Yam, Kit L. (ed.). The Wiley encyclopedia of packaging technology

Подождите немного. Документ загружается.

The combination is quenched in heated water until the

product achieves a temperature of approximately 601C

and then held at the elevated temperature for 6 minutes.

The heated bottle and contents are then cooled in stages.

The entire heat process takes approximately 20 minutes.

It is not the rapid heating of the container that is cause for

concern but rather it is the sudden cooling of the contain-

er’s outer surface. To ensure good thermal performance, a

limit on the level of the down shock is required.

Finally, nearly all bottles will receive impacts during

their intended usage regardless of whether they are

pressure, nonpressure, or vacuum items. Although impact

loading is the most commonly applied load, it is perhaps

the least understood. Of all the major loads applied to a

container, the applied forces created during an impact will

be mostly dependent on the properties of the container. In

fact, for a given striking velocity, the impact forces will

increase with increasing mass and contact stiffness (9).

The stiffness or rigidity of the container will vary with

impact location and design, while the mass will be depen-

dent on the weight of the container and product (if ﬁlled).

In general, the stiffness will be higher at the shoulder

and heel contact points than in the sidewall. This is due to

complex curvature of the shoulder and heel design. The

sidewall, which is cylindrical in shape, is far more ﬂexible

and exhibits a lower stiffness. Basic design parameters,

such as body diameter and wall thickness, will also affect

the stiffness. Decreasing body diameter and increasing wall

thickness will increase the stiffness. The total mass will

obviously be higher in ﬁlled bottles than in empty ones.

Similarly, the mass will be greater in larger capacity bottles

than in smaller ones, and the mass of reﬁllable bottles

will be greater than for nonreﬁllable bottles. Thus, higher-

impact forces will be created in ﬁlled versus empty, in larger

versus smaller capacity, and in reﬁllable versus nonreﬁll-

able ware. The performance requirements for impact load-

ing will vary with bottle size, design, and usage.

The maximum striking velocity that a container can

survive greatly depends on the forces created during an

impact. Again, these forces depend on the container size

and shape and the location of the impact site. Such

information is useful in selecting belt speeds and equip-

ment settings in handling and transport of the ware.

Differences in expected impact forces are the reasons

that empty ware is transported faster than ﬁlled ware,

small capacity ware is transported faster than larger

capacity ware, and lightweight nonreﬁllable bottles can

be transported faster than heavier reﬁllable bottles. As

demands for improving efﬁciencies increase, knowledge of

safe handling and transport speeds will be critical for

achieving good line performance (see Figure 6).

STRENGTH OF GLASS

The strength of pristine glass is considerably higher than

any other packaging material. However, design strengths

are compromised because glass is susceptible to surface

damage and prone to static fatigue (10). Surface damage

in glass bottles can result from the manufacturing process

or it can be created by mechanical contact during

handling, ﬁlling, or distribution of ware. The damage

creates stress concentrators in the surface, and any in-

crease in severity of the damage results in lower surface

strengths. Various strengths exist for different parts of the

bottle. Also, strengths for momentary loading, such as for

impacts during handling, are signiﬁcantly different than

those for long sustained loads, such as pressures produced

by a carbonated product. Thus, in order to achieve light

and efﬁcient designs without compromising performance,

an understanding of the surface strengths of a glass

container is critical.

There are natural contacts or ‘‘touch’’ points on the

outer surface of a glass bottle where the majority of

handling damage is generated. These regions include the

shoulder contact, heel contact, and bearing surface. Each

of these regions will exhibit lower surface strengths

according to the degree of damage sustained. Noncontact

regions on a glass bottle are generally protected from

mechanical damage and will exhibit higher surface

strengths. These regions include the shoulder, neck,

heel, and bottom. The parts of the bottle are illustrated

in Figure 7.

The degree of damage created during handling is also

strongly related to the market. For the nonreﬁllable

market, the degree of handling is limited by the single

trip through the process. In addition, surface treatments

are applied to the outer surface of bottles as a thin metal

oxide ﬁlm that is overcoated with a polyethylene, oleic, or

stearate compound. These surface treatments increase the

lubricity and scratch resistance of the bottles (11). With

limited handling and protected outer surfaces, the bottles

exhibit higher surface strengths that can be utilized for

lightweighting purposes.

