/2013 5/29/2013 2 2e 0 0 Pass 0
5/28/2013 5/29/2013 2 2f 0 0 Pass 0
5/28/2013 5/29/2013 2 2h 240 0 Fail 0
5/28/2013 5/29/2013 3 3a 0 0 Pass 0
5/28/2013 5/29/2013 3 3a 0 0 Pass 0
5/28/2013 5/29/2013 3 3b 0 0 Pass 0
5/28/2013 5/29/2013 3 3c 90 0 Pass 0
5/28/2013 5/29/2013 3 3d 770 0 Fail 0
5/28/2013 5/29/2013 3 3e 28 0 Pass 0
4/22/2013 4/23/2013 1 1e 0 0 Pass 1
 Current Cycle
Current 1 Current 1
EB-Test S-Test
Total # EB_pass Total # S_Test

Total Total Main

Zone Site Fail Pass Samples Zone Site Pass Samples Menu
1 1a 2 2 1 1a 2 2
1b 1 1 2 1b 2 2
1c 2 2 1c 2 2
1d 1 1 2 1d 2 2
1e 3 3 1e 3 3
1f 1 1 1f 1 1 Curren
1g 1 1 1g 1 1
1h 1 1 1h 1 1
1i 1 1 1i 1 1
1j 1 1 1j 1 1
1 Total 2 14 16 1 Total 16 16
2 2a 2 2 2 2a 2 2
2b 2 2 2b 2 2
2c 1 1 2c 1 1
2d 1 1 2d 1 1
2e 1 1 2e 1 1
2f 1 1 2f 1 1
2g 1 1 2g 1 1
2 Total 9 9 2 Total 9 9
3 3a 2 2 3 3a 2 2
3b 1 1 3b 1 1
3c 1 1 3c 1 1
3d 1 1 3d 1 1
3e 1 1 3e 1 1
3 Total 6 6 3 Total 6 6
Total Samples 2 29 31 Total Samples 31 31
 if # of failed samples > 1
sample per zone (1, 2 or both
Total
"S"
1.2), then shift to higher risk
sample sampling plan (low --> High or
fail limit med. --> high) for 1 months,
per Till you receive a negative
month result for 2 week.
Initial Salmonella samples per month State 1
Plant Product/
Sample zone zone zone
Volume process risk
Plan Total 1 2 3 Total z1 z2 z3
1A low low 30 15 9 6 1 52 40 27
1B low medium 60 30 18 12 1 52 40 27
1C low high 90 45 27 18 1 52 40 27

2A medium low 60 30 18 12 1 90 54 36
2B medium medium 120 60 36 24 1 90 54 36
2C medium high 180 90 54 36 1 90 54 36

3A high low 90 45 27 18 1 135 81 54

3B high medium 180 90 54 36 1 135 81 54
3C high high 270 135 81 54 1 135 81 54

refer to PivotTable summary: Mo-Salm

 if # of failed samples > limit
Total EB If # failed samples in a month > for 3 consecutive months, if # of failed samples < limit for 3
sample
than limit, then increase EB then shift to higher risk consecutive months, then you may
fail limit
per samples by 2X each zone for 1 sampling plan (low --> reduce total # of sample analyses
month month till you receive under norm med., or med. --> high) for by 10% (by pooling samples) if you
from all results 1 months till you receive have a good history of your results
zones under norm results
EB samples per month State 2 State 3 State 4
zone 1 zone 2 zone 3
Total (50%) (30%) (20%) Total z1 z2 z3 z1 z2 z3 z1 z2 z3
30 15 9 6 10 30 18 12 14 8 5
60 30 18 12 20 60 36 24 30 18 12 27 16 11
90 45 27 18 30 90 54 36 45 27 18 41 24 16

60 30 18 12 30 60 36 24 27 16 11
120 60 36 24 50 120 72 48 60 36 24 54 32 22
180 90 54 36 70 180 108 72 90 54 36 81 49 32

90 45 27 18 40 90 54 36 41 24 16
180 90 54 36 70 180 108 72 90 54 36 81 49 32
270 135 81 54 120 270 162 108 135 81 54 122 73 49

refer to PivotTable summary: Mo-EB Mo-EB Mo-EB