Application of Failure Mode and Effect Analysis (FMEA), Cause and Effect Analysis, and Pareto...

14
Original article Application of failure mode and effect analysis (FMEA) and cause and effect analysis for industrial processing of common octopus (Octopus vulgaris) – Part II Ioannis S. Arvanitoyannis 1 * & Theodoros H. Varzakas 2 1 Department of Agriculture, Animal Production and Aquatic Environment, School of Agricultural Sciences, University of Thessaly, Fytoko Street, 38446 Nea Ionia Magnesias, Volos, Hellas, Greece 2 Department of Processing of Agricultural Products, School of Agricultural Sciences, Technological Educational Institute of Kalamata, Hellas, Greece (Received 23 April 2007; Accepted in revised form 21 May 2007) Summary Failure mode and effect analysis (FMEA) model was applied in conjunction with cause-and-effect analysis for the risk assessment of octopus processing (Octopus vulgaris). Critical control points were identified and implemented in the cause-and-effect diagram (also known as Ishikawa, tree diagram and fishbone diagram). The main emphasis was put on the quantification of risk assessment by determining the risk priority numbers (RPN) per identified processing hazard. Chemically contaminated product, decomposed raw materials, scombrotoxin presence in the final product, incorrectly labelled product, storage in cans (foreign matter) and defective products, were identified as those with the highest RPN (378, 294, 280, 252, 245 and 144 respectively) and corrective actions were undertaken. Following the application of corrective actions, a second calculation of RPN values was carried out, leading to considerably lower values (below the upper acceptable limit of 130). It is concluded that the incorporation of FMEA analysis within the ISO2200 system of an octopus-processing industry is imperative. Keywords Cause-and-effect diagrams, FMEA analysis, octopus processing. Introduction The quality, including safety, of products has tradition- ally been controlled by inspection of the final product. This is the case with food as well as with other products. Veterinary food (meat) inspection began almost 150 years ago, after veterinary pathologists and para- sitologists had found ways to identify human pathogens in food animals. Meat inspection is still a large-scale function of veterinary medicine (http://nsgl.gso.uri.edu/ flsgp/flsgph01002.pdf). Current meat inspection is based on 100% inspection of the live animals, their carcasses and organs, ante mortem and postmortem. This system has not changed significantly over the last 100 years. The problem is that 100% inspection does not mean 100% detection of human pathogens, and this has led to a search for other ways to ensure food safety (FAO, 2001). The inefficiency of the present inspection system is associated with changes in the ways meat (food) is produced, processed, marketed and prepared when compared with that done 100 years ago (http://www.codexalimentarius.net/ download/report/633/al28_18e.pdf). Hazard Analysis Critical Control Point (HACCP) was developed in the 1960s by the US food industry and National Aeronautics and Space Administration (NASA) as a ‘zero-defect’ approach for feeding astro- nauts. The bases of HACCP are that it is a process control rather than a product control and that it focuses control on steps in the processing system that are critical to consumer health (Arvanitoyannis & Tzouros, 2006). HACCP has won wide acceptance as a voluntary control programme in the food industry, and is now a prescribed part of meat and poultry slaughter in the US, as well as seafood and juices. There is also a growing interest in using HACCP to control the safety of live animal production pre-harvest food safety (http://www.codex alimentarius.net/download/report/633/al28_18e.pdf). There can hardly be HACCP without Good Manu- facturing or Management Practices (GMP). Briefly, GMP is a description of all the steps (which should represent good practice) in a processing facility, while HACCP is a documentation that the steps important to *Correspondent: Fax: +302421093144; e-mail: [email protected] International Journal of Food Science and Technology 2009, 44, 79–92 79 doi:10.1111/j.1365-2621.2007.01640.x Ó 2008 The Authors. Journal compilation Ó 2008 Institute of Food Science and Technology

Transcript of Application of Failure Mode and Effect Analysis (FMEA), Cause and Effect Analysis, and Pareto...

Original article

Application of failure mode and effect analysis (FMEA) and cause

and effect analysis for industrial processing of common octopus

(Octopus vulgaris) – Part II

Ioannis S. Arvanitoyannis1* & Theodoros H. Varzakas2

1 Department of Agriculture, Animal Production and Aquatic Environment, School of Agricultural Sciences, University of Thessaly, Fytoko

Street, 38446 Nea Ionia Magnesias, Volos, Hellas, Greece

2 Department of Processing of Agricultural Products, School of Agricultural Sciences, Technological Educational Institute of Kalamata, Hellas,

Greece

(Received 23 April 2007; Accepted in revised form 21 May 2007)

Summary Failure mode and effect analysis (FMEA) model was applied in conjunction with cause-and-effect analysis

for the risk assessment of octopus processing (Octopus vulgaris). Critical control points were identified and

implemented in the cause-and-effect diagram (also known as Ishikawa, tree diagram and fishbone diagram).

The main emphasis was put on the quantification of risk assessment by determining the risk priority numbers

(RPN) per identified processing hazard. Chemically contaminated product, decomposed raw materials,

scombrotoxin presence in the final product, incorrectly labelled product, storage in cans (foreign matter) and

defective products, were identified as those with the highest RPN (378, 294, 280, 252, 245 and 144

respectively) and corrective actions were undertaken. Following the application of corrective actions, a

second calculation of RPN values was carried out, leading to considerably lower values (below the upper

acceptable limit of 130). It is concluded that the incorporation of FMEA analysis within the ISO2200 system

of an octopus-processing industry is imperative.

Keywords Cause-and-effect diagrams, FMEA analysis, octopus processing.

