Effective Sampling Plan & Sampling Techniques To Get The Best Analytical Samples

Assessing the quality of liquid milk is done through performing quality assurance tests, which begin by sampling of the liquid milk. It is important for safety and economic reasons. All dairy products must be tested for quality assurance purposes.

Sampling facilitates comprehensive analysis of any given product for quality assurance. A sample represents the quality of the entire volume of the product under consideration.

Sampling Dairy Products: Basics

Sampling is necessary because it is practically impossible to test the whole batch. We will examine some of the reasons why quality assessment is important, especially in the dairy industry and some important things to consider while doing this.

Reasons to examine the quality of milk:

Examination will reveal the chemical composition of milk such as fat, protein, lactose contents and vitamins
Analysis will help determine the quantity of developed acidity in the milk
To determine the presence of adulterants or preservatives in the milk
Examination will help estimate bacterial content and the amount of sediments in the milk

It is important to carry out a proper sampling to obtain accurate results.

Important terms to consider when sampling

Sampling: – drawing out of a portion of substance/material from the total population

Population: – total quantity of material whose characteristics you are seeking through sampling

Sample: – a small portion of the population which you obtain in such a way as to represent the characteristics of that particular population as closely as possible. Only a representative sample will produce consistent results.

You can have a single sample from different sub-samples, which you collect from different parts of the population. You get a representative sample by randomly drawing from the population.

Random sample: – a random sample occurs when every object in the population has an equal chance of selection

Biased sample: – a sample where some objects in the population have unlikely chance of selection

Laboratory sample: – when a sample is too large for a normal lab examination procedure, you take its fraction. This fraction is the lab sample.

The primary objective of sample selection is to ensure that the features of the lab sample represent the population.

It is necessary to limit sample size to cut down on time, cost, and personnel required for that analytical work. Limited samples exhibit limited estimates of the whole population.

Equipment for sampling dairy products

i) Stirrer/Planter

This equipment is useful for attaining consistency in the milk/cream for a representative sampling of these dairy products. It is made of stainless steel, aluminium or any other metal that does not adversely affect the milk.

It consists of a perforated disk with a long handle fixed at its center to help in moving the equipment through the milk to achieve effective mixing.

ii) Dipper/Scoop

Consists of a small cup fixed to one end of a long handle. It is mainly used to collect sample from the container. They vary in capacity.

iii) Tube sampler

Consists of an open-ended tube with provision for closing one of the ends. It has the following advantages:

You can obtain a representative sample regardless of how long the milk has stood before sampling
You can collect a sample from a column of milk representative of the batch (from the top to the bottom of the container).

Sampling of milk and milk products differs in terms of materials you use and the purpose of the sample.

Things to remember when sampling dairy products

If you have to transport the milk over long distances, chill and transport it using insulated containers/refrigerated trucks.
Ensure that you examine the samples as soon as you draw them since chemical and microbiological changes may affect the results obtained
Avoid violent mixing of the milk since the viscosity of milk will not allow air bubbles to rise to the top. Churning of the fat globules will also take place
Unfreeze (thaw) milk (in case of frozen milk) before sampling begins
Ensure you have a uniform mix of the milk from which you intend to draw the sample
Fat is the most variable constituent of a standing milk sample
You should only use chemically preserved milk for a chemical analysis.

Sample selection and sampling plans used for sampling dairy products

Sample selection is necessary because the destruction of the population occurs over time. The process is also tedious and time consuming.

Sampling plan

A sampling plan is necessary because it ensures that the lab sample is truly representative of the entire population.

You should clearly write down your sampling plan in a document that contains the details the analyst will use. Such details may include the sample size, the population, method of collection, and preservation methods.