For the reﬁllable market, the degree of handling ex-

tends over multiple trips, with most reﬁllable bottles

surviving 25 trips or more. Obviously, the degree of

handling damage is signiﬁcant from repeated usage as

evidenced by the visible wear bands created at the touch

points. Although surface treatments have been used on

reﬁllable bottles, their effectiveness is lost following 3 to 5

trips due to the caustic washes employed between cycles.

Once the surface treatments have been removed, the

Figure 6. High-speed transport of ﬁlled bottles.

558 GLASS BOTTLE DESIGN AND PERFORMANCE

bottles will then again become susceptible to scufﬁng,

resulting in lower surface strengths. It is noteworthy

that some limited success has been achieved with scuff-

free coatings applied to reﬁllable ware (12) that is paving

the way for lightweight returnable bottles.

The strength of the outer surface of bottles can also be

affected by the presence of labels or decorations and may

need to be taken into account in the evaluation of the

design. Shrink wrap labels that encapsulate the bottle can

protect the outer surfaces from handling damage that

may, in turn, result in higher surface strengths. Applied

Ceramic Labels (ACLs) are silk-screened ceramic paints

that are ﬁred onto the glass surface and become a perma-

nent part of the bottle. The strength of the labeled glass

depends on the voids and cracks in the ceramic label

which can lower the glass strength. Although the ACL

label was cost effective when applied on heavier reﬁllable

bottles, its strength lowering characteristics can limit the

degree of light-weighting in nonreﬁllable ware. Also,

frosted surfaces, which are created with an acid etch or

sandblasting, exhibit a pitted surface. These processes

remove the surface treatments, and the pitted surfaces

become more prone to handling damage.

Also, the strength of the inside surface of a container is

signiﬁcantly higher than the outer surface. This is due to

its pristine nature and lack of any mechanical contact.

However, there are differences in inside surface strengths

that depend on the manufacturing process. The inside

surface of blow-and-blow ware is stronger than that for

either press-and-blow or NNPB ware. Air is used to create

the internal cavity of the parison for blow-and-blow ware,

whereas the metal plunger used to create the internal

cavity of the parison in either the press-and-blow or

NNPB process can deposit small foreign particles into

the molten glass surface. These particles, often referred

to as ‘‘black specks,’’ can act as stress concentrators and

lower the strength of the glass. Thus, these differences

must be taken into account when evaluating the design for

performance.

Although glass is considered to be chemically inert,

freshly formed surface ﬂaws will react chemically with

moisture in the environment. This reaction will cause

stressed cracks to grow slowly, which in turn lowers the

breaking strength of the glass. This slow crack growth

mechanism is commonly referred to as ‘‘static fatigue’’ and

is responsible for delayed failures, such as pasteurization

or warehouse breakage. Because of its chemical nature,

increases in either moisture content or temperature will

accelerate this process. Thus, wetted glass is weaker than

dried glass and hot glass is weaker than cold glass.

When the ﬂaws are only momentarily stressed, such as

during an impact, there is very little time for any reactions;

and thus little, if any, crack extension or strength loss occurs .

However, when ﬂaws are subjected to sustained stresses,

such as those produced by a carbonated product stored

inside a beverage bottle, there is ample time for cracks to

react and extend, provided that the stress levels are sufﬁ-

ciently high. The time-to-failure depends on the initial ﬂaw

size and level of stress (see Figure 8). In general, as the

stress levels increase, times-to-failure decrease. Also, as the

initial ﬂaw size decreases, times-to-failure increase when

subjected to the same stress level. Nevertheless, if the length

of the extending ﬂaw becomes critical, spontaneous failure

can occur without warning. A thorough and prudent design

review takes these factors into account to avoid any issue

involving static fatigue .

PERFORMANCE

The ﬁrst step in evaluating the performance of the design

is to gather the facts. Identifying the product and any

special characteristics will dictate which load types will be

applied to the bottle. It is important to identify the

magnitude of each load type, since it may change depend-

ing on product, processing, or warehouse conditions. It

is important to identify the duration in which loads will

be applied to account for differences in static fatigue

strengths. Identifying the intended market is also impor-

tant because strength and weight requirements for reﬁll-

able markets are distinctly different from nonreﬁllable

markets. Determine whether any special labels will be

applied that might alter the performance characteristics—

for example, shrink wraps or ACLs. And ﬁnally, determine

the manufacturing process that will be used to produce

the bottle since it may affect glass weight, thickness

distribution, and inside surface strength. This informa-

tion will help determine the targeted weight, strength

Finish

Neck

Shoulder

Sidewall

Heel

Bearing Surface

Bottom

Touch Point

Recessed Label Panel

Figure 7. Parts of a bottle.