Introduction

The quality, including safety, of products has tradition-ally been controlled by inspection of the final product.This is the case with food as well as with other products.Veterinary food (meat) inspection began almost150 years ago, after veterinary pathologists and para-sitologists had found ways to identify human pathogensin food animals. Meat inspection is still a large-scalefunction of veterinary medicine (http://nsgl.gso.uri.edu/flsgp/flsgph01002.pdf).Current meat inspection is based on 100% inspection

of the live animals, their carcasses and organs, antemortem and postmortem. This system has not changedsignificantly over the last 100 years. The problem is that100% inspection does not mean 100% detection ofhuman pathogens, and this has led to a search for otherways to ensure food safety (FAO, 2001). The inefficiencyof the present inspection system is associated withchanges in the ways meat (food) is produced, processed,

marketed and prepared when compared with thatdone 100 years ago (http://www.codexalimentarius.net/download/report/633/al28_18e.pdf).Hazard Analysis Critical Control Point (HACCP)

was developed in the 1960s by the US food industryand National Aeronautics and Space Administration(NASA) as a ‘zero-defect’ approach for feeding astro-nauts. The bases of HACCP are that it is a processcontrol rather than a product control and that it focusescontrol on steps in the processing system that are criticalto consumer health (Arvanitoyannis & Tzouros, 2006).HACCP has won wide acceptance as a voluntary controlprogramme in the food industry, and is now a prescribedpart of meat and poultry slaughter in the US, as well asseafood and juices. There is also a growing interest inusing HACCP to control the safety of live animalproduction ⁄pre-harvest food safety (http://www.codexalimentarius.net/download/report/633/al28_18e.pdf).There can hardly be HACCP without Good Manu-

facturing or Management Practices (GMP). Briefly,GMP is a description of all the steps (which shouldrepresent good practice) in a processing facility, whileHACCP is a documentation that the steps important to

*Correspondent: Fax: +302421093144;

e-mail: [email protected]

International Journal of Food Science and Technology 2009, 44, 79–92 79

doi:10.1111/j.1365-2621.2007.01640.x

� 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology

Table 1 Failure mode and effect analysis – basic table – CCP 1

FMEA table of hazardous processing methods

Defective products

Estimated

corrective

actions result

Failure mode Consequences S Failure causes O Ways of control D RPN

Corrective

actions S O D RPN

Can come

from a

possible

malfunction

of the filler

Due to the high

technique control

this case could be

impossible for the

most of the

products. So that it

is not a noticeable

point

2 1) Unsuitable control

equipment

2) Poor maintenance

3) Insufficient

personnel skills

4) Insufficient use

5) Old machinery

(Kreuzer, 1984)

5 1) More funding,

better management

2) Frequent

Validation

3) Completed

staff training

4) Completed staff

training, more

funding

5) More funding,

Better management

6 60 Not required 2 5 6 60

Can come

from the low

quality of

raw

materials

Due to the low

ingredient quantity

present, even if

there are unsuitable

raw materials 1%, it

allows certification

towards to

customers that the

final product have

been repeatedly

controlled.

3 1) Men’s hygiene

2) Microorganism

and bacteria from

the toxic area

3) Polluted water for

washing

4) Insufficient

conditions of

storage of raw

materials

5) High temperature

(Vaz-Pires et al., 2004)

7 1) Extra hygiene

measures and staff

training

2) Extra hygiene measures

3) Water control,

pumping potable

water- non from

industrial area

4) Keeping of raw

materials in cool

and shady place

5) Keeping low

temperature by

using isothermal

packing

4 84 Not required 3 7 4 84

Can come

from

insufficient

personnel

skills

Even if the case of

Insufficient

personnel skills it is

not remarkable from

the appearance of

the final product by

the customer, the

most of products

could not pass all

control tests

6 1) No staff

cooperation

2) Incomplete staff

training

3) No senior

management

commitment ⁄cooperation

4) Unreliable supplier

8 1) Extra funding

2) Extra funding

3) Completed

management

training

4) Extra control

measures

3 144* Extra control

measures,

Extra Funding,

Better

management.

GMP and GLP

application.

3 6 2 36

Can come

from the

low quality

of the

packaging

materials

Important effects by

heavy metals

presence in ingredient

higher than the critical

limit. Possible

obligation for special

labelling for

non-permitted CL of

heavy metals. All the

customers are

affected. Possible

trade failure of

product

8 1) Contaminated raw

material over transfer

2) Low-purity reference

materials, insufficient

inspection

3) Bad maintenance of

the canned products

4) Insufficient ⁄ low

sterilisation

5) Low quality of

methods

3 1) Extra control

measures

2) Better

management, extra

control measure,

completed training

3) Keeping low

temperature and use

of ear driers to avoid

the moisture low.

4) Completed training

5) Extra funding

6 144* Extra control

measures, extra

funding, better

management

6 2 5 60

*When RPN is over 130, corrective action is required.

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas80

International Journal of Food Science and Technology 2009 � 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology

consumer health are under control. Proposed GMPs forfood production are often called Good AgriculturalPractices (GAPs); at this point, GAPs are advisory,rather than mandatory. Sanitation standard operatingprocedures (SSOPs) are also a needed (and oftenrequired) prerequisite to HACCP (http://www.afdo.org/afdo/upload/Sushi.pdf; Miliou et al., 2005).

Using FMEAs to assess risk

Describe the product or process

A clear and specific description of the product or processundergoing failure modes and effects analyses (FMEA)must first be articulated. The creation of this descriptionensures that the responsible engineer fully understandsthe ‘form, fit and function’ of the product or process(Kececioglu & Dimitri, 1991).

Draw a block diagram of the product or process

A block diagram of the product ⁄process needs to bedeveloped to show the logical relationships betweenthe components of the product or the steps ⁄ stages ofthe process. A block diagram may be in the form ofboxes connected by lines, with each box correspondingto a major component of the product or a major step

of the process. The lines correspond to how theproduct components or process steps are relatedto each other (http://www.npd-solutions.com/pdforum.html).

Complete the header of the FMEA table

Failure modes and effects analyses table (Table 1)headers vary from one to the next, as they are supposedto be customised according to the requirements of thecompanies using them. Generally the header requires,among others that you may wish to add, the followinginformation: Product ⁄Process ⁄System Name, Compo-nent ⁄Step Name; Product Designer or Process Engineer,Name of the Person who prepared the FMEA form;FMEA Date; Revision Level (letter or number);and Revision Date (http://www.npd-solutions.com/pdforum.html).

Enumerate the items (components, functions, steps, etc.)that make up the product or process

Break down the product or process being subjected toFMEA into its major components or steps. List downeach of these components or steps in column 1 of theFMEA table. The items must be listed down in a logicalmanner (http://www.fmeainfocentre.com/).