The sampling plan should also stipulate the required documentation of the sampling procedures. The choice of sampling plan will depend on:

Purpose of the analysis
Property to be analyzed
Nature of the total population and individual samples
Type of analytical technique to be used in characterizing the sample

More details on sampling plan discussed below.

a) Purpose of Analysis

Analysis is done for various reasons, which will affect the sampling plan. They include:

i)Official samples: – Necessary for legal requirements by government labs. This ensures that the suppliers/producers/manufacturers produce safe products that meet legal and labeling requirements e.g. whole milk sampling for butterfat content determination.
ii)Raw material: –Necessary to ascertain the quality of materials a factory uses for processing are compliant. This ensures that the material will withstand the manufacturing processes.

iii) High quality products: – Quality of the products depends on the quality of the raw materials used for making them

iv)Process control samples: – Samples analyzed during processing to ensure that the process is operating in an effective manner. It aims at detecting problems and adjusting accordingly to end up with good quality products.

Techniques used in process controls should produce process results within a short time. This can be done in the following two ways:

Measure the quality of the food online e.g. flow diversion valve
Select/remove samples and test them in a quality assurance laboratory e.g. standardization

v)The finished product sampling: – Aims at ensuring that the finished products are safe and they meet legal requirements

b) Property you intend to measure

Specify the particular property you want to analyze e.g. fat content, microbial load, developed acidity, etc. You can classify these properties as either attributes or variables.

i) Attributes: – properties that a product should either have or not.
ii) Variables: – properties that you can measure on a continuous scale such as the butter fat content, weight, moisture content of milk powder, etc.

Variable sampling requires less samples than attribute samples. The type of property measured should determine the seriousness of the outcome (such as when measuring for harmful microorganisms, there is need for high precision due to the danger it poses than when measuring for fat content).

c) Nature of the population from which you draw the sample

i)Finite population: – has a defined size e.g. tanker full of milk.
ii)Infinite population: – has no defined size e.g. sampling from a conveyor belt which continuously supply the product e.g. packaging machine.

Analysis of finite population provides properties about that population. Analysis from an infinite population gives information about the processes.

Therefore, to facilitate sampling population, it is convenient to divide the infinite population into a number of finite populations e.g. all products produced in one shift forms one finite population. This will help to trace the source of problems in the process.

iii) Continuous population: – no physical separation between the different parts of the sample e.g. well stirred milk in a tanker

iv) Compartmentalized:– split into a number of separate sub-units
v)Homogeneous: – properties of an individual sample is the same as those of the other samples in different regions of the population
vi) Heterogeneous: – characteristics of one sample vary within different locations of the population e.g. raw milk and its butterfat content; a well stirred milk is homogeneous while standing milk is heterogeneous.

Practically, most samples are heterogeneous hence the importance of selecting samples from different locations. You will then mix the parts well to obtain a representative sample.

d) Nature of the test procedure

Test procedure usually takes into account:

Accuracy and precision
The speed of measurement
Cost per analysis
Whether the technique is destructive/non-destructive

When you realize that the procedure is cheap, rapid and non-destructive, take more samples. More samples increase accuracy of the results you intend to achieve by sampling dairy products for analysis.

Effective Sampling Plan: Important Factors to Consider

After looking at the factors to be considered when sampling milk, it is now possible to draft a sampling plan. The sampling plan has a specific objective of ensuring that the obtained sample will be reflective of the qualities that one is seeking to establish about the group.

The plan will clearly indicate the size of the sample and all the important features for examination under every group.

Common features of a sampling plan:

a) Sample size

Determined by:

The variations in the populations
Seriousness of the outcome if a bad sample is not detected
Cost of analysis
Variations of the populations

b) Sample location

No problem when dealing with homogenous population; however, heterogenous population requires consideration of the sampling plans e.g.

Random sampling – random picking of the samples from the population
Systematic sampling – you pick the samples following a given order.
Judgment sampling – you pick samples following past experience.

c) Sample collection

State clearly whether the sampling will be manual or by specialized mechanical devices

Preparation of lab samples

These are the steps to follow:

i) Make the sample homogenous

Most samples are heterogenous due to inter-unit variation or intra-unit variation.

