WWW.PHARMAMANUFACTURING.

COM VOLUME 11, ISSUE 1

Novartis Edges out Pzer for
Digital Buzz p.9
Working With Vent Filters:
A Risk-Based Approach p. 27
Can Architects and
Process Developers
Row Together? p. 32
Ciurczak on Meaningful
Sampling p. 50
J
A
N
U
A
R
Y

2
0
1
2

Modular construction and disposable
process equipment are maximizing
agility and minimizing risk
Groundbreakers
Zeta Plus Activated Carbon
Cartridges and Capsules
Zeta Plus Activated Carbon media incorporates our latest activated carbon
technology to meet the needs of pharmaceutical manufacturers by
decolorizing and reducing contaminates from the process stream. It can be
used in any application where bulk activated carbon is used.
Zeta Plus Activated Carbon cartridges and capsules are available in a range
of sizes from laboratory-scale to process development through to full scale
production.
3M Purication Inc.
3M Purication Inc.,
400 Research Parkway,
Meriden, CT 06450 U.S.A.
3M is a trademark of 3M Company.
Zeta Plus is a trademark of
3M Company used under license
3M 2010. All rights reserved.
Applications
Pharmaceuticals

Decolorization in production
of active pharmaceutical
ingredients

Parenterals

Blood fractionation

Catalyst reduction from API
To learn more about how 3M Purication Inc. ltration products
can improve process efciency and save costs, contact us at:
www.3Mpurication.com or call 1-203-630-4574
Sustainability
At DSM, our purpose is to create brighter lives for people today and
generations to come. This mission is supported by sustainability as
a core value and one of four pillars in our Quality for Life