DefectiveProducts

Malfunction of Filler

Low Quality ofPackaging materials

Law Quality of Raw Material

Personnel

Insufficient Personnelskills

Insufficient Use

Old Machinery

Insufficient Conditions of Storageof Raw Materials

Polluted Waterfor Washing

Unsuitable ControlEquipment

Law Funding

Low Funding

Low Funding

Low Funding

Low Funding

Low Funding

Low Funding

Badmanagement

Badmanagement

Bad management

Infrequent Validation

Poor Maintenance

Incomplete Staff Training

Incomplete Staff Training

Incomplete management Training

Incomplete Staff Training

Raw MaterialContaminated over Transfer

Unreliable supplier Unreliable suppliers

Unreliable supplier

Insufficient inspection

Inadequate traning

Inadequate traning

High temperature

Incomplete training procedures

Low Sterilization

Low Quality of Metals

Bad Maintenance ofCanned Products

No Staff Cooperation

Inadequate quality controlNo Senior Management

commitment/cooperationDiscontinuingelectricity flow

High temperature

Industrial district

Low Hygiene Personnel’s Habits

Toxic Area

Heavy Metals

Polluted area

InadequateHygiene

Microorganismand Bacteria

Comes fromindustrial area

Figure 1 Cause-and-effect diagram for raw material cephalopod reception – CCP 1 (defective products).

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas 81

� 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology International Journal of Food Science and Technology 2009

Identify all potential failure modes associated withthe product or process

A failure mode is defined as how a system, product,or process is failing. Now here arises some confusionin the semiconductor industry, which usually measuresits failure modes in terms of how the product orprocess is deviating from its specifications. A productor process can have hundreds of different failuremodes based on this definition, most of which arehighly correlated to each other because of a commonfailure mechanism behind them (Kececioglu & Dim-itri, 1991). A failure mechanism is defined as thephysical phenomenon behind the failure mode(s)observed, e.g. die cracking, corrosion, electro-migra-tion, etc. To simplify the use of FMEA in thesemiconductor industry, therefore, the engineer maychoose whether to construct the FMEA table in termsof failure modes or in terms of failure mechanisms(Stamatis, 1995).

List down each failure mode using its technical term

Using an official technical term for listing the failuremode prevents confusion. All potential failure modesshould be listed down for each item (product component

or process step). Column 2 of the FMEA table shallbe used for this purpose (http://www.dyadem.com/downloads/trial).

Describe the effects of each of the failure modes listedand assess the severity of each of these effects

For each of the failure modes in column 2, a corre-sponding effect (or effects) must be identified and listedin column 3 of the FMEA table. A failure effect is whatthe customer will experience or perceive once the failureoccurs. A customer may either be internal or external, soeffects to both must be included. Examples of effectsinclude inoperability or performance degradation of theproduct or process, injury to the user and damage toequipment (Stamatis, 1995).Assign a severity rating to each effect. Each company

may develop its own severity rating system, dependingon the nature of its business. A common industrystandard is to use a 1-to-10 scale system, with the ‘1’corresponding to ‘no effect’ and the ‘10’ to maximumseverity, such as the occurrence of personal injury ordeath with no warning, or a very costly breakdown of anenormous system. Column 4 of the FMEA table is usedfor the severity rating (SEV) of the failure mode(Kececioglu & Dimitri, 1991).

Decomposed Frozen

Raw Material

Freezer Malfunction

Low Quality ofRaw Materials

InsufficientFreezing Methods

Insufficient Personnel Skills

Unreliable Suppliers

Inadequate quality control

No Senior Management Commitment/ Cooperation

IncompleteManagement training

Incomplete Staff training

No Staff Cooperation

Low Funding

Low Funding

Unreliable Suppliers

Raw MaterialDecomposition over transfer

Insufficient inspection

Insufficient Maintenance of Raw MaterialsHigh Temperature

Discontinuing Electricity flow

InsufficientFrozen Storage

High Level of Parasites

Insufficient SterilizationIncomplete Training procedures

InsufficientFrozen Storage

Old Machinery

Bad maintenance

Bad Management

Bad Management

Infrequent ReplacementLow Funding

Low Funding

Infrequent Machinery Control

Low Quality Machinery

Insufficient personnel training

Bad Management

Bad Management

Bad Management

Unsuitable ControlEquipment

Bad Management

Bad Management

Bad Management

Bad Management

Low Funding

Low Funding

Low Funding

Low Funding

Low purity of water

Unreliable Suppliers

Pure Maintenance ofControl Equipment

Low Purity materials

Infrequentvalidation

Insufficient UseOf the Freezing Method

Comes from Industrial area

Incomplete Staff Training

Inadequate Quality Control

No Suppliers Control Plan

No Incoming MaterialsControl Plan

Incomplete training procedures

Unsuitable Environmental

Conditions

Inadequate Cephalopod Industrial

Storage Conditions

Inadequate Cephalopod Industry Storehouse Conditions

Inadequate SuppliersStorage Conditions

InadequateTransfer Conditions

Incomplete or Insufficient Contracts Procedures

Inadequate SuppliersStorehouse Structure

Incomplete or Insufficient Contracts Procedures

Low Funding

Low Funding

Low Funding

Low Funding

Figure 2 Cause-and-effect diagram for freezing & frozen storage – CCP 2 (decomposed frozen raw materials).

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas82

International Journal of Food Science and Technology 2009 � 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology

Identify the possible cause(s) of each failure mode

Aside from its effect(s), the potential cause(s) of everylisted failure mode must also be enumerated. Apotential cause should be something that can actuallytrigger the failure to occur. Examples of failure causesinclude improper equipment set-up, operator error,use of worn-out tools, use of incorrect softwarerevision and contamination. The potential causes arelisted in column 5 of the FMEA table (http://www.fmeainfocentre.com/).

Quantify the probability of occurrence PF of eachof the failure mode causes

The likelihood of each of the potential failure causeoccurring must be quantified. Every failure cause willthen be assigned a number [probability factor (PF)],indicating this likelihood or probability of occurrence.A common industry standard for this is to assign a ‘1’to a cause that is very unlikely to occur and a ‘10’ to acause that is frequently encountered. PF values foreach of the failure causes are indicated in column 6 of

the FMEA table (http://www.dyadem.com/downloads/trial).