GLASS BOTTLE DESIGN AND PERFORMANCE 559

criteria, and complexity of modeling. Furthermore, this

information is necessary to establish the criteria against

which the design is judged for acceptable performance.

The next step is to create a computer model of the

design. This can be achieved by various methods. How-

ever, the outer surface proﬁle is usually predeﬁned with

all the drawing parameters listed on a bottle print. These

proﬁles may be completely new or a modiﬁcation to a

preexisting design. In most cases, the proﬁles can be

directly imported into the model software from a CAD

drawing ﬁle. Otherwise, the outer surface is manually

created to match the concept drawing. The inside surface

is initially created by using thicknesses that are equal

to industry minimum speciﬁcations in the sidewall and

bottom regions. Adjustments to these thicknesses may

become necessary following the analysis. Some software

packages will automatically deﬁne the inside surface from

simulations of the glass-blowing process (13). In addition

to glass weight, other bottle properties, such as overﬂow

capacity, percent headspace, tip angle, and centers of

gravity, are readily determined in the solid model. Any

changes to the geometry of the bottle can be quickly made

with these software packages to correct for any deﬁcien-

cies with bottle properties and/or weight.

The third step is to select and construct the appropriate

ﬁnite element model for determining the stress distribu-

tion resulting from applied loads. There are generally two

types of ﬁnite element models used to evaluate the per-

formance of the design. An axisymmetric model can be

used if the bottle is round and the applied loads are

axisymmetrically applied, such as pressure or top loading.

Such models are easily created and greatly reduce compu-

tational time. However, if the bottle is nonround or the

loading is nonaxisymmetric, such as impact, a three-

dimensional model is necessary. To reduce the size of the

three-dimensional model and the computational time,

axes of symmetry are often used. In most cases, either

half-bottle or quarter-bottle symmetry will sufﬁce.

Once the model is selected and the boundary conditions

are set, mainly for axes of symmetry, the loads are applied

to either individual nodes (special locations on the model)

or to surfaces. For internal pressure loading, the inside of

the bottle is deﬁned as a surface over which the pressure is

uniformly applied. In the case of impact or vertical load-

ing, individual forces are applied to the impact site or

along the sealing surface in the ﬁnish region of the bottle,

respectively.

The magnitude for the load is usually set to one. Thus,

for pressure, the load is 1 bar. For top loading, the total

load is 1 newton and for thermal shock, the temperature

difference is 11C. The corresponding stress results are

then indexed per unit load. Since glass is linearly elastic,

the total stress can simply be determined by multiplying

the stress index by the actual load. This approach also

conserves time and effort. For example, the stress distri-

bution plot shown in Figure 9 corresponds to a 12-oz long-

neck beer bottle subjected to a unit internal pressure load.

It is obvious that the stresses are not uniformly distrib-

uted throughout the bottle but vary from location to

location. In general, regions of the bottle most susceptible

to pressure performance problems are the contact points

where the strengths are the lowest. Regions of maximum

stress in the noncontact region are also susceptible to

performance problems; however, the surface strengths are

higher. This also includes the inside surface where surface

strengths are the highest.

Because these results are indexed, the total stress

generated from the counter ﬁlling pressure during heat

pasteurization and/or during storage in a warehouse can

be easily determined in spite of the fact that the pressure

levels and durations are different. The pressure levels are

determined from the (ﬁlling) machine settings or from the

expected carbonation levels and temperatures in the

pasteurizer or warehouse. The strength criteria are de-

termined from the durations of the loads and locations on

the bottle. Thus, one analysis is quickly turned into three

evaluations. The performance is determined by ensuring

that the stress levels do not exceed the expected strengths

in the various areas of the container for each of the load

conditions.

This same bottle will require a sealed closure for the

carbonated product. In general, the applied top force will

cause a compression or shortening of the bottle. Even

though glass does not break in compression, two bands of

tensile stress are generated at the touch points. As illu-

strated in Figure 10, one band is at the shoulder contact

1,000

10,000

100,000

1.E-04 1.E-02 1.E+00 1.E+02 1.E+04 1.E+06

log (t, sec)

log (S, psi)

Inside B & B

Inside NNPB

Non-Refillable Outside

Refillable Outside

Mild Abrasions

Figure 8. Approximate fatigue strength of soda lime container

glass.

