Validation-of-cleaning-programs.pdf

Validation, as a food safety management requirement, has been around since the inception of HACCP and specifically for Critical Control Points (CCPs). In recent years, however, changes in global food safety standards and legislation have introduced the requirement to validate non-critical controls such as PRPs, oPRPs and general preventive controls. This white paper focuses on validation of cleaning programs. We also address the importance of cleaning, why validation is essential and the best practice methods of validation typically employed.

1.

Introduction 1.1 What is Cleaning? 1.2 What is Disinfection? 1.3 What is Validation? 1.4 Cleaning Validation Methodology 1.5 When is Cleaning Validation Required?

2.

Steps to Validate a Cleaning Procedure 2.1 Step 1: Establish Validation Team 2.2 Step 2: Define and Document Cleaning Validation Methodology 2.3 Step 3: Collect Scientific Data 2.3.1 Hazard and Risk Data 2.3.2 Customer Complaints 2.3.3 Legislation and Commercial Requirements (Allergens/Pathogens) 2.3.4 Food Recalls, Alerts, and Notifications 2.3.5 Research Reports 2.3.6 Hazard Datasheets 2.3.7 Cleaning Chemicals 2.3.8 Machinery 2.3.9 Cleaning Methods and Equipment 2.3.10 Analytical and Sampling Methods 2.3.11 Establishing Acceptable Limits 2.4 Step 4: Conduct Hazard Analysis 2.5 Step 5: Conduct Risk Assessment 2.6 Step 6: Categorization of Cleaning Programs 2.7 Step 7: Document, Approve and Use SSOP’s 2.8 Step 8: Define Cleaning Schedules and Verification 2.9 Step 9: Training 2.10 Step 10: Conduct Validation 2.10.1 Conduct Cleaning Runs 2.10.2 Conduct Tests and Collect Data 2.11 Step 11: Analyze Results 2.11.1 Capability Studies 2.11.2 Performing a Capability Study 2.12 Step 12: Document Validation & Conclusion 2.13 Step 13: Revalidation

3.

Cleaning Validation Excel Tool

Validation, as a food safety management requirement, has been around since the inception of HACCP and specifically for Critical Control Points (CCPs). In recent years, however, new versions of global food safety standards and legislation have introduced the requirement to validate noncritical controls such as PRPs, oPRPs and general preventive controls. In this white paper we will focus on validation of cleaning, which is a widely used and important preventive measure. We will address the importance of cleaning, why validation is essential and best practice methods of validation typically employed. Food safety management systems often place HACCP at the centre of control for specific food hazards. This makes sense, particularly where a specific hazard linked to adverse impacts for the consumer can be identified. However, no matter how robust the HACCP system may be, it requires a sound program of basic controls that addresses general hazards, many of which may not be identified. These programs are described as Good Manufacturing Practices (GMP), Prerequisite Programs (PRP) and Preventive Controls (PC) to name a few. Regardless of the term used, they all represent the basic requirement for good practice to provide a safe environment for the manufacturing of food. Among the most important of these is the need to clean and sanitize your plant and equipment sufficiently to produce food free of physical, allergenic, chemical and microbiological hazards. In addition, it is important to understand the reasons why a food plant must be cleaned. These include:



To reduce the risks from food hazards – food poisoning and foreign body contamination



To comply with local and international legislation



To meet specific customer requirements, e.g., Tesco



To meet the requirements of global food safety standards (GFSI)



To maintain positive audit and inspection outcomes



To allow maximum plant productivity



To present a hygienic visual image



To promote safe working conditions for staff, contractors and visitors



To maintain product shelf-life



To avoid pest infestation

At the most basic level, the visual appearance of a food factory is an indication of the standards and culture of the company. It has a strong impact on the perception of an auditor or visitor and can influence the overall outcome of audits and securing new business. For this reason, the visual cleanliness of a company is as important as detailed HACCP plans.