Identify all existing controls (current controls) thatcontribute to the prevention of the occurrence of eachof these failure mode causes

Existing controls that prevent the cause of the failuremode from occurring or detect the failure before itreaches the customer must be identified and evaluatedfor its effectiveness in performing its intended function.Each of the controls must be listed in column 7 of theFMEA table (http://pagead2.googlesyndication.com/pagead/iclk).

Determine the ability of each control in preventingor detecting the failure mode or its cause

The effectiveness of each of the listed controls mustthen be assessed in terms of its likelihood of prevent-ing or detecting the occurrence of the failure modeor its failure cause. As usual, a number must beassigned to indicate the detection effectiveness

Scombrotoxinpresence in

final product

Insufficient Personnel Skills

Methods

Machinery’sMalfunction

Low QualityRaw Materials

UnsuitableStorehouse Place

No incomingMaterials Control Plan

No SuppliersOf Storied Materials

Control PlanBad Management

Low FundingIncomplete training

ProceduresBad Management

Low FundingIncomplete Corrective

And preventive actions

Inadequate Training

Insufficient FMEA Plan

Raw MaterialsContaminated over Storage

No Detoxify Method Application

Insufficient training

Low Funding

Workers HabitsInsufficient training

Bad Management

Raw MaterialsContaminated over Transfer

No Detoxify Method Application

Insufficient training

Low Funding

Insufficient Inspection

Raw Materials Come fromToxic area

Microorganism andBacteria

Poor HygieneRaw Materials Contaminated

During their Cultivation

Low Quality of FoodLow Quality of Water

Poor Hygiene

Non-certifiedexternal laboratory

Bad Management

Low Funding

Insufficient Control Methods

Low Funding

Bad Management

Inadequate LaboratoryEquipmentOld Machinery

Low Funding

Non-replaced Machinery

Bad Management

Low FundingPoor Maintenance

Of storage Machines Equipment

InfrequentReplacement

Low Funding

Unreliable Suppliers

Inadequate Quality Control

Incomplete Stuff Training

Bad Management

No Staff Cooperation

Low Funding

No Senior Management Commitment/

CooperationInadequate StorageIndustry’s Conditions

Toxic Area

Industrial Area

No Detoxify Application

Insufficient training

Inadequate CephalopodIndustry Structure

Incomplete or Insufficient Contract Procedures

Bad Management

Figure 3 Cause-and-effect diagram for raw material storage – CCP 3 (scombrotoxin presence in final product).

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas 83

� 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology International Journal of Food Science and Technology 2009

(DET) of each control. DET numbers are shown incolumn 8 of the FMEA table (Kececioglu & Dimitri,1991).

Calculate the RPN

The risk priority number (RPN) is simply the product ofthe failure mode severity (SEV), failure cause probabil-ity (PF), and control detection effectiveness (DET)ratings. Thus, RPN = SEV · PF · DET). RPN, whichis listed in column 9 of the FMEA table, is used inprioritising which items require additional quality plan-ning or action (http://www.semiconfareast.com/fmea.htm).

Identify action(s) to address potential failure modes thathave a high RPN

A high RPN needs the immediate attention of theengineer as it indicates that the failure mode can result inan enormous negative effect, its failure cause has a highlikelihood of occurring and there are insufficient controls

to catch it. Thus, action items must be defined to addressfailure modes that have high RPN values (SAE Aero-space Recommended Practice ARP5580, 2001). Theseactions include but should not be limited to thefollowing: inspection, testing, monitoring, redesign,de-rating, conduct of preventative maintenance, redun-dancy, process evaluation ⁄optimisation, etc. Column 10of the FMEA tables is used to list down applicableaction items (http://pagead2.googlesyndication.com/pagead/iclk).

Implement the defined actions

Assign a responsible owner and a target date ofcompletion for each of the actions defined. This makesownership of the actions clear-cut and facilitates track-ing of the actions’ progress. The responsible owner andtarget completion dates must be indicated in column11 of the FMEA table. The status or outcome ofeach action item must also be indicated in column 12of the FMEA table (http://www.semiconfareast.com/fmea.htm).

Chemically Contaminated

Product

Personnel

Methods

UnsuitableMachinery

Unsuitable Materials

Ingredients ContaminatedOver their production

No GMP, GAP, GLPapplication

Raw Material/Ingredient

ContaminatedOver transfer

Worker's habits

Inadequate Training

No GMPapplication

Inadequate TrainingInsufficientInspection

Raw Materials come from Contaminated area

No Heavy MetalDetection Plan

Bad Management

No Suppliers Control Plan

Bad Management

Low Funding Incomplete Training

Procedures

Bad Management

Low FundingIncomplete Review

Procedures

Incomplete CorrectiveAnd Preventive Actions

Inadequate Training

Incomplete StaffTraining

Unreliable Supplier

Bad Management

Old Machinery

Unsuitable Control Equipment

Low Funding

Bad Management

No-certified External Laboratory

Unsuitable Laboratory Equipment

No GLP Application

Bad Management

Low Funding

Poor MaintenanceOf Control Equipment

Low PurityReference Materials

UnsuitableEnvironmental

Conditions

Inadequate Supplier’sStorehouse Structure

Incomplete or InsufficientContracts Procedures

Inadequate transferconditions

Incomplete or InsufficientContracts Procedures

Inadequate SuppliersStorage conditions

Inadequate Cephalopod Industry Storehouse Stracture

Low Funding Inadequate Cephalopod

Industry Storage Conditions

Insufficient RecallProcesses of non- conformed

No Senior ManagementCommitment/Cooperation

No Staff Cooperation

Low Funding

Inadequate QualityControl

Bad Management

Low Funding

Bad Management

Low Funding

Figure 4 Cause-and-effect diagram for butchering ⁄ packaging – CCP 4 (chemically contaminated product).

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas84

International Journal of Food Science and Technology 2009 � 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology

Review the results of the actions taken and reassessthe RPN values

After the defined actions have been completed, theiroverall effect on the failure mode they are supposed toaddress must be reassessed. The engineer must updatethe SEV, PF and DET numbers accordingly. The newRPN must then be recalculated once the new SEV, PFand DET numbers have been established. The new RPNshould help the engineer decide if more actions areneeded or if the actions are sufficient. Columns 13, 14,15 and 16 of the FMEA table are used to indicate thenew SEV, PF, DET and RPN respectively (http://www.fmeainfocentre.com/).

