Techniques in Shelf Life Evaluation of Food Products

Yuthana Phimolsiripol, Faculty of Agro-Industry, Chiang Mai University, Chiang Mai, Thailand; and Food Innovation and Packaging
Center, Chiang Mai University, Chiang Mai, Thailand
Panuwat Suppakul, Faculty of Agro-Industry, and Center of Advanced Studies for Agriculture and Food (CASAF), Kasetsart
University, Bangkok, Thailand; and Center for Intelligent Agro-Food Packaging, College of Life Science and Biotechnology, Dongguk
University, Seoul, Republic of Korea
Ó 2016 Elsevier Inc. All rights reserved.

Basic Concept 1
Factors Affecting the Shelf Life Test 1
Understand the Product 1
Understand the Factors Affecting Product Quality 2
Identify the Critical Quality Based on Several Guidelines 2
Understand the Concerted Series 2
Types of Shelf Life Tests and Design 2
Static Tests 2
Accelerated Tests 2
Use/Abuse Tests 2
Storage and Sampling Design 2
Basic Design Sampling 3
Reverse Design Sampling 3
Semireverse Design Sampling 3
Storage Test and Trial Conditions 3
Storage Conditions 3
Commonly Used for Fixed-Storage Conditions 3
Samples for Storage Trials 3
Sampling Schedule 3
Kinetic Reactions 3
Accelerated Shelf Life Simulation 5
Steps of Accelerated Shelf Life Simulation 5
Arrhenius Model 5
Conclusion 8
References 8

Basic Concept

Nowadays, consumers are increasingly demanding high food quality, and they have corresponding expectations that the quality of
their food will be maintained at a high level during the period between purchase and consumption. Shelf life is an important attri-
bute of all foods, including raw materials, ingredients, and manufactured products. It is defined as the time during which the food
product will (1) remain safe; (2) be certain to retain desired sensory, chemical, physical, and microbiological characteristics; (3)
comply with any label declaration of nutritional data; and (4) be acceptable to the consumer (Earle and Earle, 2008).
For shelf life determination, the evaluator is required to understand basic food science and technology including food processing,
food analysis, food packaging, and statistical techniques. Shelf life studies must be carried out only when foods are correctly processed,
packed, and stored, ready for purchase and consumption (Steele, 2004). The end of shelf life can be determined from (1) relevant food
legislation; (2) guidelines given by enforcement authorities or agencies; (3) guides provided by independent professional bodies such
as IFT; (4) current industrial best practice; (5) self-imposed end-point assessment; and (6) market information.

Factors Affecting the Shelf Life Test

Understand the Product
The evaluators should ensure that they understand the product very well (Man, 2002). In general, the food product can be classified
into three types as follows: (1) Highly perishable foods (milk, fresh meat, fresh fruits, and vegetables) – These are very short-shelf
life products that are subject to microbiological and/or enzymatic deterioration. Measurements are taken every day in order to deter-
mine shelf life. (2) Semiperishable foods (pasteurized milk, smoked meats, cheeses, and some bakery products) – These are
short-to-medium shelf life products and may contain natural inhibitors or have received minimal preservative treatment.

Reference Module in Food Sciences http://dx.doi.org/10.1016/B978-0-08-100596-5.03293-5 1

2 Techniques in Shelf Life Evaluation of Food Products

Measurements are made every week in order to determine shelf life. (3) Highly stable foods (dried food, canned food, and frozen
food) – These are medium-to-long shelf life products that have received a thermal process or are maintained in specific conditions.
Measurements are made every week or monthly in order to determine shelf life.

Understand the Factors Affecting Product Quality

Many factors can influence shelf life and can be categorized into compositional and environmental factors. Compositional factors
are the properties of the final product including food composition; water activity (aw); pH value; total acidity; type of acid; redox
potential (Eh); available oxygen, nutrients, natural microflora, and surviving microbiological counts; natural biochemistry of the
product formulation (enzymes, chemical reactants); use of preservatives in product formulation (e.g., salt); and concentration of
reactant, inhibitor, and catalyst.
For environmental factors, they are those factors the final product encounters as it moves through the food chain including
time–temperature profile during processing; pressure in the headspace; temperature control during storage and distribution; rela-
tive humidity (RH) during processing, storage, and distribution; exposure to light (UV and IR) during processing, storage, and
distribution; environmental microbial counts during processing, storage, and distribution; atmospheric composition within
packaging; subsequent heat treatment (e.g., reheating or cooking before consumption); and distributor, retailer, and consumer
handling.