Cleaning is one of the most basic pre-requisites for producing safe food and is normally defined as a clear requirement in food safety standards and legislation. Cleaning is a physio-chemical process involving a number of factors, including soil, substrate, energy and reactions. In food processing operations, soils and deposits originate from the ingredients used in the prepa-

ration of the product. These soils include fats, oils, greases, proteins, carbohydrates, starches, limescale and corrosion. Not all the deposits mentioned pose a serious risk; however, all create a poor visual appearance. Some act as an excellent substrate for absorbing other soils and microorganisms. The key point is that each is chemically different and requires different cleaning methods. Therefore, it is important to identify the typical soils present and design your program accordingly. The substrate is the materials of construction found in food processing plants. The standard of materials can vary with its ease of cleaning and resistance to corrosion with chemicals. The ideal material standard is smooth, non-porous, abrasion resistant and inert, e.g., stainless steel. Energy is the core driver behind all cleaning processes. This energy is normally made up of a combination of factors including time, thermal energy, mechanical energy and chemical energy. It is the interaction of the above factors that determines the effectiveness and rate of the cleaning method. Often the aim is to obtain a balance consistent with cost, efficacy and food safety. This is where a knowledgeable chemical supplier can assist in the development of effective cleaning regimes. Chemical and physical reactions of cleaning are the processes by which the action of cleaning is achieved, taking into account the various factors of cleaning. These include wetting, penetration, emulsification, dispersion, solubilization, hydrolysis, saponification, chelation and oxidation. Detergents: The nature and complexity of the detergent employed depends on the variation of soils, water hardness, temperature of the method, plant surfaces and safety. Detergent suppliers normally have a range of detergents to be employed in varying and specific circumstances. The range of products will include:



Alkalis: caustic soda, caustic potash, coronate, silicate, phosphate



Acids: phosphoric, nitric, citric, glycolic



Chelates: EDTA, NTA, gluconate, glucoheptonate, citrate, polymeric



Solvents: isopropanol, propylene, butyl diglycol, ethers



Surfactants: anionic, cationic, non-ionic, amphoteric



Inhibitors: organic, inorganic



Enzymes: protease, lipase, amylase



Oxidising agents: hypochlorite, isocyanurates



Stabilisers



Viscosity modifiers

A detergent solution may contain between 2 and 15 components, blended carefully to specification. It is important to work with a good supplier to correctly identify the correct detergent for your operation. This will save money in the long term as cleaning will be more effective. The failure of a product to work is usually not due to a poor quality product but rather the wrong one. Application and use are also important factors and a good supplier will usually provide training in the correct use of the product. A ‘detergent’ is designed to remove soils. Another term used is ‘sanitizer’ and is often used to describe similar products. However, a ‘sanitizer’ is usually used to refer to a product containing both a detergent and disinfectant. A ‘disinfectant’ is a product which kills microbes without employing a soil removal action.

Soil deposits can harbour potentially harmful (pathogenic) microorganisms or allergens which if left to grow can present a serious risk to the health of the consumer. In order to control this risk the soil must first be removed using an effective cleaning method, normally including a detergent as previously discussed. Typically, the reduction achieved by cleaning is in the order of 3-4 logs per cm2. If the initial loading was 106 cm2, there will remain counts of 102-103 cm2 after cleaning. It is normally necessary to reduce the levels further to a few hundred, and this is where the process of disinfection is used. It should be noted that sterilization, which is the elimination of all microorganisms, is neither practical nor normally necessary in the disinfection of food plants. The group of chemicals known as disinfectants share many attributes with detergents but are different in terms of their function, which is to kill microorganisms that are left on the surface after cleaning. The biocidal effect varies depending on the active component used in the disinfectant. It can be achieved by affecting the integrity of the cell wall or by interfering with critical metabolic reactions inside the cell. Most disinfectants are oxidizing and will react with organic materials, including microorganisms. These particular disinfectants include chlorine, iodophors and peracetic acid. They are quick acting and broad spectrum. They are normally not stable in hot water and corrosive on a range of metals and other materials. Non oxidising disinfectants are typically based on quaternary ammonium compounds, which are a class of cationic surfactant, amphoterics, alcohols and aldehydes. They are usually heat stable, less corrosive and have a residual biocidal or biostatic effect. The kill effect required from a disinfectant can vary for each microorganism and, therefore, it should be carefully formulated to ensure it is effective. Some may be ineffective at low temperatures and unsuitable for a chill. Well-designed disinfectants may employ several different biocidal components including surfactants and chelates to support the killing action. Disinfectants should be chosen in conjunction with the supplier, taking into account the surface materials, soils and the specific microorganisms to be controlled. Other considerations include the ambient and solution temperature and the time required. Cleaning and disinfection may, in some cases, be combined into one operation using a sanitizer which has the action of both a detergent and a disinfectant. However, it is believed that the two stage approach is more consistent and effective than the single stage sanitizer approach. It is important that non scented chemicals are used in food operations, due to the risk of taint.