commitment.
Its philosophies and metrics are evident in everything we do, highlighted
by a top ranking in the Dow Jones Sustainability Index in the global
chemical industry for 10 consecutive years. Sustainability is also
an increasingly valued criterion for vendor selection, so its not
only a responsible approach, but a strategic business driver.
DSM Pharmaceutical Products
45 Waterview Boulevard, Parsippany, NJ 07054-1298 USA
Tel: +1 973 257 8011
www.dsmpharmaceuticalproducts.com
www.dsm.com
Visit us at Informex booth #1101
Quality
Reliability
Traceability
Sustainability
MULTI-TIP TOOLING MULTIPLIES PRODUCTION
In todays economy, everyone is looking to cut costs. In order to thrive in the tablet
manufacturing industry of tomorrow, companies must simplify their processes to reach
ultimate efciency. By adopting multi-tip tooling, tablet manufacturers can dramatically
reduce the number of presses, tooling sets and operators needed to achieve the same
production output as traditional single-tip tooling.
Multi-tip tooling is available in two common confgurations: assembly (or multi-piece)
and solid. Both styles have their advantages, and our tooling experts can help you
determine which is right for you. Contact us today to learn more about the many benefts
of multi-tip tooling. You can conveniently fnd detailed information and request a free
quote by visiting our website, or speak directly with a live representative now!
Scan the QR code with your smartphone
or visit natoli.com/multi-tip to request a
free quote in less than fve minutes.
Departments
7. FROM THE EDITOR
Can Pharma Conquer the Culture of Arrogance?
It had better start benchmarking. Analysts say 2012
could be the industrys toughest year yet.
BY AGNES SHANLEY, EDITOR IN CHIEF
9. DIGITAL INSIGHTS
For Digital Buzz, Novartis Tops Pfzer
And J&J re-enters the top ten despite continued
product recalls.
BY MICHELE V. WAGNER, SENIOR EDITOR, DIGITAL MEDIA
10. UPFRONT
Virtual quality inspections; troubleshooting R&D;
Funny Pharm; Compliance Quiz; pharma quotes
15. OUTSOURCING EXCELLENCE
Modular Gains Momentum
Modularly constructed facilities will aid outsourcing
and emerging markets, says expert Pr Almhem.
32. FACILITY DESIGN
Finding the Flow
Can architects and process designers see eye to eye?
BY ALAN A. LIDDY, AIA, NCARB, PMP, SSOE GROUP
41. PHARMA VIEW
Can Manufacturing Be Sexy for Millennials?
If not, maybe high-paying jobs will do the trick.
BY PAUL THOMAS, SENIOR EDITOR
49. CLASSIFIEDS
50. THERAPEUTIC DOSE
Sampling: Good News, Bad News
Meaningful testing will satisfy both FDA and ASTM.
BY EMIL CIURCZAK, CONTRIBUTING EDITOR
Features
27. A RISK-MANAGEMENT-BASED APPROACH
TO TANK VENT FILTRATION
Best practices to ensure compliance, and avoid prob-
lems during operation, installation, CIP and SIP.
BY MICHAEL FELO, EMD MILLIPORE
34. TECHNOLOGY ROUNDUP: MACHINE VISION
Machine vision technologies are getting simpler and
cheaper, and yet tackling more varied applications.
BY PAUL THOMAS, SENIOR EDITOR
37. IMAGING THE BLENDING PROCESS
Hyperspectral imaging can be used to optimize blend-
ing, by monitoring the distribution of excipients and
APIs in formulation.
BY GABOR KEMENY AND GINA STUESSY, MIDDLETON RESEARCH
43. SPLIT DECISIONS: TABLET SCORING COMES
UNDER SCRUTINY
FDA has released new draf guidance; we summarize
and talk with tableting expert Dale Natoli.
BY PAUL THOMAS, SENIOR EDITOR
45. BOOSTING PRODUCTION LINE RESULTS
Forget about conventional wisdom and take advantage
of employees natural working rhythms.
BY TOM MCNAMARA AND SARAH HUDSON, RENNES SCHOOL OF BUSINESS,
AND SABRY SHAABAN, GROUPE ESC LA ROCHELLE
Pharmaceutical Manufacturing (USPS number 023-188) is published monthy except bi-monthly in July/Aug and Nov/Dec, by Putman Media Inc. (also publishers of Food Processing, Chemical Processing, Control, Control Design, and
Plant Services), 555 W. Pierce Road, Suite 301, Itasca, IL 60143 (Phone: 630-467-1300 Fax: 630-467-1179). Periodicals postage paid in Itasca, IL and at additional mailing ofces. POSTMASTER: send change of address to Pharmaceutical
Manufacturing, Post Ofce Box 3431, Northbrook, IL 60065-3431. SUBSCRIPTIONS: To receive a complimentary subscription go to www.pharmamanufacturing.com. Subscription rate for non-qualied U.S. subscribers is $68/yr. Single
copy rate is $15.00. Foreign rate is $115/yr. (surface mail) and $200/yr. (airmail). Copyright 2012 by Putman Media Inc. All rights reserved. The contents of this publication may not be reproduced in whole or in part without consent of
the copyright owner. Reprints are available on a custom basis. For a price quotation contact reprints@putman.net. Subscriptions/Customer Service: (888) 644-1803
43
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM ANUA|Y 2012 S
Cover Feature
18. GROUNDBREAKERS
Driven by modular construction and
disposable process equipment, facility designs
now aim to maximize agility and minimize risk.
BY AGNES SHANLEY, EDITOR IN CHIEF, AND PAUL THOMAS, SENIOR EDITOR
18
I NSI DE
Fast, Sterile
and Precise Filling
3800 CAMP CREEK PARKWAY
BlLDlNG 200 SlTE 120
ATLANTA, GECRGlA 30331
TEL: 78-553-3400 FAX: 78-553-345 www.gemu.com
GEMs 660 diaphragm
valve is designed for
dosing and lling
in rotary and linear
lling machines.
GEM 660
Featuring:
High speed and
high cycle applications
Unique actuator concept
with low air consumption
and extended diaphragm
cycle life
High repeatability
and reliability
Integrated opening and
closing stroke adjustment
Adjustable volume and
lling speed
Multi-port valve option
offers two-stage ow
CIP/SIP suitable
Can be used for cyclical
sterilization up to 150C
I RECENTLY invited a prominent fgure in academia to
write a commentary for this magazine on what the indus-
try needs to do in the year ahead. He has been a leading
advocate for manufacturings strategic importance, and
for the need for improvement and change.
He declined, but his reasons were revealing. I have to
be careful about what I say, to avoid alienating leadership
in the industry, he said. Apparently, some previous
writings had angered people at companies whose
participation was critical to research.
Tis surprised me. Wouldnt executives want
constructive criticism from a neutral source? Afer all,
isnt the end goal continuous improvement? He is a
professor and was discussing science, nothing more and
nothing less. Tere was no political agenda.
Tis reaction has been around for a while. It started
before and has continued long afer the famous potato
chips and soap fakes comment by former FDA
commissioner McClellan. Sure, some pharma companies
are embracing best practices from other industries. GSK,
for instance, has formed a joint venture with the McLaren
Group, of Formula One fame, with the goal of applying
modeling, analytics, engineering and technology to drug
development and manufacturing. But even with projects
like this going on, and all the change of the past decade,
the Not Invented Here syndrome is alive and well in the
drug industry, we hear.
Apparently, back in the early days of Process Analytical
Technology (PAT) and CDERs Science Advisory
meetings at FDA, at least one key executive at one of the
worlds largest pharmaceutical frms openly questioned
the need for the industry to change its practices or
benchmark them against those of automotive, electronics
or consumer goods manufacturers. We could teach them
about better manufacturing, he reportedly said.
Now, Im sure that drug companies could teach those
industries everything about making drugs. But how
about waste and cycle time reduction? In biotech ventures
involving electronics giants Fuji and Samsung, we may
see what happens when the best of both worlds collide.
Change is clearly coming. Fitch Ratings has predicted that
2012 will be the industrys most challenging year ever. Can
anyone aford to be arrogant, when R&D pipelines are so
thin and drug manufacturing, as it is practiced today, has
been acknowledged to waste over $50 billion per year?
According to Emory University professor Jagdish
Sheths book, The Self-Destructive Habits of Good
Companies, highly successful companies lose their edge
through complacency and arrogance. Perhaps the same
holds true for entire industries. Sheth wrote, Te most
dangerous competition comes from low-quality/low-price
competitors. Utilizing price as their most tantalizing
selling point, they establish a presence in the marketplace.
Teir upstream competitors generally malign them as
easily dismissed peddlers of junk or just ignore them.
But if these inferior competitors improve quality while
maintaining their relative cost advantage, they become
irresistible value propositions to customers. Sheths book
came out over four years ago, and this transformation has
already begun within pharma.
Not Invented Here needs to be torn out by its roots.
Te smartest companies are learning from every source
possible. Tis new year, one of our resolutions is to include
more practical benchmarking examples from outside of
pharma that can be applied in your facilities. We also
hope to get more case studies directly from you and from
independent academic groups.
In this issue, for instance, youll see an article on
operational improvement, written by professors at the
Rennes School of Business in France. We also promise to
get more input from generics manufacturers, who face
competitive margin pressures closer to those of other
manufacturing industries.
Te future belongs, not only to the agile, but to those
who know how to learn. Heres to a good year!
Agnes Shanley, Editor in Chief
ashanley@putman.net
FROM THE EDI TOR
A Culture of Arrogance
Analysts say that 2012 may prove to be pharmas most challenging year ever.
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM !ANUA|Y 2012 7
AS MARGINS THIN AND COMPETITIVE
P|ESSU|ES lNC|EASE, lS THE|E ANY
ROOM FOR COMPLACENCY?
1MBUJOVNt1BMMBEJVNt3IPEJVNt3VUIFOJVNt3IFOJVNt(PMEt4JMWFS
Sabins precious metals recovery and rening
processes raise environmental protection to an advanced
state for safety and regulatory compliance. Your
advantage: assured peace of mind, and our promise
of maximum value returns from your precious metal-
bearing materials.
Learn more at sabinmetal.com
Think globally. Act responsibly.
Processing facilities and technical service offices in the Americas, Europe and Asia
ZETA INTERACTIVE, a digital marketing agency, has published its annual
end-of-year Buzz Report, recapping the most buzzworthy pharmaceutical brands
trends and stories from 2011. Zeta uses its Buzz technology and algorithm, which
scans more than 200 million blogs and online sites in real-time, and assigns a buzz
ranking to each company, factoring in both volume and tone of posts each brand
receives. From that, it determines digital winners and losers.
Te most noteworthy fnding was Novartis overcoming Pfzer, who previously
held the best buzzed pharma brand in 2010, for the most positively tonal web buzz
score (6.46). Novartis was not without its fair share of job cut news this year, but
perhaps it was Pfzer appearing in the news for thousands of job cuts, claims of
advertisements that objectify women, or shake-ups in top management that allowed
Novartis to slide into the top position.
Johnson & Johnson, which fnished as the #1 best buzzed pharma brand in 2008
and 2009 before dropping out of the top ten completely last year due to its recall crisis,
managed to make it back on this years list, fnishing at #6 overall. However, at just
61% positive, J&Js tonal buzz ranking was lower than any other brand on this years
pharma buzz list. And as expected, the product recalls of last year still continue to
be featured prominently in J&Js cluster analysis, with the word recall appearing
among the most popular words used to describe the brand online in 2011.
Surprisingly, Bayer climbed up four spots to the #2 spot overall on this years list
afer fnishing in 6th place in 2010. Additionally, at 88% positive tone, Bayers tonal
buzz ranking was higher than any other pharma brand on the list. Other interesting
Buzz Rankings to note:
- Pzer's buzz was closely associated with Advil, as was Bayers with aspirin
- Eli Lilly was viewed as a beacon of employment and growth
- Along with J&J at #6, Teva entered the top ten at #9
- Watson and Roche dropped out of the top ten
Do you think these rankings are accurate and meaningful: Let me know:
mvaccarello@putman.net.
DI GI TAL I NSI GHTS
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM ANUA|Y 2012
MICHELE VACCARELLO WAGNER, SENIOR EDITOR, DIGITAL MEDIA
555 West Pierce Rd., Itasca, IL 60143
Phone. (630) 467-!300 lax. (630) 467-!!79
www.putmanmedia.com
Subscriptions/Customer Service
(888) 644-!803 or (847) 559-7360
EDITORIAL TEAM
AGNES SHANLEY EDITOR IN CHIEF
ashanley@putman.net
PAUL THOMAS SENIOR EDITOR
pthomas@putman.net
MICHELE V. WAGNER SENIOR EDITOR
mvaccarello@putman.net DIGITAL MEDIA
KEITH LARSON V.P., CONTENT
klarson@putman.net
EDITORIAL ADVISORY BOARD
ALI AFNAN, Step Change Pharma
JIM AGALLOCO, Agalloco & Associates
CARL ANDERSON, Duquesne University
JAMES BLACKWELL, Bioprocess Technology Consultants
JOHN BLANCHARD, ARC Advisory Group
TOM CAMBRON, P&G Pharma
JAMES CHENEY, Novartis
BIKASH CHATTERJEE, Pharmatech Associates, Inc.
EMIL CIURCZAK, Cadrai Group
ROBERT DREAM, HDR Company
ERIC LANGER, BioPlan Associates, Inc.
ROBBE C. LYON, lDA
IVAN LUGO, INDUNIV, Puerto Rico
GIRISH MALHOTRA, Epcot International
RODDY MARTIN, AMR Research
FERNANDO PORTES, Stevens Institute of Technology
GARY RITCHIE, Consultant
DESIGN & PRODUCTION TEAM
STEPHEN C. HERNER V.P., CREATIVE SERVICES
sherner@putman.net
DEREK CHAMBERLAIN ART DIRECTOR
dchamberlain@putman.net
RITA FITZGERALD PRODUCTION MANAGER
rtzgerald@putman.net
ADMINISTRATIVE TEAM
JOHN M. CAPPELLETTI PRESIDENT/CEO
JULIE CAPPELLETTI-LANGE VICE PRESIDENT
JACK JONES CIRCULATION DIRECTOR
USPS number (023-188)
Novartis Beats Pfizer in
2011 Web Buzz Rankings
Brand
Volume
Ranking
Tonal
Ranking
Zeta Buzz
Ranking
Change From Last
Years Rankings
1. Novartis 75.1 86/14 6.46 +1
2. Bayer 71.3 88/12 6.27 +4
3. AstraZeneca 62.8 85/15 5.34 No Change
4. Merck 65.1 76/24 4.95 +1
5. Pzer 74.8 66/34 4.94 -4
6. Johnson & Johnson 75.3 61/39 4.59 New to list
7. Bristol-Myers Squibb 51.7 87/13 4.50 No Change
8. Eli Lilly 44.5 84/16 3.74 +1
9. Teva Pharmaceuticals 35.1 80/20 2.81 New to list
10. Mylan 33.4 84/16 2.80 No Change
Its now part of the SOP to send the recall notice
with each consignment. This may well be our
best cost saving measure. Atul Deshmukh
Funny Pharm comics, drawn by professional cartoonist Jerry King,
appear twice a month on PharmaManufacturing.com. Readers submit
suggested captions. Above is a recent cartoon and winning caption.
FUNNY PHARM
JUST HOW far has pharma R&D productivity fallen?
According to three Oliver Wyman analysts, very far
indeed. Teyre basing their conclusion on the number
of new drug approvals vs. the research dollars spent over
the past several decades. But theyre also factoring the
true value of new drugs launched each year.
In Beyond the Shadow of a Drought: Te Need for
a New Mindset in Pharma R&D, analysts Jef Hewitt,
J. David Campbell, and Jerry Cacciotti state: Its well
accepted that pharmaceutical R&D productivity has
fallen, with new drug approvals trending downward even
as costs trend up. Companies are taking some actions:
rationalizing costs, increasing outsourcing, collaborating
with academic institutions, increasing their focus on
specifc disease areas, reconfguring their organizations.
But is that enough? We dont think so. And neither,
apparently, does Wall Street. Investors remain wary
of R&D spending, rewarding companies that cut and
penalizing those that donta sign of limited confdence
in the industrys use of its capital.
Sure, its hard to be too alarmed when drug companies
are still seeing respectable revenue growth, but
fundamental problems are masked, they say.
- Drug companies are, to a degree, victims of their own
success. Tey have met many of what were formerly
viewed as unmet needs, and many of these important
medicines are now available in cheap, generic versions.
- Payers are exercising their purchasing power, helping to
rein in drug prices.
- Getting new drugs to market is even harder than
conventional wisdom believes. Oliver Wymans own
analysis suggests the likelihood that a new drug
entering Phase III will reach the market is just 50
percent, far less than the success benchmarks many
companies commonly use.
- Pharma drugs aren't fungible. (Stated differently,
pharmaceuticals do not lend themselves to substitu-
tion). Companies are realizing they cant compete
in varied therapeutic areas, lessening their ability to
conduct a more shots on goal approach to scoring
with given molecules.
e authors suggest several xes for the above
problems. Te key, they say, is to think diferently about
innovation:
- Focus on efficacy. Tis reverses the classic approach,
they write, which targeted the broadest population in
which the drug had a statistically signifcant (if mar-
ginal) result.
- Invest more up-front. Companies need to spend more
to fully understand where and for whom their drugs
work. (Editors Note: Sounds like a ringing endorsement
of Quality by Design.)
t&NQMPZUIFi&BTZ#VUUPOwBQQSPBDI Even small im-
provements in safety and ease of use for patients, as
Merck did with Januvia, can pay of.
Ultimately, the authors say, drug companies must
reconsider the importance, or defnition, of speed to
market." ey use the example of Pzer's missteps in
pursuing a new lipid therapy as Lipitors patent window
was closing. In the end, Merck chose a slower approach,
and produced a drug with fewer side efects and greater
success. Rather than massing investment against a single
shot, could [Pfzer] have accelerated back-up molecules
to determine if they had a diferent safety profle? the
UPFRONT
R&D Productivity: Worse Than We Thought?
Drug companies are, a new report says, victims of their own success.
10 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
BY PAUL THOMAS, SENIOR EDITOR
Your Photonics Partner
SPECTROMETERS LASERS TOTAL SOLUTI ONS
Call 1-302-368-7824 or visit us
at www.bwtek.com/nanoram
Learn more!
Ideal Solution For:
t*ODPNJOH.BUFSJBM*EFOUJDBUJPOBOE7FSJDBUJPO
t"UMJOF4BNQMJOHBOE'JOBM*OTQFDUJPO
t$PVOUFSGFJU%SVH%FUFDUJPO
N
E
W
RAM
HANDHELD RAMAN SPECTROMETER
to prevent BIG PROBLEMS.
T
h
e sm
a
llest w
a
y
CFR 21 Part 11 Compliant Software
and API Spectral Libraries Available
RAM
HANDHELD RAMAN SPECTROMETER
The NanoRam is the most sensitive and repeatable handheld spectrometer ever
designed to identify harmful non-conforming materials before they reach production.
report asks. More radically, could they have used their leading position to
strike a partnership with Merck or Roche to share the investment and risk
across CETP inhibitors, and decrease the chance that they might miss out
on a major therapy breakthrough in a chosen focus area?
For a link to the report, see PharmaQbD.com.
Drug Manufacturers Seek
Remote Control
AS FACILITIES of the future become more far-f lung and distant
from drug manufacturers home bases, and as outsourcing continues to
increase, the challenge of monitoring varied and vast activities around
the globe becomes greater. One obvious solution for this dilemma is to
make use of remote videoconferencing capabilities. Videoconferenc-
ing has long been used in boardrooms and amongst researchers, and is
gaining traction as a means of ensuring manufacturing consistency and
quality worldwide.
FDASmart Inc. is one solution provider banking on manufacturers
need for greater, faster global oversight. Its SmartInspect system enables
live, secure, Internet-based, mobile video transmissions between two
locales. In other words, says founder and CEO Ram Balani, complete
global visibility without travel and without a boatload of money.
SmartInspect has garnered interest from Pfzer, which is testing the
system for various purposesassisting technology transfer, performing
quality audits remotely, and for remote training.
Te drug company has experimented with using the system for API
screening at sites in China, for instance, notes Balani. Whereas Pfzer
might typically send three people abroad for a lengthy trip to perform
such screening, it could potentially do the entire task via videoconference.
Bristol-Myers Squibb is also doing Proof of Concept testing of
SmartInspect to facilitate tech transfer for biologicsfor sharing
complex information between, for example, Syracuse, New York, and a
contract manufacturer such as Celltrion in South Korea.
Having a direct link to our plants and subject matter experts
around the world is obviously advantageous, says Kirk Leister,
director of process analytical sciences for BMS. Part of Leisters job is
to find technological innovations that can facilitate development and
manufacturing.
Leister uses the example of having spent two months troubleshooting a
peptide map in a Korean facility, eventually sending an expert to the site,
only to fnd out that the problem was related to a fairly simple setting.
Being able to set up and operate instrumentation via mobile video would
have made a signifcant diference, he says.
SmartInspect users pay a $24,000 fee upfront for the equipment and
sofware, and may sign up for a maintenance contract as well. While BMS
and Pfzer have shown interest, Balani envisions his technology being used
more by small and mid-size companies with limited resources, and by
clinical trial teams. Paul Thomas
UPFRONT
12 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
PHARMA REPLAY
Most drug shortages we re-
viewed in detail were report-
edly caused by manufacturing
problems.
U.S. Government Accounting
Ofce (GAO) investigators, on the
root causes of recent chronic drug
shortages
This limited product availability
does not foreshadow the poten-
tial for any additional supply of
Doxil in the immediate future, as
we have no further information
from BVL on when manufactur-
ing will resume.
A letter from Janssen to health-
care providers regarding future
shipments of Doxil following
the halting of production at Ben
Venue Labs.
There is no proceduralized pro-
cess to notify senior management
of issues that could potentially
impact the safety or quality of
product.
One of many faults cited by FDA
in a recent warning letter to Ben
Venue
We apologize for any inconve-
nience this may have caused.
Facebook, citing an adminis-
trative error that allowed the
Facebook page of the Germanys
Merck KgaA to be used by Merck
& Co. of the U.S.
The problem with R&D is its
not always consistent. Its not
like engineering where you can
incrementally innovate and make
another version of the iPhone.
Merck CEO Ken Frazier, justifying
his commitment to maintaining
R&D spending
USP Proposes
Guidelines for
Supply Chain
Integrity
THE U.S. Pharmacopeial Conven-
tion has proposed a set of recom-
mended best practices that will
help ensure improved supply chain
security and integrity. (USP is cur-
rently seeking feedback; www.usp.
org/USPNF/notices/generalChap-
ter1083.html.)
The standards are contained
within the proposed USP General
Chapter <1083> Good Distribution
PracticesSupply Chain Integrity.
The proposal is intended to serve
as general guidance for essential
elements of an effective supply
chain strategy.
Tere is incentive for all players
in the pharmaceutical industry
large and small companies,
regulators and standards-setting
bodiesto come to some agreement
on hot-button issues such as
track and trace technology and,
at the larger level, to codify what
constitutes a solid, universal
approach to global supply chain
integrity, says Praveen Tyle, Ph.D.,
chief science of cer for USP. USP
has developed an initial proposal
that we expect to evolve as industry,
FDA and others weigh in. Our role
as an independent body provides
an opportunity to convene all
these parties and advance this
critical issue. . . . USP can move
forward something more concrete
than a technical report, as part of
a mechanism that can be regularly
updated to best meet the needs of all.
Te proposed standard covers four
main areas:
- Importation
- Counterfeit Drugs and Medical
Devices
- Best Practices to Combat Counter-
feit Drug and Medical Devices
- Diversion and e
e dra general chapter and
comments submitted to USP will be
discussed at a Supply Chain Integrity
Workshop that USP is hosting on
May 22-23, 2012, in Rockville,
Maryland.
UPFRONT
PHARMACEUTICAL MANUFACTURING
Biologics Modulars pre-constructed,
pre-tested CGMP manufacturing
facilities can be delivered in a
matter of months.
A Biologics manufacturing
platform is expandable and
grows as you grow.
Join the control group with
a Biologics facility.
Rapid-Deployment