Identify the Critical Quality Based on Several Guidelines

These guidelines include (1) government law – the critical quality is mainly for microbial safety; (2) customer standard – this may
be higher than the government standard; (3) competitors – the critical quality can be related to a competitor’s product; and (4)
consumers – they provide the best judgment of the critical quality of each food.

Understand the Concerted Series

These series of biochemical/physicochemical reactions are required to insightfully understand and identify mechanisms responsible
for spoilage or loss of desirable characteristics. Several reactions involve food deterioration such as microbiological spoilage, chem-
ical and enzymatic activity including rancidity, browning reactions, and moisture and/or other vapor migration.

Types of Shelf Life Tests and Design

The shelf life test can be divided into three types (Kilcast and Subramaniam, 2000) including the following:

Static Tests
Product stored under a given set of environmental conditions. This test requires a long time to observe changes and is expensive. It
gives no information on the effects of stress and comes the closest to distribution conditions.

Accelerated Tests
Product stored under a range of environmental conditions (usually temperature or relative humidity). The conditions of this test are
selected to cover the expected range encountered and can be achieved in a relatively short period of time. It also provides kinetic data,
and the test conditions should not alter the normal anticipated path affecting shelf life. The results must be interpreted with care.

Use/Abuse Tests
Product cycled through environmental variables. The test is used to assess the product and package as a unit. These tests use cycles of
variables that are equal to or beyond that expected under actual conditions. It is often used to determine the effects of transport.

Storage and Sampling Design

Design of the shelf life test is very important (Hough, 2010). There are several relevant criteria for shelf life testing. The factor has
enough effect on the quality changes, and the critical parameter is directly related to quality and consumer acceptance. In addition,
the parameter is easy to determine with simple and practical design. There are enough samples, and the quality of sample can be
controlled. The evaluator should concern about the method of measurement: it is standard and appropriate. This is relevant to the
cost of testing. Homogeneity of the sample such as yoghurt with fruit pieces should be prepared very well to make sure that the
sample represents the whole product.
For microbial safety consideration, it is very critical. The sample should not be tested for other qualities (especially sensory tests)
prior to microbiological testing results. For frozen product, it is noticed that biochemical changes in food can occur during frozen
storage. Therefore, the relationship of several parameters can be related and should be considered.
For accelerated test involved with temperature, the temperature differences can vary from 5 to 10
C, and extremely high storage
temperatures should be avoided. The sample during storage tests should include around six to eight points of sampling time. The
quality changes of the product should be 30–50% of the original to represent an accurate trend of deterioration. In general, product
sampling during storage tests can be divided into three types.
 Techniques in Shelf Life Evaluation of Food Products 3

Basic Design Sampling

The sample will be collected for product quality assessment following the sampling plan, for example, once a week until the end of
shelf life analysis. The advantage of this design is that the frequency of sampling can be increased or decreased as appropriate.
However, if the sample would be tested for sensory panels, especially trained panels, the cost of this design will be high due to
repeated training of panels before each test (Hough, 2010).

Reverse Design Sampling

The sample will be collected following the sampling plan; however, the sample will be maintained under controlled conditions. The
controlled condition ensures that the sample quality is quite stable, such as freezing or refrigeration. The sample will be assessed
after the completion of all samplings at the same time. The advantage of this design is the lower cost for sensory testing. However,
a problem may arise if the sample changes are faster or later than expected (Hough, 2010).

Semireverse Design Sampling

This is another option for sampling and is a combination of the basic and reverse sampling. The sample may be collected for testing
after two to three rounds of sampling or in the middle of the sampling plan to ensure that the product quality changes are within the
expected range of deterioration.

Storage Test and Trial Conditions

The guidelines for shelf life testing conditions (Kilcast and Subramaniam, 2000) are as follows.