560 GLASS BOTTLE DESIGN AND PERFORMANCE

and extends upward into the shoulder. The other band is

at the heel contact and it extends into the lower heel.

Since the results in Figure 10 are indexed, the total stress

can be determined by multiplying the index values by the

expected top loads. In the case of the capper, the indices

are multiplied by the total capping force. In the case of

warehouse storage, the indices are multiplied by the

enhanced stacking force. The strength criteria are deter-

mined from the durations of the loads and locations on

the bottle. Thus, one analysis is quickly turned into two

evaluations. Again, acceptable performance is determined

by ensuring that the stress levels do not exceed the

expected strengths in the various areas of the container.

An example of where the stresses are not axisymme-

trically distributed is shown in Figure 11, where an impact

is delivered to the heel contact of a supported 12-oz

long-neck beer bottle. In this case, a three-dimensional

model with quarter symmetry is employed due to the

bottle and loading conditions. There are three types of

stress generated during an impact (14). The ﬁrst and by

far the largest are the contact or Hertzian stresses. These

stresses are localized and created on the outer surface at

the impact site. They are mainly dependent on the hard-

ness and geometry of the contacting object (15). The

second highest stress is the ﬂexure stress, which occurs

on the inside surface at the impact site. These stresses are

also localized and become more of a performance-related

issue with NNPB or Press & Blow ware due to the lower

inside surface strengths. The third and lowest of the three

stresses are the hinge stresses as shown in the ﬁgure.

They occur on the outer surface and away from the impact

site. For the heel impact, there are hinges on either side of

Max Principal

3.5

2.5

1.5

0.5

Figure 10. Vertical load stress indices

in a 12-oz long-neck beer bottle (psi/lb

Max Principal

Figure 9. Internal pressure stress in-

dices in a 12-oz long-neck beer bottle

(psi/psi).

GLASS BOTTLE DESIGN AND PERFORMANCE 561

the impact site. There are also heel hinges that occur

directly beneath the impact site, which often extend

downward into the bearing surface where the surface

strength are lower.

As shown in Figure 11, the results are also indexed.

However, the unit force is taken as a unit striking velocity,

where the results are expressed for a ﬁlled bottle striking

a rigid object. As such, the performance of the bottle is

measured against a severe trade impact. The maximum

striking velocity that ﬁlled containers can survive is then

determined by dividing the expected impact strengths by

the various stress indices for the severe impact loading.

Such information is useful in selecting belt speeds and

equipment settings in the handling and transport of the

ware. Alternatively, if the transport velocities in a ﬁlling

line are known, the total stresses generated in the ﬁlled

bottle subjected to a severe impact can be determined by

multiplying the stress indices by the line velocity. Again,

acceptable performance is determined by ensuring that

the stress levels do not exceed the expected strengths in

the various areas of the container.

Obviously, total performance is dependent on the in-

dividual performances for the various loading conditions.

To ensure that the design will be a safe package for the

consumer, the stresses generated for all of the various

loads at all locations in the bottle must be less than the

expected surface strengths associated with the glass bot-

tle. If any of the stresses equal or exceed the surface

strengths, the entire design is then considered question-

able or unacceptable until all of the performance criteria

are met for the expected usage. Modiﬁcations to the design

are then necessary to lower the stresses into an acceptable

range. This may involve adjusting the thickness distribu-

tion by moving glass into more sensitive regions without

changing the glass weight. It may involve altering the

shape of the shoulder or heel of the bottle without

violating the theme of the package. Finally, the suggested

or targeted weight may simply be insufﬁcient, and addi-

tional glass has to be added to the design. Whatever

approach is taken, the goal is still to offer the consumer

an appealing and safe container package.

MODIFICATIONS

Failure to meet a performance requirement is no reason to

abandon a design. One of the advantages of computer

modeling is the ease and quickness of changing a design.

However, to utilize the efforts of marketing research in

developing the look and feel of the concept design, changes

must be made thoughtfully so that the theme or ﬂavor of

the design, which is often used for brand recognition, is

maintained. Consider the changes made to the Michelob

bottle as shown in Figure 12. When viewed together, the

changes may not appear as modest; however, these design

changes occurred over a 30-year span. However, during

that time, 127 g of glass were removed from the original

design, making today’s bottle a true lightweight. Even at a

lighter weight, the design has better performance char-

acteristics than the original design due to computer

modeling and evaluation. It is worth noting that the shape

still exhibits the long tapered neck and pinched waist

associated with the original Michelob bottle style.