CGMP Manufacturing Facilities

CONTROL YOUR SPACE. CONTROL YOUR COSTS. CONTROL YOUR TIME.
Inquiries: Clark Byrum, Jr.
President/CEO
317.456.9191
cbyrum@BiologicsModular.com
JOIN THE
CONTROL GROUP
Leasing available through MEI Healthcare Capital. Patent pending.
Welcome to Compliance Quiz. (Find each months quiz,
and more details on answers, on PharmaManufacturing.
com.) Januarys quiz focuses on some general rules of
good GMP compliance. For answers, see below, right.
1. Which three criteria are FDA investigators most likely
to focus on during GMP inspections?
A. Your ability to debate the broad language used in
FDA guidance documents, refreshments provided, and
your sense of humor.
B. Process validation, lab/facility operational response to
OOS results, your rms response to any observances
of deviations.
C. Cost savings from shortcuts in your quality system, lab/
facility dress code, the kindness of the receptionist.
2. 21 CFR Part 211.22 (d) states: The responsibilities
and applicable to the quality con-
trol unit shall be in ...
A. software & compliance
B. accessories & arrears
C. procedures & writing
D. regulations & an electronic format
3. Which of the following is NOT an example of an ob-
servance of a deviation?
A. Inadequate process failure investigation.
B. Responsibility of Quality Control Unit not document-
ed or followed.
C. Laboratory controls do not include scientically sound
test procedures to assure that drug products conform
to standards of identity, strength, quality and purity.
D. Examination and testing of samples is not done
frequently enough to improve the likelihood that in-
process materials conform to specications.
4. One of the most critical aspects of a successful Form
483 response is that it:
A. is quick and indicates the immediate x (for the devia-
tion) to be implemented in the shortest amount of time.
B. comes from one single person in the rm, wherein each
observation is addressed by one person in one step.
C. goes beyond addressing the Forms observations, but
determines root causes and underlying issues behind
the observed deviation.
D. leaves an overall impression of self-righteous indignation.
5. An effective corrective and preventive action (CAPA)
system of procedures must entail:
A. The precise details of one corrective and one preven-
tive action taken after a single defective batch.
B. An assumption that once the corrective action is tak-
en, the problem is solved for the foreseeable future.
C. A dynamic program for continuous reappraisal of pro-
cesses and test methods that will determine areas that
are at risk for producing an unacceptable product.
UPFRONT
14 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
A n s w e r s
1 . B 2 . C 3 . D 4 . C 5 . C
Compliance Quiz
Sponsored by
www.Vaisala.com
ICH Fleshes Out Q8,
Q9, and Q10
THE ICH Quality Implementation Working Group has
released its fnal three Points to Consider (PtC) docu-
ments, following three it released last June. Te docu-
ments are intended to provide further clarity in regards to
ICH Q8, Q9, and Q10. Te six PtCs are:
- Criticality of Quality Attributes and Process Parameters
- Control Strategy
- Level of Documentation in Enhanced (QbD) Regulatory
Submissions
- Role of Models in Quality by Design
- Design Space
- Process Validation/Continuous Process
Verication
Put those welding torches away and pack up that old roll groover
too. Now, there is a better way to connect stainless steel pipewith
Viega ProPress

. Press technology is faster, safer and more consistent

compared to traditional methods. Well prove it! Contact Viega for
more information or for a demonstration on the future of pipe joining.
The Torch Is Past
(and so are welding and grooving)
TheTorchIsPast.com
866.766.7805
The gl obal l eader
i n pl umbi ng, heat i ng
and pi pe j oi ni ng syst ems
For bio-pharmaceutical piping applications, Viega ProPress
for stainless steel offers substantial savings in man-hours when
compared to welding and grooving. The system also provides a
safer, cleaner, higher-quality installation.
A.W. Stiles Contractors recently installed Viega ProPress
for stainless steel at a facility that manufactures personal
care products such as mouthwashes, shampoos and baby
powder. The upgrade involved switching from PVC piping
to stainless steel for the peroxide lines.
CUSTOMER: Health/beauty products manufacturer
APPLICATION: Chemical transport lines
LOCATION: Smyrna, TN
CONTRACTOR: A.W. Stiles Contractors
Viega ProPress

for stainless
saves labor, enhances safety
for Bio-Pharm applications
The Viega ProPress system allowed us to pre-run
all the pipe, except for the connections at each
end, parallel to the existing lines while production
still ran. Viega can be installed so fastwe were
able to do the fnal cut-ins during lunches, so they
had very little downtime.