Storage Conditions
The sample should be stored in several conditions including optimum conditions, typical or average conditions, and worst-case
conditions.

Commonly Used for Fixed-Storage Conditions

l Frozen: 18
C or lower (RH is usually close to 100%)
l Chilled: 0–5
C, with a maximum of 8
C (RH is usually very high)
l Temperate: 25
C, 75% RH
l Tropical: 38
C, 90% RH
l Control: optimum conditions for each product

Samples for Storage Trials

Number of samples should be concerned. Size and packaging of samples should be the same and close to real conditions.

Sampling Schedule
The sampling plan is depended on the typical shelf life: (1) Short-shelf life products: Up to 1 week (e.g., ready meals), samples can
be taken off daily for evaluation. (2) Medium-shelf life products: Up to 3 weeks (e.g., some ambient cakes and pastry), samples can
be taken off on days 0, 7, 14, 19, 21, and 25. (3) Long-shelf life products: Up to 1 year (e.g., some breakfast cereals), samples can be
taken off at monthly intervals or at months 0, 1, 2, 3, 6, 12, and (perhaps) 18.

Kinetic Reactions

The simplest technique for shelf life testing is the kinetic reaction approach. To predict the shelf life, the kinetic data is used to eval-
uate how the deterioration process behaves as a function of time. It is the concept of quantification of the quality of food products
based on reaction change (Labuza and Riboh, 1982; Corradini and Peleg, 2007). The kinetic equation may be expressed as:

d½A
rA ¼  ¼ k½An
dt
Following the chemical reactions, k is the kinetic constant, t is time, and n is order of reaction. The change in concentration A
of a component of interest is monitored. The quality factors [A] are usually quantifiable chemical, physical, microbiological, or
sensory parameters, such as the loss of a nutrient or characteristic of flavor or formation of an off-flavor. The time to reach the
value of the quality index (Ats ) at a specified condition (i.e., the shelf life) ts is inversely proportional to the rate constant at these
conditions.

fq ðAts Þ
ts ¼
k
4 Techniques in Shelf Life Evaluation of Food Products

Table 1 Quality function of reaction order

Apparent reaction order Quality function

0 At  A0 or At /A0
1 ln (At  A0) or ln (At /A0)
2 1/At  1/A0
n (n s 1) (1/n  1)$ðAt1n  A01n Þ

concentraon concentraon concentraon

zero order 1st order 2nd order

me me me

Figure 1 Pattern of concentration changes in different reaction orders.

The forms of the quality function of the food for an apparent zero-, first-, second- or nth-order reaction are presented in Table 1
and Figure 1, which show the different patterns of the reaction orders. For zero-order reactions, the reaction rate is independent of
the concentration of a reactant. First-order reactions depend on a single reactant, and the exponent value is one. For second-order
reactions, the reaction rate may be proportional to one concentration squared or to the product of two concentrations (Labuza and
Riboh, 1982). Most reactions that are responsible for shelf life loss based on a characteristic physicochemical, chemical, or microbial
index include the following: zero order (e.g., frozen food overall quality, Maillard browning) and first order (e.g., vitamin loss,
oxidative color loss, microbial growth), as shown in Table 1 and Figure 1. Several published research has been applied to the kinetic
model to describe the temperature dependence. The quality parameters can be chemical properties such as peroxide value in extra
virgin olive oil (Calligaris et al., 2006); vitamin C loss in citrus juice concentrate (Burdurlu et al., 2006); hydrogen ion in coffee
liquids (Manzocco and Nicoli, 2007); physical property such as color loss in fresh-cut asparagus (Sothornvit and Kiatchanapaibul,
2009); weight loss in frozen bread dough (Phimolsiripol et al., 2011); or sensory property such as sensory attributes in frozen
shrimp (Tsironi et al., 2009).
The simplest and frequently used method to find the reaction order is the integration method (van Boekel, 2008). The
process starts from (1) guess reaction order; (2) integrate; (3) linearize by linear regression; (4) plot experimental data in line-
arized form; and (5) if data fit a straight line, then guess is right, if not start again. The simple selection is higher coefficient of
determination (R2), and slope is reaction rate constant. Determination of reaction order has been exemplified in Example 1 as
follows:

Example 1: Determination of reaction order.