According to the Codex Alimentarius definition, validation is: Obtaining evidence that a control measure or combination of control measures, if properly implemented, is capable of controlling the hazard to a specified outcome

In the context of cleaning, validation is proof that the approved cleaning procedure is capable of producing safe food within the criteria set for specific hazards. In other words, if we say that a cleaning program conducted between a line change-over is required to remove a specific allergen prior to the next production batch, then validation is positive and scientific proof that the program is in fact capable of doing this.

It is different from verification, which checks that the cleaning program has in fact been completed and was done according to the approved procedure. Verification merely checks you are doing what you say you are doing while validation asks the question; is what you are doing capable of eliminating the hazard?

The process of validating any preventive control, including cleaning, should be logical and clear in terms of the required outcomes. For cleaning, the process should be scientific and based firmly on the identification of hazards and robust risk assessment. Remember, cleaning is a costly activity for any food business and the above approach will help in managing costs and focusing resources on the areas with the maximum impact. While there is no set way to conduct validation of cleaning for every business, there are reasonable and logical steps which can be taken to ensure the validation study is following best practice. The table on page 7 is a suggested process for conducting cleaning validation and can be adapted by any site to meet its specific requirements. In this model, there are 13 steps to follow when validating your internal cleaning and sanitizing activities.

Validation of cleaning programs is normally required when legislation or food safety standards you are operating under require it. For example, the BRC Standard for Food specifically requires validation of PRP’s, especially where there is a high risk to the consumer. Validation of cleaning procedures should also be considered where risk assessment has identified it as being critical to legality and food safety, e.g., between manufacturing of one product and another, for allergen or species cross-contamination. Validation is not normally necessary for non-critical cleaning, e.g., between batches of the same product or non-contact surfaces such as floors, walls and the outside of vessels.

The first step in the process is to establish your validation team. The purpose of the validation team is to take responsibility for the validation process and ensure that the preventive control (in this case cleaning) is effective and capable of achieving the desired outcome. Validation is a scientific process and, therefore, the validation team should be made up of individuals who have a scientific understanding and grounding in collecting, collating, and conducting a structured analysis and reporting. Given this, the existing HACCP or Food Safety Plan team are usually the logical choice for conducting validation. Under FSMA, the qualified individual may also be ideally suited to adopt the role of leading or conducting validation of controls. For smaller companies where resources are limited, consideration should be given to using external resources and specialists. The following table, on page 8, should be considered when establishing your team:

Figure 1:

Suggested process for conducting cleaning validation

Establish Validation Team

This is the team who will conduct the validation. It is usually made up of employees and external service providers and experts where applicable

Define and Document Cleaning Validation Methodology

This is the documented procedure which clearly defines how the validation of cleaning will be conducted, by whom etc.

Collect Scientific Data

This is the process of collecting the various data and documents required to complete a validation of cleaning procedures, e.g., hazards, risks, chemicals, machinery, equipment, sampling methods, etc.

Conduct Hazard Analysis

This is the process of clearly identifying the specific hazards associated with the product, process and cleaning

Conduct Risk Assessment

This is the process of determining the significance of the specific hazards and cleaning procedures in the operation

Categorize Cleaning Programs

This is the process of grouping cleaning procedures which are similar in terms of activity, hazards and risks in order to minimise the amount of validation work required

Document, Approve and Use SSOP’s

In this step, the validation team will clearly document the actual cleaning procedures in use in the operation

Define Cleaning Schedules & Verification

This is the process of defining the frequency of cleaning to be conducted and how the efficacy of the cleaning will be assessed post cleaning (verification)

Training

This is the requirement that all employees engaged in cleaning activities are fully trained and competent in applying the documented and approved cleaning procedures

Conduct Validation

This is the activity of conducting cleaning runs, collecting samples, testing them and recording the results

Analyze Results

This involves conducting simple and statistical analysis of the collected and collated data

Document Validation & Conclusion

This is the action of drawing conclusions regarding the validity of cleaning programs and documenting them.