Prior to the new installation, whenever a different percentage

of alcohol was used in the product, the lines had to be
fushed. That resulted in about $20,000 a month in waste
materials. With the upgrade, not only was the fushing no
longer necessary, but the problem of melting PVC lines was
also eliminated.
According to Tommy Stiles, it would have cost $200,000 to
weld all of the stainless piping. That amount didnt include
the lost production from downtime. The project ended up
costing only $60,000 with A.W. Stiles using Viega ProPress
for stainless. The adaptability of the system allowed the
contractor to integrate the new piping with the existing pumps
and other equipment in the customers system.
Downtime was not an option, Stiles said. The Viega
ProPress system allowed us to pre-run all of the pipe parallel
to the existing lines, except for the connections at each end.
And, we did it while production still ran. The system can be
installed so fast, we were able to do the fnal cut-ins during
lunches, so they had very little downtime.
Pressing was fve times faster than if we had welded all of
the joints, Stiles estimated. Of all the joints we installed,
not one leaked. Thats almost unheard of with other joining
methods. In some environments, welding is not allowed
especially with the alcohols and other fammable materials
running through the lines.
To learn more about Viega ProPress for stainless in
Bio-Pharm applications, visit TheTorchIsPast.com
or call 866.766.7805.
OUTSOURCI NG EXCELLENCE
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM !ANUA|Y 2012 1S
PR ALMHEM has been involved in modular con-
struction for several years, previously at Pharmadule,
and, now, with the recent startup ModularPartners. As
the pharmaceutical industry increases its use of modular
construction, new players are entering the market. At
the same time, designs are becoming more standardized,
which promises to reduce project timelines even further.
Will pharma embrace modular to the same extent that
other industries have? Almhem recently shared his per-
spectives on mod con, where it is today, and where it is
going, in the pharma and biopharma industry.
PhM: Modular construction continues to grow in popu-
larity within pharma and biopharma. How far along are
we in terms of a paradigm shif?
P.A.: I think we are still in an early stage. Modular solu-
tions have gone from being a novelty and something very
few would consider, to now being discussed and consid-
ered in many (most?) cases. However, few have started to
take advantage of the real opportunities in modulariza-
tion in the way that, for example, the sofware and elec-
tronics industries do. Tis requires a change toward much
more standardization of building blocks (not necessarily
the complete process or facility), use of confgurable sys-
tems rather than customized systems, etc. Only when we
adopt these strategies can we talk about a paradigm shif.
PhM: Whats been the most signifcant change youve
seen in the past ten years in regards to the industrys
understanding and acceptance of modular construction?
P.A.: Te biggest change is that in the last few years, most
end users have become aware of modular solutions and
more and more service provider are now saying they ofer
modular solutions. Modular solutions have become an
established methodology.
PhM: What have been the most signifcant hurdles to
adoption? Are there limitations to modular facilities
that you didnt foresee in the past?
P.A.: Te most signifcant hurdles have been the tradi-
tion of how projects are planned and executed. If their
full beneft is to be realized, modular solutions need to be
considered in the early stages of a project. Many times, if
the question about modularization has been asked, it has
been too late. Modular facilities have also had a reputa-
tion for being expensive. Tis has sometimes been true,
partially because. in the past, there hasnt been enough
volume to really drive down cost. Other limitations like
layouts and fexibility have, for the most part, been solved
by new engineering solutions.
PhM: Modular facilities require more intense, focused
initial planning. You previously wrote: Team building,
interface management, automation, and integrated
validation must be incorporated into the plan at the pre-
project planning phase. Has this been a challenge for
manufacturers?
P.A.: Tis is actually relevant for all fast-track projects,
not just modular projects. It has been a challenge for
owners, but there is an increasing understanding that this
is mostly a matter of good project management.
PhM: Are major manufacturers gravitating towards
modular design for emerging markets in particular, as a
way to better standardize, control, and maintain facili-
ties there?
P.A.: Yes, modular solutions are even more valuable in
these markets, for the reasons you mention. However,
in addition, modular constructions can mitigate risk for
project quality and schedule, and can enhance security
by allowing companies to keep intellectual property in
house and under control.
PhM: Have you collected data on time and cost savings
of modular vs. traditional projects? What specifcs can
you share?
Pharma Facilities: Modular Gains Momentum
Modularly constructed facilities are an inevitable part of the industrys future, especially for emerging markets, says Pr Almhem
o Mouu|a|Pa|L|J|s.
BY PAUL THOMAS, SENIOR EDITOR
PHARMA IS NOT TAKING ADVANTAGE
O| MODULA|lZATlON lN THE WAY THAT
SOFTWARE AND ELECTRONICS ARE.
P.A.: Tere are case studies and some statistics, but un-
fortunately very few really relevant, objective reports. CII
has done benchmarking, but only members have access to
this. Merck made a detailed benchmark for their Summit
S6 project that was one of the ISPE Facility of the Year
category winners this year. Tis was presented at the ISPE
Annual Meeting 2009 and published by ISPE.
PhM: Have there been advances/changes in how modules
are shipped and transported? Are they typically moved
in the same fashion as standard shipping containers?
P.A.: Building modules are generally shipped in a fashion
similar to shipping containers. Te main diferences are
that the modules are larger and shipped with greater care
(for instance, steps are taken to ensure weather protec-
tion). Tousands of modules have been shipped around
the world with very few incidents. Process modules are
shipped in similar ways, depending on size, sensitivity,
and other factors.
PhM: Finally, can you clarify the circumstances of your
leaving Pharmadule and helping to start Modular
Partners?
P.A.: Pharmadule AB, the Swedish parent company of
the Pharmadule group, was owned by a European private
equity frm. Afer a couple of difcult years in 2009 and
2010, they decided in early 2011 to close Pharmadule, and
Pharmadule AB fled for bankruptcy in February 2011.
Pharmadule AB is now closed. Pharmadule, Inc., the U.S.
subsidiary that I ran, is also closed. We did not fle for
bankruptcy but closed. Te other subsidiary, Pharmadule
O, the manufacturing plant in the Pharmadule group,
was taken over by local management in a management
buyout. Pharmadule O is now owned partly by manage-
ment, and partly by Telstar of Spain, and is core part of
the ModularPartners network.
Afer closing Pharmadule, Inc., I started a company
ModWave (www.modwave.com), with a colleague from
Pharmadule, Camilla Sivertsson. From that platform,
we took the initiative to form ModularPartners with
KeyPlants of Stockholm, Sweden, and Yonkers Industries
of Cary, NC. Please see www.modularpartners.com for
additional information.
KeyPlants is a spin-of from Pharmadule AB, formed
in mid 2010. Just like Pharmadule O, the company is
owned by management and Telstar. Yonkers Industries is
a privately held construction management frm.
OUTSOURCI NG EXCELLENCE
16 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
NEWS & NOTES
Amgen and Watson Pharmaceuticals will develop and
commercialize follow-on versions of several antibody
biotech drugs for cancer, excluding Amgens products.
Amgen will be in charge of development, production
and initial marketing, while Watson will provide as much
as $400 million for product development and contribute
expertise in commercialization and marketing.
Baxter and Momenta Pharmaceuticals will also collabo-
rate to develop and commercialize biosimilars. Baxter
will provide expertise in clinical development and bio-
logic manufacturing, while Momenta will add expertise
in analytics and product and process development.
Soligenix has begun developing a next-generation
anthrax vaccine, a modied anthrax toxin protein, with
Harvard University.
ImmunoGen and Eli Lilly will jointly develop antibody-
based cancer drugs. Lilly will get exclusive licensing
rights to some of ImmunoGens cancer drugs in ex-
change for a $200 million milestone fee for each drug
licensed and additional royalties.
DSM Pharmaceutical Products has contracted with New
Jersey-based QRxPharma to manufacture MoxDuo
capsules, an opioid pain medication, to supply the U.S.
market through 2015.
Accelrys has acquired VelQuest, which specializes in
paperless lab execution systems, for $35 million.
AAIPharma Services has acquired Celsis Analytical
Services, which performs material testing services for
pharma and biopharma. The new company will feature
an integrated development and material testing service
model, said Patrick Walsh, AAIPharma CEO.
Ben Venue Labs will extend the voluntary suspension of
manufacturing at its Bedford, Ohio facility. The company
said it can no longer continue to manufacture and re-
mediate simultaneously. Ben Venue hopes to get some
manufacturing begun this winter, but said that its north
facility, which makes key sterile injectables that are in
short supply on the market, requires major reconstruc-
tion and will not manufacture any new products before
the fourth quarter of this year.
PHARMACEUTI CAL COATI NG SYSTEMS
SENSI ENT