Obviously, all designs cannot afford 30 years to evolve

into high-performance lightweight bottles. With markets

and tastes rapidly changing, responses must be quick and

direct. The quickest approach to improve a design’s per-

formance is to shift glass into sensitive regions without

changing the total glass weight. No changes are required

to the outer surface, leaving the theme of the design

untouched. By keeping the weight ﬁxed, there are no

changes to overﬂow capacity or head space with only

Max Principal

126

108

Side

Hinge

Impact Site

Heel Hinge

Bearing Surface

Hinge

Figure 11. Heel impact stress indices

in a 12-oz long-neck beer bottle (psi/ips).

562 GLASS BOTTLE DESIGN AND PERFORMANCE

slight changes occurring in the center of gravity and tip

angle. This obviously requires removing glass from other

areas of the container. In many designs, the bases are

under designed while the upper shoulder and neck regions

are over designed. Thus, shifting glass from the neck and

shoulder regions into the base is often recommended. In

fact, a study by Davis and Shott showed that the shift in

glass into the lower heel region is one of the best ways to

enhance the pressure performance of lightweight glass

bottles (16).

Improvements to the performance of a design can also

be achieved by changing its shape. However, in many

instances, shape changes in the design can lead to com-

promise. For example, the effect of increasing the contact

radius of a simple shoulder design is shown in Figure 13

where the conical shape is transformed into a rounded or

bulb-like design. The stress distribution plots in the upper

half of the ﬁgure show that for internal pressure, stress at

the shoulder contact increases with increasing shoulder

contact radius. Also, the maximum stress in the shoulder

region decreases as the shoulder contact radius increases.

If the shoulder contact stress is too high, reducing the

radius will help lower its value. Keep in mind that as the

shoulder contact radius is reduced, the maximum stress in

the shoulder is also increasing. Thus a compromise may be

needed on the radius change in order to maintain both

levels of stress in an acceptable range.

There may be other compromises to consider. The

results for the same shoulder contact radius changes for

vertical loading are shown in the lower half of the ﬁgure.

Figure 12. Evolution of the Michelob bottle.

Figure 13. Effect of shoulder shape on bottle perfor-

mance: (a) Internal pressure (4–18 psi/psi) and (b) vertical

loading (0–1.4 psi/lb).

GLASS BOTTLE DESIGN AND PERFORMANCE 563

These results are opposite to those found for internal

pressure loading. As the shoulder contact radius in-

creases, the stress at the shoulder contact decreases while

the maximum stress in the shoulder region increases.

Thus, while making shape changes to the design, it is

important to consider the effect of any shape change on

the stresses created by all of the applied loads, thus

preventing the introduction of a new issue while attempt-

ing to resolve another. Similar compromises may be

required when altering the push-up, heel contact height

or toe-in of the design.

Lightweighting is perhaps the most important innova-

tion in keeping glass packaging competitive. The rules for

performance of lightweight designs are no different; stres-

ses generated from the applied forces must not exceed

the surface strengths. However, as glass is removed from

the design, the stresses will increase with decreasing wall

thickness and there will be a limit to lightweighting when

the surface strengths of the bottle are exceeded. Addi-

tional lightweighting can only then be achieved by redu-

cing stresses and/or increasing design strengths. Attempts

have been made to increase strengths by tempering glass

bottles (17) or enhancing the surface strengths with

acrylate coatings (18).

The design strength can also be enhanced through

creative design methods. Since the design strength is

dependent on the degree of handling abuse, containers

can be designed to focus the handling damage to lower

stressed regions and protect the higher stressed regions

from handling abuse. Including recessed label panels in a

bottle design will help protect both the label and the side-

wall from normal bottle-to-bottle handling damage. Having

a higher design strength in the sidewall paves the way for

removing glass in this region. In fact, the most efﬁcient

place to remove glass is from the sidewall. The recessed

label panel also focuses any handling damage at the

shoulder and heel touch points where the stresses decrease

with increasing label panel depth (19). Also, including

knurling on the bearing surface in the design concentrates

any handling damage to the knurl tips. The knurls act as

stress relievers, exhibit lower stresses, and protect the base

glass from becoming damaged (see Figure 14).