Revalidation

This is the process of redesigning cleaning programs in advance of conducting additional validation or validating existing procedures for a defined period of time to ensure the procedure remains valid

Figure 2:

The differing roles associated with validation teams

The next step is to define and document your cleaning validation methodology. This defines how as a company you validate cleaning to ensure it is capable of meeting the desired outcomes. You can use this whitepaper as a guide for preparing your own internal procedure. Every company is different, so the validation mythology should be adapted to suit local needs. This procedure should be an approved document within your general food safety management systems and should be drafted by a responsible member of the management team. The following is a suggested list of areas which should be covered in the procedure:



Objectives & responsible people



Cleaning SOP’s



Cleaning chemicals, concentration, solution volume, water quality



Time and temperature



Flow rate, pressure and rinsing



Number of cleaning cycles



Description of the equipment - make, model, complexity of design



Training of operators



Equipment used for monitoring (e.g., conductivity meters, pH meters)



Sampling procedures (e.g., direct sampling, rinse sampling, in process monitoring and sampling locations) and the rationale for their use



Analytical methods



Acceptance criteria (with the rationale for setting the specific limits)



Revalidation requirements

Because validation is a scientific process, it is essential that the team undertake an exercise to collect all relevant and available data to support the analysis and decision making. Much of this data will already be available as part of developing the existing food safety plan. Where relevant data is not at hand it should be sourced and retained for review by the HACCP team. Data which is typically required to conduct validation of cleaning includes:



Hazard and risk data



Legislation and commerical requirements



Cleaning chemical



Machinery



Cleaning Methods and Equipment



Analytical and sampling methods

As with all studies conducted under food safety management, the first step is to conduct a hazard analysis. Before you can do this, data on hazards is essential, since without this it is impossible to determine risk and the effectiveness of control measures. Data on hazards can be obtained from a variety of sources including: 

Customer complaints



Legislation (Allergens / pathogens)



Food recalls, alerts and notifications (e.g. RASFF / FDA)



Research reports



Hazard datasheets (FDA Bad Bug Book)

This data should be gathered by the team and collated for analysis later in the validation process. Care should be taken to obtain the best quality data possible. Where data is limited and only poor quality or no data is available, this should be noted in the validation report and accounted for in determining the risk.

While customer complaints are usually unwelcome for most food companies, they can be a valuable source of hazard and risk data associated with your products and processes. Collecting and collating this data can permit trending the numbers and causes, which can inform a business of these hazards.

Very often the objective of cleaning, in addition to producing safe food, can be the production of legal food products. Legislation and commercial standards such as the GFSI can often prescribe

what this should be and, therefore, it is important to define what the legal and commercial framework is for cleaning. Generally speaking, the target hazards arising from this will include:



Allergens



Pathogens



Species

These hazards arise from a lack of control and/or cross contamination risks from poor cleaning. For example, food safety legislation will define a list of allergens which need to be controlled and declared if present in the food product. Similarly, criteria regarding the presence of specific pathogens in food are legally controlled. Recent food scandals such as ‘horse-gate’ have also led to tighter legal and commercial controls on specific contamination in meat products.

When conducting any hazard analysis and specific hazard identification, a good source of data are the online public registrars of food recalls, alerts and outbreaks maintained by regulatory agencies such as the EU and the FDA. This provides data of real and up-to-date food safety and legal issues relating to food products and commodities. This data can provide a reliable input into your risk identification exercise and ensure good validation of your cleaning programs. The following are links to two of these sources: European Union RASFF website: http://ec.europa.eu/food/safety/rasff/index_en.htm US FDA Recalls, Market Withdrawals, & Safety Alerts website: http://www.fda.gov/Safety/Recalls/

Various sources regularly publish scientific data on hazards and food safety risks and these can form a reliable input into your hazard analysis step. These usually come in the form of research reports, opinions and papers. While these can be very scientific in nature and time consuming to source and read, they can support validation of your food safety system in general. For example, the European Union’s food safety agency (EFSA) publishes reports and opinions on food safety hazards and risks. These can be located at https://www.efsa.europa.eu/en/publications.

A number of reputable sources produce what are known as hazard datasheets. These are concise documents providing excellent and accessible data on hazards, including their character, source, infectious dose, disease symptoms, occurrence and typical controls. These can be used as valuable reference material and validation of hazard identification. A good example is the FDA’s Bad Bug Book which can be found here: http://www.fda.gov/Food/FoodborneIllnessContaminants/CausesOfIllnessBadBugBook/

Data on the chemicals currently being used to perform cleaning within the operations must also be sourced and collected. This data is normally obtained from the chemical supplier in the form of

datasheets. The purpose of collecting data on cleaning chemicals is mainly to determine their suitability for cleaning. Upon review the following should be considered:



Solubility of the materials to be removed



Design and construction of the equipment and surface materials to be cleaned



Minimum temperature and volume of cleaning agent and rinse solution



Manufacturer's recommendations



Released by quality control and meets food standards or regulations



Composition known



Easily removed with rinsing - demonstrated - with acceptable limits defined



If residues persist - avoid



Also, consider detergent breakdown

Another essential area where data needs to be collected is relating to the machinery being used in the operation. Consideration should be given to the following:



Cleaning of contact surfaces to be validated



Critical areas should be identified



Dedicated equipment for: 

products which are difficult to clean



equipment which is difficult to clean



products with a high safety risk



The design of equipment may influence the effectiveness of the cleaning process



Which are the critical areas for sampling?



What would be considered an appropriate approach for cleaning validation for this piece of equipment?



What is important about cleaning validation for components / parts of equipment?



Also, consider the different materials, e.g., stainless steel contact surfaces, silicon seals and others

Data should also be gathered relating to the cleaning methods and equipment used. This will assist in determining the appropriateness of the methods employed. There are a number of methods which can be used to apply detergents and disinfectants, such as manual cleaning using cloths, mops, brushes, pads, etc. Manual cleaning is normally used in small areas, equipment that is nonwaterproof or requires dismantling or areas which are difficult to clean by other methods. It is a labor intensive method and may limit the use of certain chemicals for safety reasons. To ensure cleaning is effective, the method must be clearly defined and staff trained to an appropriate level. Foam cleaning is the common method for cleaning most food operations. A foam blanket, created using a wide range of available equipment is projected from a nozzle and allowed time to act on the soil. It is then rinsed off with the released deposits. Large areas such as floors, walls, conveyors, tables and well-designed production equipment are ideal for foam cleaning. Foam is a carrier for the detergent. The foam should be applied in an even layer. Coverage rates are quick and chemical usage is economical. Your chemical supplier will advise on the most appropriate chemicals and equipment for your operation. The equipment itself may be mobile, centralized or satellite. Spray cleaning uses a lance on a pressure washer with chemical induction by Venturi. This method can be wasteful of chemicals and can be slow to produce a foam. It should be used where foaming properties are not essential for the cleaning action. Aerial fogging uses compressed air or other equipment to generate a fine mist of disinfectant solution which hangs in the air long enough to disinfect airborne organisms. It will also settle on surfaces to produce a bactericidal effect. The system can come in a small portable device or built-in automatic central systems. Fogging should never be used as a primary sanitizing method. It should be used in conjunction with other methods. It is also important to ensure that coverage and saturation is sufficient and the mist is fine to allow proper action. Machine washing is normally an automatic or semi-automatic washing process conducted within a purpose built machine. There are many machine designs depending on the application, e.g., crate washing or utensil washing. They represent a significant capital investment and need to have a clear business case before purchasing. They tend to consume a large amount of chemicals and water. Failure to maintain them correctly can lead to a contamination risk to the product. Chemicals used in these machines should be low foaming. An effective system for controlling the dose of chemical should be employed and temperature control systems should be used where critical. CIP or cleaning in place is used extensively for the interior cleaning of pipes, vessels, tankers, heat exchanges, fillers and other enclosed process systems. CIP involves a programed cycle including timed pre-rinse, cleaning and rinsing stages and can be fully automatic or semi-automatic with a system of valves, pumps and detergent tanks controlled by a microprocessor. There are a number of parameters that need to be specified and controlled for effective CIP.

The quality of the validation conducted will depend heavily on the sampling and test methods employed. Here we are seeking to identify and select the most appropriate method for the validation and this starts with research and data collection.

Sampling There are two main methods of sampling:



Direct surface sampling and



Rinse sampling

A combination of the two is the most desirable. Direct Surface Sampling (Direct method) This is the most commonly used method and typically involves the use of “swabs”. The type of sampling material should not interfere with the test. Factors to be considered include the supplier of the swab, area swabbed, number of swabs used, whether they are wet or dry swabs, swab handling and swabbing technique. In addition, the location from which the sample is taken should be considered, including worst case locations. Critical areas and those hardest to clean should receive special attention, e.g., in semi-automatic/fully automatic clean-in-place systems. The composition of the equipment (e.g., glass or steel) may also impact the sampling method. Ensure you select the appropriate sampling medium and solvent. Rinse Samples (Indirect method) Rinse sampling is an indirect method and allows sampling of a large surface and areas that are inaccessible or that cannot be routinely disassembled, e.g., enclosed pipework systems. This method provides an "overall picture" and is useful for checking for residues of cleaning agents. This method can be used in combination with other sampling methods, such as surface sampling. Analytical Methods There are a wide variety of analytical methods available for testing your samples. Again, it is important to research the appropriate one for your needs. It is important that you use validated analytical methods which are able to detect residuals or contaminants specific for the substance(s) being assayed and at an appropriate level of cleanliness (sensitivity). Sensitive and specific methods may include ELISA, Specific Allergen Testing and Non Specific Testing. Validation of the analytical method should include the limit of detection (LOD) and reproducibility.