PHARMACEUTI CAL COATI NG SYSTEMS

A unit of Sensient Technologies Corporation, a leading global
manufacturer of colors, flavors and fragrances
2515 N. Jefferson, St. Louis, Missouri 63106
|
800.325.8110
Manufactured in the USA
2012 Sensient Colors LLC, all rights reserved. The SENSIENT trademark and the Sensient Technologies Corporation logo are owned and registered by
Sensient Technologies Corporation. All rights in trademarks are reserved.
*Image for comparison purposes only. Actual results may vary.
A Hi gh Sol i ds, HPMC Coati ng System
THE DIFFERENCE IS CLEAR
Sensient

Pharmaceutical Coating Systems leads the industry in the development and manufacture of innovative
coatings and colors for the pharmaceutical and nutraceutical industries.
Spectrablend II is a patented high solids, HPMC coating system that delivers a premium appearance, uniform
coverage and increased stability. Flexible and ready-to-use in 30 minutes, Spectrablend II offers manufacturing
advantages including ease of dispersion and the elimination of dust, foam or pan scuffing.
Servicing leading pharmaceutical companies from 35 locations globally, Sensients comprehensive range of
versatile and novel coating systems offers visual and functional attributes necessary for brand definition, product
identification and trademark protection.
Define and protect your brand with superior, high-quality coating systems. Visit sensientpharma.com.
2012 MARKS what some analysts are
already expecting to be the most chal-
lenging year in the pharmaceutical indus-
trys history. As patent expirations and global
competition step up, and the mantra changes from
blockbuster to nichebuster, manufactur-
ers are chipping away, wherever they can, at
the years and millions of dollars required to
launch each new product.
If they must fail, more pharmaceutical
manufacturers acknowledge, theyll need
to fail fast and move on quickly. Pharmas older
infrastructure has not always allowed this to happen.
Over the past few years, the pace of pharma plant
consolidations and closures has increased. In the U.S.
alone, 38 drug manufacturing facilities were shut down
last year, and 65 the year before, according to Pharmalots
Ed Silverman, who analyzed data from the Sugarland,
Texas-based research frm, Industrial Information
Resources, in December. However, he notes that 106 new
plants and laboratories, worth an estimated $4.3 billion
in contracts, are now planned or under construction.
Companies in pharmas newest sector,
biopharmaceutical manufacturing, are leading the
way in harnessing technologies and methods to reduce
the timeline and risk of building facilities to make
new products. Tis will be important, says BioPlan
Associates principal Eric Langer, since biopharm
companies will be increasing production
capacity by 25%, globally, over the next
few years. Biosimilars promise to
intensify the competition,
and companies are gearing up for production in fexible
facilities throughout the world.
Enabling pharma facilities transformation are modular
construction, which has been around for decades but
is now possible in shipping-container-sized units (Box,
p. 23), as well as disposable bioprocessing
equipment, standardized solutions allowing
for continuous operation and quick
validation.
At the same time, modeling and simulation
tools are becoming more important,
particularly for training and startup, as are standard
templates for IT and process control. For MedImmunes
new biopharmaceutical manufacturing facility in
Frederick, Maryland, which won ISPEs Facility of the Year
competition last year, best IT and automation practices not
always seen in traditional pharma led to improved results
(Box, p. 20).
Tis article will look at enabling technologies, some of
them already in action at existing and prototype facilities,
that suggest drug manufacturings future.
Enablers are combining in interesting ways that
promise to allow greater use of continuous processing
in the plants of the future. Peter Watler, CTO of Hyde
Engineering + Consulting (San Francisco), notes,
A simple 45-cm-diameter chromatography column
operating continuously could replace a complex and
costly 200-cm column operating once per batch, he
says. Tis will drive a need for enhanced unit operation
science and superb automation, but it will result in
reduced facility footprint, complexity and cost.
Tey are also shrinking design-to-production
timelines, which, Watler notes, have moved from 4-6
years a decade ago, to as little as 12 months and will
soon shrink, for smaller facilities, to just six months.
Eventually, he says, timelines will be dictated by shipping
and assembly times.
18 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
Modular construction and disposable
process equipment are maximizing
agility and minimizing risk
By Agnes Shanley,
Editor in Chief,
and Paul Thomas,
Senior Editor
At the same time, globalization has complicated the plant
construction picture. Despite progress with harmonization, there is
still some discontinuity of requirements from global regulators and,
especially in more remote locations, there are questions of local suppliers
and support, says Bikash Chatterjee, president of Pharmatech Associates
(Hayward, Calif), which is actively consulting in India and China.
Tere is also a need, Chatterjee adds, to make information management a part of the
plant design process from the very start. We should take into consideration developing and
applying IT infrastructure as part of facility design, assuming a global supply chain, he says.
In addition, he says, plant layout must be optimized early in the drug development process, to
foster collaboration within the company and across cultures, which can be an issue for some Asian
societies. Multicultural operations require collaboration, he says. In Asia, the focus is ofen on
doing the opposite. Most of ces and conference rooms have card key or key code access. Designing
spaces that foster collaboration and integrate telecommunication technology that can bridge cultural
diferences requires thought.
State of the art facilities in the future, Chatterjee says, will need:
- social buildings" that foster interaction and team-based research and development
- open and closed laboratory designs
- exibility to accommodate changes
- design for technology to provide access to electronic communication systems
- location in science parks to facilitate partnerships between government and the private sector
THE FUTURE? MODULAR AND PLASTIC
For now, though, observers agree that many, if not most, of pharma's new facilities will be modular. De-
spite the technologys higher construction and transportation costs, manufacturers should continue to
be attracted by the fact that it allows for smaller facility footprints and allows construction to progress
in locations where cleanroom and piping expertise are hard to come by, says Hyde's Watler.
He foresees layouts becoming simpler and more standardized, driving costs down and
enabling the construction of smaller, closed processing type plants in both developed and
developing countries. Developing countries no longer want to be 'Coca-Cola bottlers,'
simply lling imported drug substance. ey are moving to self-su ciency for the
therapeutics and vaccines unique to their regions, and small, modular facilities featuring
single-use systems are well suited to meeting this need," he says.
Working in synch with modular technology is disposable process equipment, which
can help reduce investment and operating costs, and potential fnancial risk for anyone
launching a new pharma product. Howard Levine, head of BioProcess Technology
Consultants, summarized some of the economic benefts recently at the World
Vaccine Manufacturing Congress in France (Table, p. 22).
Disposables can reduce labor costs by about 30%, Levine noted, although
they result in 20% higher raw materials costs, ofering a net savings, per
manufacturing campaign, of about 10%, he said. However, additional
savings come from the elimination of column packing and elastomer
changeouts, as well as lower validation, calibration and equipment
preparation requirements, as well as shorter processing time.
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM ANUA|Y 2012 1
Te leading edge of pharmas new manufacturing base
may be seen in vaccine manufacturing and niches such as
personalized medicine, in such concepts as the pandemic-
ready vaccine facilities being developed by Texas-based
G-Con, LLC using its GMP-ready modular cleanroom
technology. G-Cons partners in various ventures include
Xcellerex (Marlborough, Mass.) and GE Healthcare
(Chalfont St. Giles, UK). GE is also collaborating with
M+W Group (Stuttgart, Germany) ofering turnkey
construction solutions aimed at global markets.
Initiatives driven by the Gates Foundation, WHO and
GAVI will spur development of more modular vaccine
manufacturing facilities, allowing countries to tailor
development to their unique regional needs, predicts
Hydes Watler. One of the frst such facilities was CPL
Biologicals vaccine plant, a partnership between Novavax
(Rockville, Md.) and Indias Cadilla Pharmaceuticals,
which was completed in 2010 in Dholka, India.
In the U.S., increased government investment is
stimulating more research and commercial activity in
modular vaccine facilities using new plant, insect or
animal cell culture platforms. Working with various
plant cell culture platforms originally developed by
Fraunhofer USA Center for Molecular Biotechnology
20 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
FUTURE FACI LI TI ES
Last year, MedImmune received ISPEs Facility of the Year
Overall Winner award for its new large-scale mammalian
cell culture-based production facility adjacent to its Freder-
ick Manufacturing Center (FMC) in Frederick, Maryland.
Brent Hill, director of automation within MedImmunes
Global Engineering organization, and Victor Ronchetti, Sr.
VP and technical director for systems integrator Auto-
mated Control Concepts, explain how they harnessed IT
and process control tools to meet an extremely aggressive
product schedule.
PhM: What was the impetus to do something extraor-
dinary and, in your mind, what is truly groundbreaking
about the Frederick project?
Hill, Ronchetti: We had recently just nished a large Pilot
Plant project similar in nature to the one in Frederick. We
had learned that we would need to do something that was
extraordinary to meet the aggressive schedule. Therefore,
we set in motion several systems, such as the Factory Ac-
ceptance Test Process Automation Core (FATPAC) and code
control, to position ourselves for success.
As most projects go, there are unforeseen events that
involve patience and adaptability. Near the completion of
the project we were given a completely new set of built
P&IDs. The control system had already been designed
around the design documents. We were now faced with
the task of re-coding and re-execution of the Process Con-
trol System (PCS) SAT.
Due to competing deadlines, MedImmune was now
faced with a nearly vertical schedule where the constraint
was the physical equipment in the facility. To stay on
schedule, the automation team used Rockwell Softlogics
software to completely replicate the process control
system in its entirety. As shakedown activities took pre-
cedence in the schedule, it was necessary for the project
team to perform activities related to the commissioning
and qualication of the PCS in parallel, without interrupt-
ing shakedown runs.
Under the strain of limited time on equipment as a
result of competing project phases (start-up and debug-
ging of applications, train production operators, and
running test batches), the project team developed a
separate PCS simulator (in addition to the PCS simulator
used for operator training) that allowed them to commis-
sion and qualify major aspects of the PCS without having
to perform the work on the plant oor. This system,
which simulates every Programmable Logic Controller
(PLC) in the facility, provided a safe, equivalent environ-
ment to perform testing.
USING ORDINARY TOOLS IN EXTRAORDINARY WAYS AT MEDIMMUNE
Simulated training helped prepare the Frederick team in advance.
FUTURE FACI LI TI ES
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM ANUA|Y 2012 21
PhM: What about S-88 (aka ANSI/ISA-88) worked for
this project?
Hill, Ronchetti: Giving the specic control modules, equip-
ment modules and phases to each of the appropriate
vendors was key to the projects success. With the regula-
tory nature of the pharmaceutical industry, companies are
becoming more informed about the value of only validat-
ing these standard modules and then using them time and
time again. We are seeing the majority of the companies
in the pharmaceutical industry going this way on most, if
not, all hardware/software platforms.
PhM: You had 44 skids from different vendors all over the
world and expected them to have common data storage,
a single domain controller, form, t, function, etc. Did this
prospect seem ludicrous when the project began?
Hill, Ronchetti: At the beginning of the project, we sent
out in the bid specications, our requirement that all skid
vendors were to use our standard modules and our inter-
system communications specication. Once the vendor
understood what we were trying to accomplish, all but
one not only accepted it, but were enthusiastic about it.
They realized how much easier it would make start-up and
integration of the system as a whole.
PhM: Explain a bit more about FATPAC, if you could.
Hill, Ronchetti: A common issue in automation projects,
especially of this size, is interfacing skidded systems
with the PCS and with each other. Using a proactive ap-
proach to solve this potential issue, the team developed a
FATPAC, a portable interface package to support FAT. This
package of servers replicated MedImmunes high-level
process network and allowed MedImmune to test the
equipment in the appropriate environment, at each site.
The FATPAC included a domain controller used to preset
user access, a PLC, and an HMI client from the PCS system.
Communications were set up with the PCS PLC/HMI and
used in the FAT for each of the skids. Any problems identi-
ed during FAT were resolved and retested prior to ship-
ment. When the skids arrived onsite, minimal setup was
required to integrate them into the PCS system.
PhM: Does the simulation project provide a template that
you can now apply to most any process going forward?
Hill, Ronchetti: As discussed earlier, the PCS simulator was
a complete replication of the live PCS system. If a change
was needed due to errors found in testing or errors found
on the live system the code was changed on the live
system rst. Each night the code from the live system was
then downloaded into the simulator. Asset Centre was
used to maintain code equivalence between systems and
allowed for traceability. Test scripts were then generated
and the errors were then retested to insure the quality
and integrity of the codes. This process is now the stan-
dard for all of MedImmune Control system projects.
PhM: Brent has said, Automation is always to blame for
slow startup. Does the Frederick project change this?
Hill, Ronchetti: The very nature of automation is considered
a risk. The errors or problems in coding have an unknown
duration, which is what has historically has given automation
a bad name. By careful planning, the industry as a whole
can help make the paradigm shift. Once it is obvious that al-
though critical to the success of any project, the automation
need not be one of the greatest unknowns of a project.
are Project Greenvax, based in Texas, whose partners
include Xcellerex, G-Con and Texas A&M University,
and iBio, Inc. (Newark, Del), whose iBioLaunch
platform was patented last October.
HYBRID FACILITIES: FINDING THE RIGHT MIX
While mobile facilities would, by defnition, be based on
disposable equipment, it is unlikely that a large tradi-
tional type facility would be built with 100% dispos-
able equipment. At larger scale, their cost and ef ciency
attractions diminish, says Watler. Most facilities on the
ground today are opting for a hybrid approach, com-
bining biodisposables with traditional stainless steel
equipment. Te challenge that manufacturers will have is
determining the right ratio of each and where disposables
can convey signifcant cost and operational advantages
over traditional equipment.
Finding the right mix has been a challenge for DSM
Biologics as it builds a major new facility in Brisbane,