Keep the FMEA table updated

Update the FMEA table every time the product designor process changes or new actions or information causethe SEV, PF or DET to change (SAE AerospaceRecommended Practice ARP5580, 2001).

Cause-and-effect diagram

The cause-and-effect diagram is the brainchild ofKaoru Ishikawa, who pioneered quality management

processes in the Kawasaki shipyards, and in theprocess, became one of the founding fathers ofmodern management. The cause and effect diagramis used to explore all the potential or real causes (orinputs) that result in a single effect (or output). Causesare arranged according to their level of importance ordetail, resulting in a depiction of relationships andhierarchy of events. This can help anyone search forroot causes, identify areas where there may beproblems, and compare the relative importance ofdifferent causes (http://mot.vuse.vanderbilt.edu/mt322/Ishikawa.htm).Causes in a cause-and-effect diagram are frequently

arranged into four major categories. The most impor-tant ones are the following: manpower, methods,materials and machinery (recommended for manufac-turing); equipment, policies, procedures and people(recommended for administration and service) (http://www.semiconfareast.com/ishikawa.htm).The cause-and-effect diagram is also known as the

fishbone diagram because it was drawn to resemble theskeleton of a fish, with the main causal categories drawnas ‘bones’ attached to the spine of the fish (http://www.isixsigma.com/tt/cause_and_effect/).Cause-and-effect diagrams can also be drawn as tree

diagrams, resembling a tree turned on its side. From a

IncorrectLabeledProduct

Personnel Methods

Machinery Materials

No LabelingControl Plan

Bad ManagementNo Labeled Product

Control Plan Bad Management

IncompleteTraining Procedures

Bad Management

Low FundingIncomplete

Review Procedures

Bad ManagementIncomplete Corrective And Preventve Actions

Inadequate Training

IncorrectLabels

Bad Management

Insufficient Training

Low Funding

Product changeDuring the procedure

No DetectionControl Plan

Bad Management

Unsuitable Labeling Materials

Unreliable Suppliers

Bad Management

Unsuitable Control Equipment

Bad ManagementLow Funding

Inadequate or IncompleteMachinery Equipment

Bad ManagementLow Funding

Poor Maintenance ofLabeling Machines

Bad Management

Infrequent Validation

Low Funding

Low Funding

Old MachineryNon-replaced

Machinery

Bad Management

Low Funding

No Stuff Cooperation

Insufficient Training

Incomplete Staff Training

Low Funding

Bad ManagementLow Funding

Unreliable SupplierInadequate

Quality Control

No GMP Application

Figure 5 Cause-and-effect diagram for packing – CCP 5 (incorrect labelled product).

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas 85

� 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology International Journal of Food Science and Technology 2009

single outcome or trunk, branches extend that representmajor categories of inputs or causes that create thatsingle outcome. These large branches then lead tosmaller and smaller branches of causes all the waydown to twigs at the ends. The tree structure has anadvantage over the fishbone-style diagram. As a fish-bone diagram becomes more and more complex, itbecomes difficult to find and compare items that are ofthe same distance from the effect because they aredispersed over the diagram. With the tree structure, allitems on the same causal level are aligned vertically(http://www.saferpak.com/cause_effect.htm).Ishikawa diagrams have been drawn for the different

CCPs. Figure 1 depicts the cause-and-effect diagram forraw material cephalopod reception – CCP 1 (defectiveproducts). Figure 2 shows the cause-and-effect diagramfor freezing & frozen storage – CCP 2 (decomposedfrozen raw materials). Figure 3 shows the cause-and-effect diagram for raw material storage – CCP 3(scombrotoxin presence in final product). Figure 4shows the cause-and-effect diagram for butcher-ing ⁄packaging – CCP 4 (chemically contaminatedproduct). Figure 5 depicts the cause-and-effect diagram

for packing– CCP 5 (incorrect labelled product) andfinally Fig. 6 depicts the cause-and-effect diagram forstorage in cans (mechanical) – CCP 6 (productscontaining foreign matter).

Mollusk processing line

Table 1 shows the FMEA – basic table for CCP 1(cephalopods reception). Table 2 shows the FMEA –basic table for CCP 3 (scombrotoxin formation).Causes of failure are the following: (1) no incomingmaterials’ control plan; (2) no suppliers’ control plan;(3) incomplete training procedures; and (4) incompletecorrective actions. Ways of control are the following:(1) extra control measures; (2) more funding, bettermanagement and better staff training; and (3) betterFMEA plan.Table 3 shows the FMEA for CCP 4 (chemically

contaminated product). Causes of failure are thefollowing: (1) raw materials ⁄ ingredients contami-nated over their production; (2) raw materials ⁄ ingre-dients contaminated over transfer; and (3) rawmaterials obtained from contaminated area. Ways of

ProductsContaining

Foreign Materials

Personnel InsufficientControl Methods

MachineryMaterials

No incoming MaterialsControl Plan

Bad Management

IncompleteTraining Procedures

Bad Management

Low FundingIncomplete Review

Procedures

Bad Management

Incomplete CorrectiveAnd Preventive Actions

Inadequate Training

Ingredient contain Objectionable Matter

Unreliable Suppliers

Bad Management

Low Funding Product Include Low Quality Raw Material

Low Funding

Low Purity of Water

Low Funding

Low Quality ofCanning Materials

Unreliable Suppliers

Bad Management

Insufficient Recall Processes

Of Non- conformed

No Staff Cooperation

Low Funding

Pure Staff Hygiene

Workers Habits

Inadequate TrainingUnreliable Supplier

No GHP, GMP application

Inadequate Quality Control

Insufficient Staff Training

Bad Management

UnsuitableControl Equipment

Low Funding

Bad Management

Non CerrtifiedMachinery

Unreliable Suppliers

Inadequate or IncompleteMachinery Equipment

Bad Management

Low Funding

Pure Maintenanceof Machinery

Infrequent Control

Bad Management

Incomplete Staff Training

Low Funding

Figure 6 Cause-and-effect diagram for storage in cans (mechanical) – CCP 6 (products containing foreign matter).

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas86

International Journal of Food Science and Technology 2009 � 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology

control are: (1) better workers’ training; (2) extracontrol and detection measures; and (3) bettermanagement.