Find the reaction order of ascorbic acid loss on multivitamin storage. The concentrations of ascorbic acid (A) at different times when stored at 40
C
are presented in Table 2.
Solution 1:
Guess reaction order by applying the quality function as shown in Table 3. Then, plot graph to compare each order as shown in Figure 2. From the
plot in Figure 2, the best fit is the first-order reaction.

Table 2 Concentrations of ascorbic acid

at different times when stored at 40
C

Time (days) Ascorbic acid (mg ml1)

10 271
20 109
30 58
40 30.5
50 18
60 10
 Techniques in Shelf Life Evaluation of Food Products 5

Table 3 Comparison of reaction orders

Time (days) Zero order (A) First order (ln A) Second order (1/A)

10 271 5.602119 0.00369

20 109 4.691348 0.009174
30 58 4.060443 0.017241
40 30.5 3.417727 0.032787
50 18 2.890372 0.055556
60 10 2.302585 0.1
R2 0.7522 0.9916 0.8758

Accelerated Shelf Life Simulation

Food industries require a relatively short time to obtain the necessary information for determining the shelf life of their products.
For practical reasons, when the actual storage time is long, the industry usually uses accelerated test techniques that considerably
shorten the process of obtaining the necessary experimental data. Accelerated shelf life simulation will refer to any method that
is capable of evaluating product stability, based on data that is obtained in a significantly shorter period than the actual shelf
life of the product (Steele, 2004). As aimed at shortening the time required to estimate a shelf life, the concepts of accelerated shelf
life simulation include: (1) The assumption is that by storing food at a higher temperature, any adverse effect on its storage behavior
and hence shelf life may become apparent in a shorter time. (2) The shelf life under normal storage conditions can be estimated by
extrapolation using the data obtained from the accelerated determination.

Steps of Accelerated Shelf Life Simulation

When the accelerated shelf life testing is performed, the compositional factors must be kept constant (Hough, 2010). For full
concern, the microbial safety and quality parameters must be defined and determined as the first priority for shelf life of product.
Then, key deteriorative reactions that will cause quality loss and consumer unacceptability are selected. In case of proper packaging,
the cost-effective packages are used except the effect of packaging materials is concerned. The different packaging materials will be
used for comparison. Otherwise, the other environmental factors, including temperature, relative humidity, light, etc., as the desired
kinetically active factors for acceleration of the deterioration process are selected.
Next step, the evaluator should determine how long the product is to be held at each test temperature and determine the
frequency of test time between tests at temperature. The overall number of samples that must be stored at each temperature value
will be calculated.
Final step is that a kinetic study of the deterioration process at such levels of the accelerating factors is run. However, if the rate of
deterioration is too fast or too slow, then the frequency of sampling can be increased or decreased as appropriate. The shelf life
prediction can be predicted by constructing the shelf life plots to determine the shelf life under normal storage conditions using
the extrapolating data or the kinetic model to predict shelf life at actual storage conditions.

Arrhenius Model
The Arrhenius model is a classical model that relates the rate of a chemical reaction to the changes in temperature. This model is
widely applied in several processing and storage tests as affected by temperature (Corradini and Peleg, 2007; Phimolsiripol et al.,
2008, 2011). The model is represented by:
 
Ea
k ¼ k0 exp  ;
RT
where k0 is the rate constant, Ea is the energy of activation, R is the gas constant (1.9872 cal mol1 K1 or 8.3144 J mol1 K1), and
T is the absolute temperature (Kelvin, K).
The Arrhenius equation can be put in standard slope-intercept form by taking the natural logarithm by:

ln k ¼ ln A Ea =RT or
ln k ¼ ln A ðEa =RÞ ð1=TÞ
h h h h
y ¼b þa x
The steps of the Arrhenius model for shelf life determination are as follows: (1) Find the reaction order following the kinetic
reaction concept as stated in the Section Kinetic Reactions. (2) Follow the Arrhenius relationship. (3) Plot the Arrhenius
6 Techniques in Shelf Life Evaluation of Food Products