When establishing acceptable limits for your testing, you should ensure they are practical, achievable and verifiable. Your rationale for establishing them should be logical and based on knowledge of materials. Each situation should be assessed individually. Ideally, the limits should be measurable in the form of a control chart, where your actual results (or sub-group of results) can be plotted and viewed against the established limits. Figure 3:

Control chart

When setting limits you should be product-specific, group products into families and choose a worst case product or group products according to risk. Limits may be expressed as a concentration in a subsequent product (ppm), limit per surface area (cfu / cm2), in rinse water as ppm or visual. Certain allergenic ingredients and highly potent material should be undetectable by the best available analytical methods.

The purpose of conducting a hazard analysis is to clearly and specifically identify and list the hazards associated with your products and processes arising from ineffective or poor cleaning activities. Typically, this can be an output from your HACCP or PCP study and may not need to be conducted separately for the cleaning validation activity. However, if required, this hazard analysis should be reviewed to ensure it is comprehensive and specifically addresses cleaning related hazards.

As with HACCP, hazards may be identified under the general categories of Biological, Physical and Chemical. You should also review the data collected in the previous step to ensure all hazards are being included in the analysis. It is usual for companies to prepare a list of these hazards, including the basis for their identification. It is important that you are as specific as possible. For example, for biological hazards it is better to identify a specific pathogen, e.g., E. Coli O157 rather than “bacteria”. The reason for this is that pathogens have different characters and often require different controls to address them. Similarly you should try and identify a specific allergen and not general allergens as a hazard. Hazards may include specific hazards under the following categories.

Figure 4:

Categorization of hazards

The next step is to generate a list of cleaning steps, procedures, activities and programs, and identify which of these steps may be associated with a specific hazard(s). Again much like the approach you take in HACCP, you are looking to determine the potential hazards relating to a cleaning program or activity.

Once all potential hazards have been identified for cleaning activities, the next step is to determine the significance of these hazards. Not all hazards present enough of a risk to warrant a full control program or indeed validation. For this reason, risk assessment is used to separate these out. In regard to cleaning, risk assessment should seek to determine which cleaning programs pose the greatest risk should failure occur in the activity. The company can then use the outputs of the risk assessment to direct cleaning resources and verification activities on the high risk activities. Under the BRC Standard for Food, 4.11 states “The frequency and methods of cleaning shall be based on risk”. This implies that it is a requirement to conduct a documented risk assessment on all cleaning programs. The general procedure for conducting a risk assessment on cleaning programs is as follows: 1.

List all cleaning programs and describe each and the associated hazards

2.

Determine risk rating criteria and decision outcomes

3.

Rate each program

4.

Probability of failure to conduct cleaning correctly

5.

Severity of impact if failure occurs

The best approach to get started is to use a standard risk assessment model or matrix. Figure 5:

An example of a risk assessment matrix

In this example, a simple 3 x 3 matrix is used for estimating the probability of occurrence and severity of impact. In this model, you need to remember what you are assessing in terms of probability is the likelihood of the cleaning program or activity failing or being ineffective. In this regard, it is more like the Failure Mode Effect Analysis and not the approach usually adopted in HACCP. After rating each step, point or cleaning activity, both values are multiplied together to provide a risk estimate of high, medium or low. Prior to conducting the risk assessment, it is necessary to define the criteria for each rating number under severity and probability. The following provides a suggestion of how these ratings may be characterized. For example, a high risk cleaning program may be characterized as one where failure to conduct the cleaning activity effectively is likely and the impacts would be severe. Figure 6:

Characterization of criteria for probability and severity

Once each cleaning program or activity has been risk assessed, the outcome, action or control which should be applied should be clearly defined. For example, if the risk assessment determines a high risk rating, then you may consider performing this cleaning activity at a higher frequency and with full verification post cleaning. For lower risk rated activities, the outcome may be a lower or minimal cleaning frequency with just visual verification. This approach provides a clear and structured approach to develop cleaning programs and the logic used.