Table 4 shows the FMEA for CCP 6 (productcontains foreign matter). Causes of failure are thefollowing: (1) no incoming material’s control plan; (2)

Table 2 Failure mode and effect analysis – basic table – CCP 2

FMEA table of hazardous processing methods

Decomposed frozen raw materials

Estimated

corrective

actions

result

Failure mode Consequences S Failure causes O Ways of control D RPN Corrective actions S O D RPN

Can come from

a possible

malfunction

of the freezer

Because of the high

technique control this

case could be impossible

for the most of the

products. So that it isn’t

a noticeable point

3 1) Unsuitable control

equipment

2) Poor maintenance

3) Insufficient

personnel skills

4) Insufficient use

5) Old machinery

6) Infrequent control

of the freezer

5 1) More funding,

better management

2) Frequent validation

3) Completed staff

training

4) Completed staff

training, more funding

5) More funding,

better management

6) Frequent control

4 60 Not required 3 5 4 60

Can come from

the low quality

of raw

materials

Because of low ingredient

quantity present, even if

there are unsuitable

frozen raw materials

1%, it allows

certification towards to

customers that the final

product has been

repeatedly controlled

2 1) Unreliable

suppliers

2) Raw materials

decomposed

over transfer

3) Bad maintenance

of frozen raw

materials

4) High quantity of

parasites or toxins

5 1) Extra control

measures

2) Better inspection

3) Keeping of raw

materials in cool and

shady place

4) Extra hygiene

measures and staff

training

3 30 Not required 2 5 3 30

Can come from

insufficient

personnel

skills

Even if the case of

insufficient

personnel skills it is not

remarkable from the

appearance of the final

product

by the customer,

the most of

products couldn’t

pass all

control tests.

7 1) No Staff

cooperation

2) Incomplete staff

training

3) No senior

management

commitment ⁄cooperation

4) Unreliable supplier

6 1) Extra funding

2) Extra funding

3) Completed

management training

4) Extra control

measures

7 294* Extra control

measures, extra

funding, better

management.

approved supplier’s

certification is

provided as well

as technical cards

4 5 4 80

Can come from

the low quality

of the freezing

method

Important effects by

decomposed frozen raw

materials

as ingredients more

than 30%.

Possible obligation for

non-permitting

frozen raw materials. All

of customers

are affected. Strong

disconnect of

consume unions.

Possible trade

failure of product

8 1) Low purity of water

for the freezer

2) Insufficient use of

the freezing method

3) Inadequate quality

control

4) No incoming

material’s

control plan

5) No suppliers

control plan

6) Incomplete

corrective and

preventive actions

5 1) Extra control

measures

2) Completed Staff

Training, More

funding

3) Extra control

measures

4) >>

5) >>

6)>>

6 240* Extra control

measures, Extra

Funding, Better

management.

Approved

supplier’s

certification is

provided as

well as

Technical Cards

4 5 4 80

*When the RPN is over 130, corrective action is required.

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas 87

� 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology International Journal of Food Science and Technology 2009

raw materials ⁄ ingredients contaminated over their pro-duction; (3) raw materials contaminated over transfer;(4) low-quality canning materials; (5) low-purity water;

and (6) no suppliers’ control plan. Ways of control are:(1) extra control measures; (2) better transfer conditions;(3) extra funding; and (4) extra control measures.

Table 3 Failure mode and effect analysis – basic table – CCP 3

FMEA table of hazardous processing methods

Scombrotoxin formation

Estimated

corrective

actions result

Failure mode Consequences S Failure causes O Ways of control D RPN Corrective actions S O D RPN

Can come from low

effectiveness of

detoxification

method

Important effects by

severe decomposed

frozen raw materials

as ingredients more

than 3%. Possible

obligation for

non-permitting

frozen raw

materials. All of

customers are

affected. Strong

disconnect of

consume unions

9 1) No incoming

materials’ control

plan

2) No suppliers

control plan

3) Incomplete

training procedures

4) Incomplete

corrective actions

4 1) Extra control

measures

2) Extra control

measures

3) More funding,

better management,

better staff training

4) Better FMEA plan

3 108 Not required 9 4 3 108

Can come from the

low quality of raw

materials

Because of low

ingredient quantity

present, even if

there are unsuitable

frozen raw materials

3%, it allows

certification towards

to customers that

the final product has

been repeatedly

controlled

8 1) Raw materials

contaminated over

transfer

2) Raw materials

contaminated over

storage

3) Raw materials

come from

toxic area

4) Raw material

contaminated

during their

cultivation

4 1) Better transfer

conditions

2) Better storage

conditions

3) Extra control

measures.

Application better

detoxify methods.

4) Extra control

measures and

suppliers detection.

Extra hygiene

measures

5 160* Extra control

measures, extra

funding, better

management.

Approved suppliers’

certification is

provided as well as

technical cards

acceptance of

products and

sampling-sending

for toxin analysis

5 3 4 60

Can come from

insufficient

personnel skills

Even if the case of

insufficient

personnel skills it is

not remarkable from

the appearance of

the final product by

the customer, the

most of products

could not pass all

control tests

8 1) No staff

cooperation

2) Incomplete staff

training

3) No senior

management

commitment ⁄cooperation

4) Unreliable supplier

5 1) Extra funding

2) Extra funding

3) Completed

management

training

4) Extra control

measures

7 280* Extra control

measures, extra

funding, better

management.

GMP, GHP and GLP

application

4 3 6 72

Can come from

unsuitable storage

place

Even if the case of the

presence of toxins

nearby or inside the

storage place of the

product exists, all

the products could

easily be

contaminated.

Possible trade failure

of the product

7 1) Inadequate

cephalopod

industrial storage

conditions

2) Inadequate

industrial structure

3) No detoxification

5 1) Better storage

conditions

2) Better plan for the

industrial structure

3) Frequent

detoxification

5 175* GMP, GHP and GLP

application.

Extra control

measures, extra

funding, better

management.