Zero order
300
250
y = -4.5871x + 243.3
200 R² = 0.7522
150

A
100
50
0
-50
0 10 20 30 40 50 60 70
Time

First order
6
y = -0.0644x + 6.0818
5 R² = 0.9916

4
ln A

0
0 10 20 30 40 50 60 70
Time

Second order
0.12
0.1 y = 0.0018x - 0.0272
0.08 R² = 0.8758

0.06
1/A

0.04
0.02
0
-0.02
0 10 20 30 40 50 60 70
Time

Figure 2 Plots of data representing quality function of reaction order.

relationship. (4) Fit the curve using linear regression. (5) Slope of plot between ln k versus 1/T is Ea/R. For instance, Arrhenius model
having been employed for shelf life prediction is demonstrated in Example 2.

Example 2: Application of the Arrhenius model for shelf life prediction.

To investigate the effect of propolis extract obtained from a domestic source in Thailand on the antioxidant properties of six mayonnaise formulas
during storage (Santiprasert et al., 2009). Propolis was extracted using 95% ethanol in a 1:3 (w/w) ratio for 14 days. The effects of propolis extract (0%,
0.25%, and 0.5%) and fish oil (0% and 2%) on mayonnaise stored at 5
C, 35
C, or 55
C for 0, 1, 2, 3, and 4 weeks were determined. The results of
acid value (AV) tests during storage as shown in Table 4 can be described by a zero-order reaction. The correlation of AV with the sensory test indicates
that the shelf life of mayonnaise will be rejected when the AV is more than 10 g oleic acid kg fat1 week1. The initial AV of both samples is 3.8. Predict
the shelf lives of mayonnaise formulas 1 and 3 stored at 30
C.
Solution 2:
From the data in Table 4, the kinetic reaction can be described by a zero-order reaction. The kinetic rate constants (k) are applied in the Arrhenius
model. The Arrhenius plot is presented in Figure 3. Table 5 shows estimated kinetic parameters for the development of lipid oxidation during storage of
different mayonnaise formulas. Figure 4 shows the appearance of mayonnaise stored at 5, 35, and 55
C. This sample was analyzed for rancidity retar-
dation. However, the temperature at 55
C was not suitable for the shelf life study because the mayonnaise sample was rejected due to emulsion
separation.
 Techniques in Shelf Life Evaluation of Food Products 7

Table 4 Rate constants for lipid oxidation assuming zero-order reaction kinetics of different
mayonnaise formulas

k (g oleic acid kg fat1 week1)

Formula Propolis (%) Fish oil (%) 5
C 35
C 55
C

1 0.00 0 0.153 0.396 0.494

2 0.25 0 0.188 0.369 0.489
3 0.50 0 0.098 0.237 0.343
4 0.00 2 0.308 0.410 0.410
5 0.25 2 0.288 0.356 0.437
6 0.50 2 0.368 0.439 0.576

0.0

-0.5

-1.0
ln k

-1.5

-2.0

-2.5
0.003 0.0031 0.0032 0.0033 0.0034 0.0035 0.0036 0.0037
1/T (1/K)

Figure 3 Arrhenius plot (ln k vs. 1/T ) for lipid oxidation (AV value) rate constants of different mayonnaise formulas during storage. A, 0% prop-
olis and 0% fish oil; ,, 0.25% propolis and 0% fish oil; D, 0.5% propolis and 0% fish oil; B, 0% propolis and 2% fish oil; *, 0.25% propolis and
2% fish oil; C, 0.5% propolis and 2% fish oil.

Table 5 Estimated kinetic parameters for the development of lipid oxidation during storage of different mayonnaise formulas

ln k0 (g oleic
Formula Propolis (%) Fish oil (%) Ea (kJ mol1) acid kg fat1 week1) R2

1 0.00 0 18.1 5.99 0.98

2 0.25 0 14.7 4.69 0.99
3 0.50 0 19.2 6.01 0.99
4 0.00 2 4.6 0.85 0.87
5 0.25 2 6.2 1.41 0.98
6 0.50 2 6.5 1.78 0.92

o o o
5C 35 C 55 C

Figure 4 Mayonnaise appearance after storage at different temperatures.