Figure 14. Effect of bearing surface knurling on

bottle performance (psi/psi).

564 GLASS BOTTLE DESIGN AND PERFORMANCE

CONCLUSION

The glass container industry has come a long way since

the wake-up call of the 1970s. Following an initial decline

in the market, the industry has stabilized. In fact, the

industry has been steadily growing (1). Today, nearly 300

billion containers are produced worldwide on an annual

basis. This is due in part to the numerous new designs and

markets. However, a weight reduction of more than 30%

in the past 20 to 30 years has led glass packages to be

economical and competitively priced. The task of achiev-

ing lightweight containers was not straightforward, and

many factors had to be taken into consideration. Invest-

ments into technology for improved melting, forming, and

coating were required. This provided larger and more

efﬁcient furnaces. Larger and faster IS forming machines

that utilized the NNPB technology replaced the smaller,

less efﬁcient machines. Improvement to the surface treat-

ment technology provided higher design strength. Invest-

ments into new equipment, plants, and warehouses were

necessary. However, it is unlikely that the successes in the

glass container industry could have been achieved without

the understanding of the importance of design on

performance.

ACKNOWLEDGEMENTS

Special thanks to the Glass Packaging Institute for the

photographs in Figures 1, 2, 3, and 12, and as well as

to William Slusser for his insightful comments and

discussions.

BIBLIOGRAPHY

1. Glass Packaging Essentials, A Multimedia Resource, 2nd

edition, Glass Packaging Institute, Alexandria, VA, 2004.

ohttp://www.gpi.org/education/materials/>.

2. Glass: Clear Vision for a Bright Future, U.S. Department of

Energy, April, 1996; ohttp://www1.eere. energy.gov/industry/

glass/pdfs/glass_vision.pdf>

3. A. L. Griff, in The Wiley Encyclopedia of Packaging Technol-

ogy, John Wiley & Sons, New York, 1986, pp. 121–124.

4. J. L. Pellegrino, Energy and Environmental Proﬁle of the U.S.

Glass Industry, U.S. Department of Energy, April 2002;

ohttp://www1.eere.energy.gov/industry/glass/pdfs/

glass2002proﬁle.pdf>

5. R. Chinella, Glass Industry April, 14 (1990).

6. V. H. Rawson and G. Turton, Glastech. Ber. 46, 28 (1973).

7. algor.com/news, December 4, 2003; ohttp://www.algor.com/

news_pub/news_releases/2003/agr /default.asp>

8. B. Moody, Packaging in Glass, Hutchinson London, 1977 pp.

361–363.

9. M. W. Davis, Glass Technology 34(4), 163–165 (1993).

10. R. E. Mould, Fundam. Phenom. Mater. Sci. 4, 119 (1967).

11. H. P. Williams, Glass Technol. 16(2), 34 (1975).

12. H. B. Lingenberg, Brauwelt 135(26), 1283 (1995).

13. ‘‘IFORM,’’ Rockﬁeld Software Ltd. Technium, Swansea, SA1

8PH, West Glamorgan, U.K.

14. M. W. Davis, Int. Glass J. 99, 42–46 (1999).

15. A. P. Boresi, O. M. Sidebottom, F. B. Seely, and J. O. Smith,

Advanced Mechanics of Materials, 3rd edition, John Wiley &

Sons, New York, 1978, pp. 581–627.

16. M. W. Davis, and A. R. Shott, Glass June 221–222 (1995).

17. C. E. Mongan, Jr., U.S. Patent 2338071 (to Hartford Empire

Co.) 1943.

18. Pennekamp, Glass Industries, January, 32 (1999).

19. M. W. Davis, and A. R. Shott, Glastech. Ber. 66(9), 221–226

(1993).

GLASS BOTTLE DESIGN AND PERFORMANCE 565

HAZARD ANALYSIS AND CRITICAL CONTROL

POINTS

BARBARA BLAKISTONE

National Fisheries Institute,

McLean, Virginia

YUHUAN CHEN

Grocery Manufacturers

Association, Washington, D.C.