4 4 3 48

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas88

International Journal of Food Science and Technology 2009 � 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology

Table 4 Failure mode and effect analysis – basic table – CCP 4

FMEA table of hazardous processing methods

Chemically contaminated product

Estimated

corrective

actions result

Failure mode Consequences S Failure causes O Ways of control D RPN Corrective actions S O D RPN

Can come from

unsuitable raw

materials

Because of the very

important effect by

several chemically

contaminated raw

materials, more than

5%, the use of those

products is not

permitted. They are

very dangerous

products

9 1) Raw materials ⁄ingredients

contaminated

over their

production

2) Raw materials ⁄ingredients

contaminated

over transfer

3) Raw materials

come from

contaminated area

7 1) Better workers

training

2) Extra control and

detection measures

3) Better

management

6 378* GMP, GHP and GLP

application.

Extra control

measures, Extra

Funding, Better

management.

6 4 4 96

Can come from

Insufficient

Personnel Skills

Even if the case of

Insufficient

personnel skills it is

not remarkable from

the appearance of

the final product by

the customer, the

most of products

couldn’t pass all

control tests.

5 1) No Staff

Cooperation

2) Incomplete Staff

training

3) No Senior

Management

Commitment ⁄Cooperation

4) Unreliable Supplier

7 1) Extra funding

2) Extra funding

3) Completed

management

training

4) Extra control

measures

6 210 GMP, GHP and GLP

application.

Better management

3 3 4 36

Can come from the

low effective

chemical clearing

method

Important effect due

to severe chemical

presence in

ingredient greater

than 0.5%. Possible

obligation for

special labelling for

non-permitting

chemicals. Large

part of customers

affected. Strong

disconnect of

environmental

organisations and

consumer unions.

Possible trade

failure of product

8 1) No incoming

material control plan

2) No suppliers’

control plan

3) Incomplete

training procedures

4) Incomplete

corrective actions

4 1) Extra control

measures

2) >>

3) More funding,

better management,

better staff training

4) Better FMEA plan

3 96 Not required 8 4 3 96

Can come from a

possible

malfunction of

machineries

Because of the high

technique control

this case could be

impossible for the

most of the

products. So that it

is not a noticeable

point (Seixas et al.,

2005)

5 1) Unsuitable control

equipment

2) Poor maintenance

3) Insufficient

personnel skills

4) Insufficient use

5) Old machinery

4 1) More funding,

better management

2) Frequent validation

3) Completed staff

training

4) Completed staff

training, more

funding

5) More funding,

better management

4 80 Not required 5 4 4 80

*When the RPN is over 130, corrective action is required.

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas 89

� 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology International Journal of Food Science and Technology 2009

Table 5 shows the FMEA for CCP 2 (decomposedfrozen raw materials). Causes of failure are the follow-ing: (1) unsuitable control equipment; (2) poor main-tenance; (3) insufficient personnel skills; (4) insufficientuse; (5) old machinery; and (6) infrequent control of thefreezer. Ways of control are through: (1) more funding,better management; (2) frequent validation; (3) com-

pleted staff training; (4) completed staff training, morefunding; (5) more funding, better management; and (6)frequent control.Table 6 shows the FMEA for CCP 5 (incorrectly

labelled product). Causes of failure are the following:(1) unsuitable control equipment; (2) poor main-tenance; (3) insufficient personnel skills; (4) insufficient

Table 5 Failure mode and effect analysis – basic table – CCP 5

FMEA table of hazardous processing methods

Incorrectly labelled product

Estimated

corrective

actions result

Failure mode Consequences S Failure causes O Ways of control D RPN Corrective actions S O D RPN

Can come from a

possible

malfunction of

machineries

Same customers may

have allergies to

some ingredients of

this product. So, it is

important to stick

the correct labels,

otherwise, this could

be dangerous or

with undesirable

effects to customers.

6 1) Unsuitable control

equipment

2) Poor maintenance

3) Insufficient

personnel skills

4) Insufficient use

5) Old machinery

5 1) More funding,

better management

2) Frequent validation

3) Completed staff

training

4) Completed staff

training, more

funding

5) More funding,

better management

6) Frequent

machineries control

2 60 Not required 6 5 2 60

Can come from

insufficient

personnel skills

Even if the case of

insufficient

personnel skills it is

not remarkable from

the appearance of

the final product by

the customer, the

most of products

could not pass all

control tests

6 1) No Staff

cooperation

2) Incomplete staff

training

3) No senior

management

commitment ⁄cooperation

4) Unreliable supplier

6 1) Extra funding

2) Extra funding

3) Completed

management

training

4) Extra control

measures

7 252* GMP, GHP and GLP

application.

3 4 4 48

Can come from the

low effective

labelling

control method

Some customers will

not take this product

again. With the pass

of time it will be the

most unreliable

trade. Although, it is

not so important

and dangerous

factor for trade

failure

4 1) No labels control

plan

2) Incomplete

training procedures

3) Incomplete review

procedures

4) Incomplete

corrective and

preventive actions

3 1) Extra control measures

2) Extra training procedures

3) Extra control measures

4) Better management

3 36 Not required 4 3 3 36

Can come from

unsuitable

materials

Because of the many

controls during the

procedure of the

product some

incorrect labelling

materials will not

affect so mach the

customers

2 1) Incorrect labels

2) Labels change

during the retail

3) Unsuitable

labelling materials

3 1) Extra control

measures

2) Stronger labelling

materials

3) Extra funding

3 18 Not required 2 3 3 18

*When the RPN is over 130, corrective action is required.

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas90

International Journal of Food Science and Technology 2009 � 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology

use; and (5) old machinery. Ways of control are thefollowing: (1) more funding, better management; (2)frequent validation; (3) completed staff training;

(4) completed staff training, more funding; (5)more funding, better management; and (6) frequentmachinery control.