Figure 7:

Definition of outcome decision criteria

All this can be recorded in a master risk assessment document as follows: Figure 8:

Master risk assessment document

The most important outcome of risk assessment is the power to identify which cleaning programs and activities are critical for producing safe and legal food. For example, risk assessment can identify which cleaning programs are crucial for preventing allergen and species cross contamination between products. In particular, it can identify cleaning programs on production lines and contact surfaces between allergen and species change overs. Once identified, criteria can be established for high risk programs including:



Clear and detailed cleaning procedures for line change over between species



Visual validation of specific contact surfaces



ATP and allergen testing positive release



Clear re-clean criteria

Every food business has limited resources. For this reason it is important to use risk assessment to focus those resources on value-adding tasks. Risk assessment is excellent for assisting in this. Vali-

dation of cleaning can use a lot of time and resources. By risk assessing cleaning programs you can categorize them into groups and, thus, avoid repeating validations on similar programs. Another option is to take the “worst case” scenario for cleaning programs or choosing a representative product which is most difficult to clean after production. Validation can also be conducted on similar equipment or the same equipment in different sizes (e.g., 300 l, 500 l and 1000 l tanks).

Cleaning is a complex process. To ensure it is conducted correctly, a defined and systematic approach is required that takes into account a number of factors previously covered. This approach takes the form of a Standard Sanitizing Operating Procedure (SSOP), and this is usually a legal requirement in addition to a fundamental requirement of global food standards. A collection of these cleaning procedures forms a Cleaning Plan or Program which is plant specific. A typical cleaning procedure includes the following:



Cleaning method



Standards



Frequency



Chemicals used



Equipment used



Time and temperature specifications

Figure 9:

Cleaning procedure example

These procedures may be collected into a cleaning manual, which should be available to those responsible for cleaning. The cleaning plan may also be summarized in a table and records should be maintained of all completed cleaning activities. As previously discussed, cleaning is a significant cost for food businesses. This may contribute to cleaners and managers combining or omitting individual steps in the cleaning procedures. This should be avoided. Training of staff and commitment by management are essential to prevent this. The correct sequence of a general cleaning procedure for surfaces in a food plant is:



Gross Clean/Preparation



Pre-rinse



Detergent application



Post-rinse



Disinfect



Terminal rinse

In terms of validation of the cleaning, it is an essential requirement that the procedure to be validated is clearly documented as it actually exists. At this point, you are not concerned with whether it meets specific requirements. This should be done on the job and in conjunction with those who perform the cleaning. Practically, the person documenting the procedure should observe how the cleaning is conducted in real time, document the steps and confirm with the relevant cleaning staff that the procedure reflects what is normally conducted. It may be necessary to observe and document the procedure against different members of the cleaning team in case there is operator variation. Other details such as times, temperatures, pressures, equipment, chemicals and PPE used should be defined and documented. At this point, the procedure is ready to be formally approved and issued for use. Remember, document the cleaning procedure as it actually exists. At this point, you are not concerned with whether it meets specific requirements or is valid. Your main objective is to define the procedure as it is for now.

Once the cleaning procedure has been documented and approved, the next step is to define how often the procedure will be conducted. This will normally be done based on the output of the risk assessment step. It is also important to identify the verification required for the program, e.g., visual, ATP, chemical testing. This will include the standard to be achieved and specific sampling points based on the risk assessment.

Now that a standardized cleaning program has been approved for use, it is important that all cleaning personnel who will be conducting the cleaning are trained in completing it. This is critical to remove the impact of variation between operators since validation will be based on data collected over a period of time and for multiple cleaning cycles. Training of employees should be conducted against the approved procedure. Full records of all training should be retained as part of the validation report.

Validation of cleaning SSOP is now possible. There are a number of approaches to doing this depending on the risk of the operation, resources available and the underlining motivation for conducting the validation, e.g., customer requirements, certification or legislation. In this example, we will be using a structured statistical model to determine the capability of the SSOP to achieve the desired outcome. This may not always be possible, however, you should endeavour to conduct as full and statistically valid an exercise as possible. In some cases, a company may base its validation on single point data, however, this should be clearly highlighted in the validation report.