ORIGIN AND HISTORY OF HACCP

Concepts underlying the Hazard Analysis and Critical

Control Point (HACCP) system originated in the 1960s

to establish risk in quality assurance systems of the

National Aeronautics and Space Administration (NASA)

(1). The concept was presented to the public at the 1971

National Conference on Food Protection (1, 2). The Pills-

bury Company, in response to the food safety require-

ments imposed by NASA for space foods produced for

manned space ﬂights, pioneered using HACCP principles

for food protection and published the ﬁrst text on the

subject in 1973 (3). While interest was strong in the early

1970’s, HACCP was not widely incorporated into the food

industry, in part due to lack of guidance and difﬁculties

industry encountered in implementation. The 1970s were

a decade of trial, error, and debate between the regulatory

agencies and industy. However, in 1985 a Subcommittee of

the Food Protection Committee of the National Academy

of Sciences (NAS) resolved the frustrations of all with a

report (An Evaluation of the Role of Microbiological

Criteria for Foods and Food Ingredients) that included a

strong endorsement of HACCP as a preventive system for

control of microbiological hazards (4). As a result of this

recommendation, the National Advisory Committee on

Microbiological Criteria for Foods (NACMCF) was formed

in 1987 to advise the Secretaries of Agriculture, Com-

merce, Defense, and Health and Human Services (5). In

1989 NACMCF published its ﬁrst major document even-

tually entitled HACCP Principles for Food Production (6),

which was updated in 1992 and published as an article

with the title ‘‘Hazard Analysis and Critical Control Point

System’’ (7). Another revision of the document took place

in 1997, and the document was published in 1998 as an

article with the title ‘‘Hazard Anlaysis and Critical Con-

trol Point Principles and Application Guidelines’’ (8).

The NACMCF documents have formed the basis for

HACCP application in the United States. On the interna-

tional level, the Codex Alimentarius Committee on Food

Hygiene has worked in concert with the NACMCF to

develop HACCP principles and guidelines for application

worldwide. In 2003, Codex adopted the latest version of its

HACCP document (9), which shares many similarities

with the 1997 NACMCF document. The NACMCF docu-

ments were used by the USDA Food Safety and Inspection

Service (FSIS) in promulgating HACCP regulations for

meat and poultry products (10) and by the Food and Drug

Administration (FDA) in promulgating HACCP regula-

tions for seafood products (11) and for juice products (12).

The ﬁrst commodity to be regulated by HACCP was

seafood based on ﬁnal regulations published in the

Federal Register on December 18, 1995, which became

effective on December 18, 1997. The regulations require

processors of ﬁsh and ﬁshery products to have on ﬁle a

HACCP plan for their operations. Based on the ﬁnal

regulations, FDA publishes Fish & Fishery Products

Hazards & Controls Guidance every two to three years

(13). A new Seafood Hazards Guide is expected in 2008

which will include a number of additions and modiﬁca-

tions, a few of which are expected to be:

. Addition of a chapter on post-harvest processing

technologies (i.e., hydrostatic pressure, Individually

Quick Frozen (IQF) with extended cold storage, mild

heat, and irradiation) affecting pathogen survival in

mollusks.

. Information will be detailed on time–temperature

indicators (TTIs) and how they must function.

. Implementation of HACCP comments from the Sea-

food HACCP Alliance for Education and Training

based on their program for the seafood industry

and federal, state, and local inspectors.

. More guidance on how to address allergens.

. Recommendations on calibration of temperature-sen-

sing devices.

HACCP CONCEPT

HACCP is a preventive system for assuring the safe

production of foods. It is based on a common-sense applica-

tion of technical and scientiﬁc principles to the food

production process and system of distribution. The princi-

ples of HACCP are applicable to all phases of food produc-

tion, including basic agriculture, food preparation and

handling, food processing, food packaging, food service,

distribution systems, and consumer handling and use.

The most basic concept underlying HACCP is the

prevention of occurrence of hazards rather than the

inspection of ﬁnished products for the hazards. A grower,

processor, or distributor, when they have gathered sufﬁ-

cient information concerning that segment of the food

system, should be able to locate where and how a food

safety problem may occur. If the ‘‘where’’ and ‘‘how’’ are

known, prevention is simpliﬁed. A HACCP program deals

with control of food safety hazards originating from the

ingredients, product processing, and packaging. The goals

of the program are to make the product safe to consume

and

be able to prove that the product presents minimum

risk

to consumers.

The where and how are included in the

HA (hazard analysis) part of HACCP. The focus on control

of processes and conditions is included in the second

567