Table 6 Failure mode and effect analysis – basic table – CCP 6

FMEA table of hazardous processing methods

Product contains foreign matter

Estimated

corrective

actions result

Failure mode Consequences S Failure causes O Ways of control D RPN Corrective actions S O D RPN

Can come from the

low quality of raw

materials

Because of low

ingredient quantity

present, even if

there are foreign

materials 1%, it

allows certification

towards to

customers that the

final product has

been repeatedly

controlled

5 1) No incoming

material’s control

plan

2) Raw materials ⁄ingredients

contaminated over

their production

3) Raw materials

contaminated over

transfer

4) Low-quality

canning materials

5) Low-purity water

6) No suppliers’

control plan

4 1) Extra control

measures

2)>>

3) Better transfer

conditions

4) Extra funding

5) >>

6) Extra control

measures

3 60 Not required 5 4 3 60

Can come from a

possible

malfunction of

machineries

Because of the high

technique control

this case could be

impossible for the

most of the

products. So that it

is not a noticeable

point

3 1) Unsuitable control

equipment

2) Poor maintenance

3) Insufficient

personnel skills

4) Insufficient use

5) Old machinery

3 1) More funding,

better management

2) Frequent validation

3) Completed

staff training

4) Completed staff

training, more

funding

5) More funding,

better management

6) Frequent

machinery control

2 18 Not required 3 3 2 18

Can come from

insufficient

personnel skills

Even if the case of

insufficient

personnel skills it is

not remarkable from

the appearance of

the final product by

the customer, the

most of products

could not pass all

control tests

5 1) No staff

cooperation

2) Incomplete staff

training

3) No senior

management

Commitment ⁄cooperation

4) Unreliable supplier

7 1) Extra funding

2) Extra funding

3) Completed

management

training

4) Extra control

measures

7 245* GMP, GHP and GLP

application.

3 4 5 60

Can come from the

low effective control

method

Because of the high

technique control

this case could be

impossible for the

most of the

products. So that it

is not a noticeable

point

2 1) No incoming

material control plan

2) No suppliers’

control plan

3) Incomplete

training procedures

4) Incomplete

corrective actions

4 1) Extra control

measures

2) More funding,

better management,

better staff training

3) Better FMEA plan

3 24 Not required 2 4 3 24

*When the RPN is over 130, corrective action is required.

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas 91

� 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology International Journal of Food Science and Technology 2009

Conclusions

The main emphasis was put on the quantification of riskassessment by determining the RPN per identifiedprocessing hazard. Chemically contaminated product,decomposed raw materials, scombrotoxin presence infinal product, incorrectly labelled product, storage incans (foreign matter) and defective products were theprocesses identified as those with the highest RPN (378,294, 280, 252, 245 and 144 respectively) and correctiveactions were undertaken. Following the application ofcorrective actions, a second calculation of RPN valueswas carried out leading to considerably lower values(below the upper acceptable limit of 130). It is note-worthy that the application of Ishikawa (cause-and-effect or tree diagram) led to converging results, thuscorroborating the validity of conclusions derived fromrisk assessment and FMEA. Therefore, the incorpora-tion of FMEA analysis within the ISO2200 system of anoctopus-processing industry is considered imperative.

References

Arvanitoyannis, I.S. & Tzouros, N.H. (2006). ISO 22000, The NewFood Quality and Safety Standard. Athens: Stamoulis S.A.

FAO, 2001. Fish and Fisheries Products. Food Outlook 1, February,vol. 11. Rome: FAO ⁄GIEWS.

Kececioglu, D. (1991). Reliability Engineering Handbook, vol. 2.Englewood Cliffs, NJ: Prentice-Hall Inc .

Kreuzer, R. (1984). Cephalopods: Handling, Processing and Products.Pp. 108. FAO Fish. Tech. Pap. 254

Miliou, H., Fintikaki, M., Kountouris, T. & Verriopoulos, G. (2005).Combined effects of temperature and body weight on growth andprotein utilization of the common octopus, Octopus vulgaris.Aquaculture, 249, 245–256.

SAE Aerospace Recommended Practice ARP5580 (2001). Recom-mended Failure Modes and Effects Analysis (FMEA) Practices forNon-Automobile Applications. Warrendale, PA: SAE International.

Seixas, S., Bustamante, P. & Pierce, G.J. (2005). Interannual patternsof variation in concentrations of trace elements in arms of Octopusvulgaris. Chemosphere, 59, 1113–1124.

Stamatis, D.H. (1995). Failure Mode and Effect Analysis: FMEA fromTheory to Execution. Milwaukee, WI: American Society for Quality(ASQ).

Vaz-Pires, P, Seixas, P. & Barbosa, A. (2004). Aquaculture potential ofthe common octopus (Octopus vulgaris Cuvier, 1797): a review.Aquaculture, 238, 221–238.

Internet sources

• http://www.npd-solutions.com/pdforum.html

• http://www.fmeainfocentre.com/

• http://www.fmeainfocentre.com/download/umich.pdf

• http://www.npd-solutions.com/fmea.html

• http://www.informatik.uni-bremen.de/uniform/gdpa_d/

methods/m-fmea.htm

• http://www.relexsoftware.com/products/fmeafmeca.asp

• http://www.ivt.ntnu.no/bat/bm/buildphys/proceedings/

68_Nielsen.pdf

• http://mot.vuse.vanderbilt.edu/mt322/Ishikawa.htm

• http://www.isixsigma.com/tt/cause_and_effect/

• http://www.saferpak.com/cause_effect.htm

• http://www.semiconfareast.com/ishikawa.htm

• http://mielsvr2.ecs.umass.edu/virtual_econ/module2/

Cause_effect.htm

• http://animaldiversity.ummz.umich.edu/site/resources/

Grzimek_inverts/Cephalopoda/Octopus_vulgaris.jpg/

view.html

• http://nsgl.gso.uri.edu/flsgp/flsgph01002.pdf

• http://www.afdo.org/afdo/upload/Sushi.pdf

• http://www.fao.org/docrep/meeting/008/j1682e/

j1682e04.htm

• http://www.codexalimentarius.net/download/report/

633/al28_18e.pdf

• http://europa.eu.int/comm/food/fvo/act_getPDF.

cfm?PDF_ID = 4573

• http://www.fish.wa.gov.au/docs/pub/SeafoodRetailers-

Guide/SeafoodRetailersGuidlelines.pdf

• http://www.nhmrc.gov.au/publications/_files/eh38.pdf

• http://www-nehc.med.navy.mil/downloads/prevmed/

MILHBK3006A.pdf

• http://www.e-haccp.org.uk/previousconference/

Conference_proceedings1.pdf

• http://www.blackwell-synergy.com/links/doi/10.1046/

j.1471-5740.2002.00038.x/pdf?cookieSet=1

Failure analysis in octopus processing I. S. Arvanitoyannis and T. H. Varzakas92

International Journal of Food Science and Technology 2009 � 2008 The Authors. Journal compilation � 2008 Institute of Food Science and Technology