Australia. Te new site has a six-story shell in place that
is being ft out in 2012. Te second and third foors are
empty for the time being, available for future build-out.
Nevertheless, when it goes online some time in 2013,
DSM Brisbane will become by far the largest biopharma
contract manufacturer down under,
says Ben Hughes, senior process
engineer for DSM who is overseeing
much of the project work.
All upstream processes at
the site, from media and bufer
preparation through to bioreactors,
are anticipated to be single-use,
Hughes says. Weve shrunk the
process suites by keeping all media
and bufer concentrates outside of the
controlled areas. Te product and all
solutions will be pumped through
the walls into 500-liter bag-lined
totes awaiting transfer or directly to
the process equipment.
Downstream, we envision single-
use for all of the fltration steps, he
says. Chromatography will be more
of a hybrid approach.
Single-use has all sorts of
advantages in a CMO environment,
he continues. Its also really nice for
us in terms of future-proofng the site.
Its extremely adaptable, with plug-
and-play skids and so forth, so its easy
to roll in new single-use technologies
and roll out the older ones.
Its also about sustainability,
Hughes notes, especially as single-use
eliminates CIP and SIP procedures and
signifcantly reduces consumption of
water, cleaning chemicals, energy and
other resources.
SENSE AND SUSTAINABILITY
Te Brisbane facility is also a sign of
the times in that it is the product of
tight collaboration between DSM and
the Queensland State Government
and the Commonwealth of Australia,
who provided fnancial assistance
for the project. Around the time of
the BIO annual meeting in 2009,
BioPharmaceuticals Australia (BPA)
requested expressions of interest to
operate a major new bio contract
manufacturing facility. It was fortu-
itous timing for DSM, says Hughes,
as the manufacturer was looking
to expand its biologics capabilities.
DSM was selected as BPAs partner
and work began in 2010.
FUTURE FACI LI TI ES
22 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
FACILITY COSTS FOR A BIOREACTOR
(Stainless Steel vs. Disposable)
Stainless Steel Single-Use
Construction time 16 months 14 months
Process area 6,372 ft2 6,781 ft
Class C 1,109 ft 667 ft
Class D 5,231 ft 3,315 ft
CNC 0 2,745 ft
Total area 12,153 ft 2,745 ft
Piping length 2,854 ft 886 ft
Process equipment cost 4 million Euros 3 million Euros
Total equipment cost 17.3 million Euros 15 million Euros
Source: Levine, H., Vaccine Manufacturing in the Coming Decade, presented at the World Vaccine Manufacturing Congress, 2011, Lyon,
France, October 11-12, 2011
continued on page 26
A mockup of
a production
space in the
DSM Brisbane
facility.
800-648-5950
Find Out More At www.ham-info.com/0249
Hamilton Bonaduz AG
Bonaduz/Switzerland
Tel. +41 (0)81.660.60.60
Fax +41 (0)81.660.60.70
contact@hamilton.ch
Hamilton Company
Tel. +1.775.858.3000
Fax +1.775.856.7259
sensors@hamiltoncompany.com
www.hamiltoncompany.com
Advanced Process Control
Dissolved Oxygen, pH, and Conductivity measurements
connect directly to your PLC
Robust 4-20 mA and Digital signals generated directly from the sensor
Eliminate weak signal problems and transmitter costs with direct connection to PLC
In lab pre-calibration and digital sensor management
Several monitoring options including wireless monitoring of multiple sensors
FUTURE FACI LI TI ES
24 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
MODULAR MOMENTUM
Malcolm McLean is a name that few people in pharma
know, and yet his 1956 invention is helping to revolutionize
drug facilities today. McLean is the father of the modern
day shipping container, which is inspiring a modular facili-
ties boom in our industry. The shipping container provides
a convenient size for, say, a cleanroom or facility production
add-on, and of course can be easily, quickly transported to
anywhere in the world. Modular facilities:
geherally require modesI up-!rohI capiIal ihvesImehI
cah be builI ahd IesIed aI home be!ore beihg shipped
cah be moved Io wherever Ihey are heeded, especially Io
emerging markets
are easily sIahdardized Io !aciliIaIe Iraihihg ahd cohsis-
tent practices between sites
cah be scaled up (or "scaled ouI") oh ah as-heeded basis
Founding its business on the shipping container concept
is Biologics Modular, headed by Clark Byrum, president
and CEO, and Chris Wernimont, VP of Engineering and
Operations and formerly of Eli Lilly. This past September,
it led for several patents related to manufacturing clean-
room and biologics production space within the shipping
container connes.
"Our origihal IargeI markeI was R&D Iype sIu!!," says
Byrum. But they soon realized there was potential for GMP
manufacture, and Biologics Modular produced its rst
"Alpha UhiI" ih 2010. A Iypical uhiI would be abouI 80%
compleIe oh desigh wheh purchased, wiIh !ull HVAC ahd
oIher essehIials, leavihg Ihe remaihihg 20% Io be modi!ed.
Biologics Modular works with outside consultants to com-
plete the process-related work, which could mean building
as many as three or four production steps in one container.
Their rst order was a stem cell processing company that
required a GMP facility as well as an analytical lab.
Still, the industry is just warming to the idea of modular
!aciliIies. "Our biggesI challehge is chahge," 8yrum says.
"We're Iryihg Io chahge Ihe mihdseI ahd show mahu!ac-
Iurers a beIIer way Io be more e!!ciehI."
One of the pioneering forces in pharma modular con-
struction was Pharmadule. That rm has gone bankrupt,
but many of its assets are now part of ModularPartners,
!ouhded by Par Almhem. "Modular soluIiohs have gohe
from being a novelty and something very few would
consider, to now being discussed and considered in many
(mosI?) cases," Almhem says. "However, !ew have sIarIed
to take advantage of the real opportunities in modular-
ization in the way that, for example, the software and
elecIrohics ihdusIries do."
1his requires a chahge Ioward more o! a "buildihg
block" mehIaliIy, he says. Ih addiIioh, "Ihe mosI sighi!-
cant hurdles have been the tradition of how projects are
planned and executed. Modular solutions need to be
considered in the early stages of a project in order to get
the full benet of them. Many times, if the question about
modularizaIioh has beeh asked, iI has beeh Ioo laIe."
Modular facilities have also had a reputation for being
expehsive, Almhem says. "1his has someIimes beeh Irue,
parIially because Ihere hash'I ih Ihe pasI beeh ehough
volume Io really drive dowh cosI."
The ModularPartners modules are also shipped is a fash-
ion similar to shipping containers. The main difference,
Almhem hoIes, is IhaI Ihe modules are larger ahd shipped
wiIh greaIer care (wiIh addiIiohal weaIher proIecIioh, !or
example). For our !ull cohversaIioh wiIh Almhem, see p. 15
of this issue.
Exterior and interior (left) views of a
Biologics Modular cleanroom.
Consistent Performance!
One source for all your process control cabinets,
worldwide. Improve your network performance
and reliability by working direct with the leading
valve manufacturer. A true global partner for
increasing logistical efficiences and standardization.
Global manufacturer of process control and factory automation solutions
For more information:
Call: 1-800-Go-Festo
1-800-463-3786
www.festo.com/us/biotech
Section J of the Australian
Building Code, enacted in 2010, calls
for advanced greening of all new
buildings, including, for example,
signifcant increases in thermal
insulation, window glazing and
shading, rainwater collection, and
high-efciency equipment (e.g.,
HVAC and piped services with
variable speed drive motors). A
sophisticated Building Management
System provides detailed operating
and energy consumption data.
Te facility will be oriented to the
north for optimal solar design, is
convenient to public transport, and
has a large area for bicycle storage.
Any of these requirements alone
is not groundbreaking, says Hughes.
But together, they signal a new way of
building drug sitesheartening to
see, he says.
All of this comes in a smaller
building than would have been
imagined in the pasta mere 8,000
square meters. DSMs proprietary
technologies for productivity
for example, for ramping up
mammalian cell culture or
expanded bed chromatography
yieldswill allow it to achieve
in single-use bioreactors what
would have required stainless steel
bioreactors with 10 to 15 times
greater capacity. Weve achieved a
footprint that is much smaller than
facilities built in the past but will
still maintain the required output,
Hughes says.
Facilities of the future will be
green, experts agree. According
to estimates from the design and
construction frm CRB, bulk biotech
manufacturing facilities of the future
can realize green savings in the
neighborhood of:
- 73-93 water usage reduction
- 30-93 chemical reduction
- 30+ energy reduction
- 23-30 smaller carbon footprint
- 30-40 smaller land requirement
When considering a hybrid
installation, Hydes Watler says,
the frst question to ask is, is the
single-use system designed from
the ground up, or is it simply a
modifcation of a conventional
system? If it is simply a
modifcation, did the workaround
result in follow-on issues which
must be addressed?
He uses the rectangular
disposable bag in bin as a good
example of a ground-up design. It is
not a mere modification of a round
stainless steel tank, Watler says, but
rather, is rectangular, stackable, fits
easily against a wall, has an open
top and side panels for access
quite different from a conventional
tank, and it works.
In contrast, buffer preparation
tanks have been challenging.
There are no baff les so mixing
is not as rapid, solids can settle
in corners and crevices and it
is a challenge to insert sensors
into the tank, Watler says. The
competition stainless steel mix
tank, of proven design, with
an optimized CIP cycle can be
changed over in 15 minutes. At
larger scales, this conventional
tank may be the best option,
operationally and cost wise.
He recommends using FMEA and
PHA risk analysis to sort through the
mix of single use and conventional
equipment. As facility design evolves
and disposable systems become further
incorporated, Watler says, there will be
similarities between diferent sites, but
not to the extent of standardization of
hybrid facility designs.
Whats clear from talking to
facility experts is that the drug
manufacturing plants of the future
are already here, being designed
and built worldwide. Tey will only
become more afordable, fexible,
green, and commonplace in the
coming years.
FUTURE FACI LI TI ES
1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
BI OPROCESSI NG
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM ANUA|Y 2012 27
MANY BIOPHARMACEUTICAL applications require
vent flters, the hydrophobic sterilizing-grade flters that
are used as air vents on processing tanks. Tese flters
maintain near-ambient pressure in the tank while ensuring
sterility. In addition, the flters remove viruses and micro-
organisms from the gas as it fows into or out of the tank.
To ensure its proper operation and sterility, a
bioreactor, for example, may have a number of vent flters
including those for the tank vent, the sparge gas inlet, and
the overlay gas inlet.
Understanding the diverse applications of vent flters is
critical to their proper implementation and use.
Following best practices can help ensure compliance, and
avoid problems during operation, installation, CIP and SIP.
By Michael Felo, EMD Millipore
A number of factors must be
considered in advance of vent flter
implementation, including flter
sizing, housing and piping design,
condensate control, regulatory
requirements, and operational
considerations such as clean- and
steam-in-place (CIP, SIP) and
integrity testing. Employing best
practices for vent flter use can
avoid common problems during
installation, CIP, SIP, integrity
testing and operation.
Tis article will describe an
application-based approach to vent
flter sizing, in situ integrity testing
of the flter design, best practices
for housing design and vent flter
operation, and a risk management
approach to implementation and
replacement of vent flters.
VENT FILTER SIZING
Air fows in and out of a process
tank commonly for two reasons: the
frst is to replace a volume of liquid
as it is pumped in or out of the tank.
Sizing the tank vent flter for pump-
out or fll rate is relatively simple, as
the air fow rate will be equal to the
pump-out or fll rate. With fow rate
and pressure determined, a fow/
pressure change (P) curve can be
used to determine sizing.
The second reason air will f low
into a process tank is to compensate
for the volume change associated
with steam condensation. At the
end of a tank SIP procedure, steam
in the tank will cool and undergo
a phase change to liquid water.
There is more than a thousand-
fold difference in volume between
water in the gas phase and water in
the liquid phase. During cooling,
sterilized ambient air must be
allowed into the tank to prevent
vacuum. Sizing the vent filter for
steam collapse requires knowing
the vacuum rating of the tank and
the convective cooling rate. These
can be calculated based on the
tank dimensions including height,
diameter, and wall thickness.
Improper tank vent sizing can
result in low pump-out rates, loss
of sterility due to a ruptured disk
or flter failure, or worst case, tank
implosion. Fortunately, proper
sizing is not difcult, as long as the
fow requirements and driving force
are understood.
Tank venting can be static or
dynamic, with each requiring flters
that are sized slightly diferently.
For static venting, the air outside of
the tank is assumed to be at ambient
pressure, so the driving force for
airfow is determined by the pressure
diference between the inside of
the tank and the atmosphere. For
dynamic venting, compressed air is
fed to the tank in order to minimize
any diference in pressure that occurs
between the tank and atmosphere.
Static tank venting is commonly
used for bufer tanks and
intermediate storage tanks. To
determine the proper size for a static
tank vent flter, four basic steps
should be followed:
1. Determine the maximum fow rate
for venting that the vent flter will
need to provide. Tis will be either
the process fow rate or steam col-
lapse rate afer an SIP cycle.
2. Select the maximum pressure
drop that you want the filter to
experience. The pressure drop
is typically less than 5 psi and
should be dictated by the vacuum
rating for the tank or rupture
disk vacuum rating. Clearly, it
is important to avoid pulling a
BI OPROCESSI NG
28 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
2.6
0 bar
2.6
0 bar
2.6
0 bar
2.6
0 bar
Compaction
Instantaneous
ow
measurement
Time
Flow
Cross
section
Stabilization
8
I\U
P
atm
Figure 1. Hydrocorr water ow integrity testing process.
5)&3&*4.03&5)"/"5)064"/%'0-%%*''&3&/$&
*/70-6.&#&58&&/8"5&3*/5)&("41)"4&"/%
8"5&3*/5)&-*26*%1)"4&
strong vacuum on the tank that
might cause collapse.
3. Calculate the number of filters
or filter area required to meet
f low rate and pressure drop
requirements.
4. Ensure an adequate safety factor
(~1.5x) and select the appropriate
flter confguration for the tank or
application.
Table 1 shows the vent sizing
process for a static vent. Tis example
represents a hold tank used for
ambient temperature water storage
with a maximum pump-out rate
of 400 liters per minute. No SIP
is necessary and the tank has no
vacuum rating.
Common uses for dynamic tank
venting include bioreactors and other
applications where steam is replaced
with compressed air afer an SIP
cycle. In this case, the process for
sizing a flter varies slightly from the
static model and is as follows:
1. Select the desired pressure drop.
Pressure drop is typically less than
or equal to 2 psi, especially when
calculating for bioreactors where
minimizing vacuum is key to
maintaining a sterile environment
in the tank.
2. Calculate the air fow rate neces-
sary to replace the steam during
steam collapse post-SIP.
3. Calculate the number of flters or
flter area needed to meet fow rate
and pressure drop requirements.
4. Ensure an adequate safety factor
(~1.5x) and select the appropriate
flter confguration for the tank or
application.
Dynamic vent sizing can be
significantly more complex than
static venting since the steam
collapse rate needs to be calculated
BI OPROCESSI NG
30 x 18 Ametek Centrifuge Hastelloy C-276
A n l
8 C
u C u
C 8
P 8
P L
l l
C
k L 8
S 8
v S u
Kd