To assist you, this whitepaper is accompanied with an Excel Tool for you to capture your data and report on it. The following is a recommended procedure for conducting the validation. It is based on actual cleaning cycles in your operation from which data is gathered and analyzed.

A Cleaning Run is an identified cleaning SSOP which you wish to validate. Only SSOP’s which have been approved and which you have defined clear criteria and trained staff on should be validated. Conduct the cleaning SSOP exactly as documented. Do this over a defined period of time and/or number of cleaning runs, e.g. 3-4, weeks / 10 cleaning activities. Make sure you define this in the cleaning report in advance. The number of cleaning runs should be sufficient to produce statistically valid data.

Following each cleaning run, conduct microbiological, allergen or other tests as required on food contact surfaces and other sample points. Ensure the test method is defined in advance and repeated for each test. The person conducting the test should also be trained and competent to conduct the test. All data should be recorded accurately and retained for analysis later on.

The next step is to analyze the data. Collate the results obtained over the validation period and conduct appropriate statistical analyses to determine the variability in the efficacy of the cleaning SOP’s. This is known as a capability study and it is the best way to determine if a cleaning procedure is actually capable of achieving the designed results over time.

The purpose of a capability study is to estimate the process average and variation, relative to the specified tolerances for the standard cleaning procedure. This is the best way to validate the cleaning programs as it is based on the actual variation of the procedure as opposed to one test result. This is a simple statistical exercise and can be completed using the Excel tool provided with this whitepaper. The following details how to use the capability study tool.

The first step is to identify the essential data and conditions relating to the cleaning activity to be

validated. Capability studies are frequently performed using a number of sample points for a given post-cleaning sample analysis. If the samples for the study are not consecutive, note the times, groupings, and conditions under which the units were produced. Measure the samples in time-order-produced if possible, to permit deeper investigation if desired. In the tool, enter the criteria maximum and minimum (if applicable) on the form where shown, e.g., allergen ppm, cfu, rlu, etc. Characteristics with only a minimum criteria are usually "higher is better" characteristics while characteristics with only a maximum criteria are usually "lower is better" characteristics. Characteristics with both a maximum and minimum criteria are usually "nominal is best". For micro testing where “lower is better” normally applies, you should enter a minimum also to create a 'normal' process. As you collect the data, it should be entered. Typically a data sample size of about 30 is adequate. The tool will automatically construct a histogram to visually estimate the shape of the process distribution, relative to the tolerance limits. From this, a Cp ratio will be calculated, which is the ratio of the tolerance width to the width of 6-sigma of process variation. A Cp of less than 1.33 signals trouble staying within the tolerance. A Cp greater than 2.0 signals a good potential to meet the tolerances. The Cp ratio is independent of the location of the process average. Processes with off-target averages can still score good Cp ratios. If the characteristic has only a one-sided tolerance, the Cp ratio does not apply. The Cpk, is the distance from process average to closest tolerance, divided by 3-sigma. Both Cp and Cpk are estimates based on random sampling, and are subject to variation from one sample to another sample. Histogram distributions with unexpected shapes deserve investigation, e.g., bimodal, having two means. Process capability estimates are misleading if the process is subject to significant special causes (changes in average or variation). For this reason, it is important that variables, such as the employee training, cleaning conditions, temperatures, concentrations, equipment, etc., are maintained as close to constant as possible. The documented and standardized procedure for cleaning should only be used and visually monitored to ensure it was conducted to standard.

Figure 10:

Capability study

Once analysis of the results is completed, a conclusion can be made regarding the validity of the cleaning method. The following should be considered and recorded as part of this step:



Document the data in a validation report



Make a clear presentation of the scientific basis, materials, methodology, results and findings



If results indicate that the SSOPs are capable of consistently delivering results that comply with the established criteria during the period, then the cleaning SOP’s can be considered effective and validated.



If results indicate that the SOPs are incapable of consistently delivering results that comply with the established criteria during the period, then the cleaning SOP’s can be considered ineffective and not validated.

Following validation and conclusion, it may be necessary to conduct another validation to ensure any modified procedure is valid. In this case the following is best practice:

Figure 11:

Revalidation best practice

In the case where a procedure has already been validated it is also good practice to repeat the validation at a pre-defined frequency e.g. two-years.

An Excel workbook tool has been provided with this whitepaper to assist you in applying the best practice covered in the document. The tool contains the following: Figure 12:

Cleaning validation tool contents

Figure 13:

Cleaning run report

Figure 14:

Cleaning validation

              