Inter-unit variation – variation of properties in different units e.g. milk fat content variations that occur in different milk cans
Intra-unit variation – variation within individual units, e.g. fat content variation in one milk can.

ii) Reduce the sample size

Smaller sample sizes are easier to manage during analysis, which reduces the chances of sample contamination and cross-contamination.

iii) Prevent changes in the milk sample

Ensure the sample does not undergo physical, chemical, or enzymatic changes. Enzymatic changes should be countered by elimination or inactivation of the enzyme, which can be achieved through:

Adding chemical preservatives
Freezing the sample
Heat treating the sample

Fat oxidation is the most likely chemical change to occur on fat based dairy products such as cream, ghee (products with unsaturated fatty acids). To reduce oxidation, limit such samples to light exposure, elevated temperature, oxygen, or from co-oxidants.

Prevent microbial growth through freezing, cold storage, heat treatment, and use of preservatives.

Physical changes likely to occur may include loss or gain of moisture, which can affect butter analysis. Another serious physical change may also include crystallization (as in the case of ice cream samples). You will be able to control most of the physical changes by monitoring the storage temperature and relative humidity.

iv) Sample identification

Carefully label all the laboratory samples so that in case of any problem, it is easy to trace the origin. The information that is usually used for the identification of laboratory samples includes:

Sample description, i.e. what is it? (whether butter, cheese, ice cream, etc.)
Note the time of sampling
The location of sampling
Identity of the person who took the sample
Method used to select the sample
Record of the procedure used for each sample

Sampling Liquid Milk and Other Dairy Products

The objective of raw milk quality assessment is to ascertain the quality of the finished product. You have to make sure that you are starting off with the best quality raw milk so that you can assure the consumers that the quality of the finished product is the best.

Since you are dealing with large quantities of milk, it is necessary to take a representative sample from which you will be able to determine the quality of the entire batch. This is why you will need to do sampling of liquid milk.

Sampling from a single container

At the dairy processing plant, you will be receiving milk in single batches, either from individual farmers or from bulking stations. You will need to obtain samples from these single containers.

To do this, mix the milk mechanically and then draw the required quantity of sample. Label the sample container appropriately and follow the general sample handling procedures.

Composite sample from several containers

A composite sample is the quantity of milk obtained by mixing proportional parts of different milks. A sample is then picked from the composite sample to represent the whole lot.

Sampling from storage tanks and road tankers

Storage tanks usually have agitators used to mix the milk. Ensure the milk is mixed gently for about 15 minutes in the tank and in case there are no mechanical agitators, mix the milk mechanically using a stirrer and then pick the sample using a dipper.

Pick the samples proportionate to the size of the tank. The minimum amount of the sample should be at least 500 ml.

Most modern tanks and tankers have automatic samplers from which you obtain the sample after the contents have been uniformly mixed.

Sampling Different Dairy Products

a) Sampling cream

Cream is thicker than fluid milk; therefore, the tester should use a sampling tube with a wider diameter. It is important to ensure that the cream has been properly mixed before picking the samples.

Make sure to examine the samples soon after collection as any delay may lead to deterioration in the sample quality. This is due to the susceptibility of cream to enzymatic degradation given its high butter fat content. This could jeopardize the results.

b) Sampling evaporated milk

Put the unopened cans in a water bath at 60°C for two hours. You will need to remove the cans from the water bath every 20 minutes for a vigorous shaking.

After the two hours have elapsed, remove the cans and cool to room temperature. Remove the lid and mix the content with a spatula. At this point, you can get a composite sample from different cans.

Depending on the type of tests to be done, you can test the sample the way it is or you can mix with distilled water. The value obtained from the test is adjusted using a correction factor.

c) Sampling sweetened condensed milk

Temper the unopened cans in a water bath at 35°C for 30 minutes. After the time has elapsed, empty all the contents of the container while still warm and mix to a uniform consistency.

Take 100 grams of the sample and mix with 500 grams of distilled water in a water bath. Get a test sample from the diluted mixture for testing and correct using the dilution factor used.

d) Sampling dried milk

Avoid sampling in a high humidity environment. Take samples from different parts/regions of the mass to be tested using a tubular trier and then transfer the collected samples to a dry clean cont