8 Techniques in Shelf Life Evaluation of Food Products

Table 6 Shelf life prediction of mayonnaise formulas 1 and 3 stored at 30
C

Formula ln k ¼ ln k0  Ea/RT k (g oleic acid kg fat1 week1) A  Ao ¼ kt Shelf life (week)

1 1.194 0.302 (3.8  10) ¼ (0.302)*t 20.5

3 1.611 0.199 (3.8  10) ¼ (0.199)*t 31.1

Results from this example showed that fish oil increased the development of lipid oxidation as reflected by a significantly
(p < 0.05) higher AV. Addition of propolis extract significantly reduced (p < 0.05) the high AV. The predicted shelf lives of mayon-
naise formulas 1 and 3 stored at 30
C will be 20 and 31 weeks, respectively, as calculated in Table 6. This confirmed that the addi-
tion of 2% propolis can extend the shelf life of mayonnaise by 11 weeks when compared to the control (no propolis).

Conclusion

Shelf life is very important for food products, and it must accommodate safety, quality, organoleptic, and appearance requirements
to assure customer acceptance. To determine the shelf life, accelerated is another option and the Arrhenius model is a classical
model that relates the rate of a chemical reaction to the changes in temperature. This model is widely applied in several processing
and storage tests as affected by temperature. However, it is concerned that shelf life determination is valid only for the exact product
composition, packaging, and processing conditions combination.

References

van Boekel, M.A.J.S., 2008. Kinetic modeling of food quality: a critical review. Compr. Rev. Food Sci. Food Saf. 7, 144–158.
Burdurlu, H.S., Koca, N., Karadeniz, F., 2006. Degradation of vitamin C in citrus juice concentrates during storage. J. Food Eng. 74, 211–216.
Calligaris, S., Sovrano, S., Manzocco, L., Nicoli, M.C., 2006. Influence of crystallisation on the oxidative stability of extra virgin olive oil. J. Agric. Food Chem. 54, 529–535.
Corradini, M.G., Peleg, M., 2007. Shelf-life estimation from accelerated storage data. Trends Food Sci. Technol. 18, 37–47.
Earle, M., Earle, R., 2008. Case Studies in Food Product Development. CRC Press, Boca Raton.
Hough, G., 2010. Sensory Shelf Life Estimation of Food Products. Taylor & Francis, Boca Raton.
Kilcast, D., Subramaniam, P., 2000. The Stability and Shelf-life of Food. Woodhead Publishing Ltd., Cambridge.
Labuza, T.P., Riboh, D., 1982. Theory and application of Arrhenius kinetics to the prediction of nutrient losses in foods. Food Technol. 36, 66–74.
Man, D., 2002. Shelf Life. Blackwell Science, Oxford.
Manzocco, L., Nicoli, M.C., 2007. Modelling the effect of water activity and storage temperature on chemical stability of coffee brews. J. Agric. Food Chem. 55, 6521–6526.
Phimolsiripol, Y., Siripatrawan, U., Tulyathan, V., Cleland, D.J., 2008. Effects of freezing and temperature fluctuations during frozen storage on frozen dough and bread quality.
J. Food Eng. 84, 48–56.
Phimolsiripol, Y., Siripatrawan, U., Cleland, D.J., 2011. Weight loss of frozen bread dough under isothermal and fluctuating temperature storage conditions. J. Food Eng. 106,
134–143.
Santiprasert, S., Sanguandeekul, R., Phimolsiripol, Y., 2009. Effect of propolis extract on rancidity of mayonnaise. In: Proceeding of the 47th Kasetsart University Annual Conference.
Kasetsart University, Bangkok, Thailand, pp. 233–240.
Sothornvit, R., Kiatchanapaibul, P., 2009. Quality and shelf-life of washed fresh-cut asparagus in modified atmosphere packaging. LWT Food Sci. Technol. 42, 1484–1490.
Steele, R., 2004. Understanding and Measuring the Shelf-life of Food. CRC Press, Boca Raton.
Tsironi, T., Dermesonlouoglou, E., Giannakourou, M., Taoukis, P., 2009. Shelf life modelling of frozen shrimp at variable temperature conditions. LWT Food Sci. Technol. 42,
664–671.