W
Parameter Value
Maximum ow rate 400 liters / minute (~13 SCFM)
Maximum pressure drop P 1 psi (due to no vacuum rating on tank)
Filter size 5 Aervent lter (EMD Millipore)
provides 20 SCFM at P = psi
Table 1. Vent sizing process for a static vent.
for the specific tank and process conditions being
used. In this case, it is recommended that available
software programs be used to calculate the proper
dynamic vent filter sizing.
IN SITU INTEGRITY TESTING OF FILTERS
Vent flters are typically tested using a process such as the
Hydrocorr water fow integrity test (Hydrocorr Valida-
tion Guide, MM document VG050) [1]. Te test measures
the resistance of the flter to water intrusion (Figure 1). A
flter is placed in a stainless steel housing which is fooded
with water. Under a pressure of 40 psi, there is compac-
tion of the flter. Over time, the compaction stabilizes
and the fow decreases. Once stabilization is complete,
an instantaneous fow measurement can be taken; if the
measurement is below the specifcation of the vent flter, it
is considered to be integral.
In situ water-based testing can be conducted when
the flter is attached to the tank using a manual or fully
automated process. With the manual process, the flter
is fooded, the Hydrocorr integrity test is conducted,
and once a passing value is achieved, the vent flter
housing is drained. Te same steps are used in the fully
automated process, which can be a more efcient way of
testing when in a commercial operations setting.
HOUSING DESIGN AND VENT FILTER OPERATION
Best practices for the implementation of vent flters
include selection of the proper housing. Tree housing
options are available (Figure 2):
- C-line. Ofers the best condensate trapping capacity on
the upstream side of the flter. If the stream is in an es-
pecially moist or humid environment, the C-line format
allows excellent removal of condensate.
- T-line. While this format is popu-
lar, it is not optimal for vent fltra-
tion, as the housing has to be tilted
for proper condensate drainage.
- In-line. Tis format allows the
downstream condensate to drain
directly into the vessel.
In all cases, a vertical mounting of
the housing is always required to
enable the most efective drainage of
condensate.
Te life cycle of the vent flter includes installation
in the housing on the tank, CIP, SIP, and operation.
Adopting best practices at each step can help ensure
proper functioning. During installation, flter o-rings
should be pre-wetted for easier installation in flter
housings. Code 7 tabs at the bottom of the cartridge
should be locked in to their housings as venting occurs
in the reverse direction and, if there is pressure pulsing,
the cartridge may be ejected from the flter housing.
During CIP, the sprayball direction should be
adjusted so that CIP liquids do not reach the flter
as caustic solutions can impact the strength of the
membrane. A heat jacket, blanket, or trace should
be used to minimize condensate buildup in the flter
housing during the CIP cycle. CIP creates a moist,
humid environment so it is important to avoid
instantaneous release of high pressure, moist gas
through the vent flter, which can cause damage.
SIP is the most mechanically strenuous step that a vent
flter will experience. A number of best practices can be
employed to prevent damage to the flter:
- Avoid pressure pulsing across the vent lter.
- Minimize a high-pressure change across the lter at
high temperature as the flter can be-
come weaker at elevated temperatures.
- Steam the vent lter in reverse direc-
tion from the tank to ensure conden-
sate removal from the core of the flter.
- Delay the vent lter sterilization step
until afer the tank heating phase is
complete.
- Air/condensate should be sent via the
drain and not the vent flter during the
heating phase.
During the operation phase, it is best to avoid fouling
of the flter from entrained liquid in the tank especially
when using the lter on a bioreactor. A heat jacket/
blanket/trace should be used during the process if the
potential exists for moisture build-up.
BI OPROCESSI NG
30 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
Figure 2. Vent lter housing options.
4*1*45)&.045
MECHANICALLY
453&/606445&15)"5
"7&/5'*-5&38*--
&91&3*&/$&
RISK MANAGEMENT
As shown in Table 2, regulatory authorities have addressed
the implementation of vent flters. All advocate a risk-based
approach be used to establish flter re-use and integrity
testing policy.
A typical risk assessment and process validation
plan should consider the criticality of the process and
application of the vent flter, the number of re-use cycles,
process implications, and the impact on flter lifetime.
Regulatory documents establish two types of
applications for vent flters: critical and moderately
critical applications. Critical applications are those in
which fltered gas is in direct contact with the sterile
fnal product. In this situation, the sterilization cycle
must be validated and performed before each use of
the vent flter. Te flter must be integrity tested upon
installation and following each use.
In moderately critical applications, the filtered
gas is not in direct contact with the final product. In
these applications, sterilization and integrity testing
frequency should be established based on the following
risk assessment parameters:
- Historical in-process lter integrity test data
- Microbiological/bioburden limits established for the lter
- Filter process conditions
It is common practice to reuse vent filters over mul-
tiple cycles. A risk-based assessment should help guide
reuse and change-out criteria. The assessment should
consider the following:
- Maximum number of sterilization cycles (a manufac-
turer or internally set limit)
- Preventative maintenance schedules for related equip-
ment (the tank on which the vent is mounted)
- Potential for cross contamination from dierent prod-
uct strains
- History of integrity test failures
- Humidity during operation
- Maximum pressure drop for non-clean air stream
- Service life of the lter at the temperature for heat
traced housing
- Specied utilization number of uses expected for the lter
References
1. Jaenchen, R., Schubert, J., Jafari, S., and West, A. Studies on the
theoretical basis of the water intrusion test (WIT). European Jour-
nal of Parenteral Sciences, 1997, Vol. 2, No. 2, 3945.
2. Technical report no. 40: Sterilizing fltration of gases. PDA Journal
of Pharmaceutical Science and Technology, 2005, Vol. 58, No.S-1.
3. Pharmaceutical Inspection Convention, Pharmaceutical Inspec-
tion Co-Operation Scheme. Recommendation on the Validation of
Aseptic Processes. Section 9.6.1, January 2011.
4. EC guide to GMP Annex 1: Manufacturing of sterile medicinal
product, 2003.
About the Author
Michael Felo is an Applications Engineer Consultant in the Merck
Millipore Biomanufacturing Sciences Network (BSN) whose focus
is optimization, implementation, and troubleshooting of late-stage
clinical and commercial scale manufacturing processes for the tech-
nologies of clarifcation, aseptic fltration, and single-use systems. Mr.
Felo has over 12 years of experience in process development, technol-
ogy transfer, and clinical and commercial GMP manufacturing of
monoclonal antibody and recombinant protein therapies.
BI OPROCESSI NG
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM ANUA|Y 2012 31
PDA TR40
No single approach applies to all applications, and an appropriate testing fre-
quency and rationale should be selected using risk analyses considering impact on
product quality and regulatory compliance.
PIC/S Section 9.6.1
It is important that the integrity of critical gas and air vent lters is conrmed im-
mediately after the lling and if it fails, the disposition of the batch determined.
In practice vent lters fail the integrity test more frequently than product lters,
as generally they are less robust and more sensitive to pressure differentials during
steam sterilization.
European Commission
Results of these checks should be included in the batch record. The integrity of
critical gas and air vent lters should be conrmed after use. The integrity of other
lters should be conrmed at appropriate intervals.
Table 2. Examples of regulations addressing vent lter implementation [2,3,4].
DESIGNING A new or expanded manufacturing facil-
ity with a new or modifed process requires two sets of
knowledgethat of process-fow architects and that of
the manufacturing production experts.
Together, process-fow architects and manufacturers
possess key components of the knowledge required to
fnd a better process. A manufacturer understands the
process fow of a current facility, has adapted it over the
years to better ft the existing facility and site constraints,
knows what will and wont work in an existing facility
and, perhaps most importantly, understands the
imperatives of the current market for its product.
A process-fow architect possesses an overview of how
manufacturing processes in diferent industries have
maximized quality and efciency. He or she can, for
instance, pluck an idea from a plastics manufacturing
process and plug it into a pharmaceutical process.
At the beginning of a project, however, neither side
has a comprehensive understanding of what knowledge
or insights the other can ofer to this specifc project.
Designing a more efcient manufacturing process begins
with each side telling the other enough about what it
knows to create a common body of knowledge that can
be shaped and moldedby both sidesinto a new and
improved process and product. While all of this may
seem like common sense, it is difcult for manufacturers
and process-fow architects to generate a mutually
benefcial planning relationship.
It is similar to the situation that a home seller and
homebuyer fnd themselves in. Te seller resents the
comments that the prospective buyer makes about
repainting or adding a room here or changing a room
there. Te prospective buyer views the homeowner as a
small thinker. But if the home seller fnds a way to like
the prospective buyers ideas and if the prospective buyer
focuses on what he or she likes about the existing house,
both sides are much more likely to fnd common ground
and, perhaps, a sales contract.
Te same dynamic can improve the results from
process-fow architects and manufacturers. Both sides
must identify the common goal: an improved process. Tey
must resist the natural resentments that can crop up, and
focus on answering each others questions until someone
has an insight that makes everyone shout, Tats it!
WHAT IF YOU CUT A HOLE IN THE CEILING?
Consider the case of a pharmaceutical manufacturer
working through the question of how to increase the
production of tablets. Te existing production process
uses forklifs to deliver powder to a compactor, which
compresses the powder into a solid ribbon of material
that could be processed into tablets.
In discussions with the process-fow architects, the
manufacturing team indicated that the key to increasing
production was to move more powder into the compactor.
Te forklifs couldnt keep up. Adding more forklifs
wouldnt work because the existing foor did not have
enough space to permit more trafc.
How about an addition? asked the architects. Te
manufacturer had already considered that option in detail.
One area of the plant ofered plenty of space for the addition,
but it was so far away from the post-compaction production
line that the time required to transport the material wouldnt
produce a measurable improvement.
Te expansion would have to go in near the existing
compactor, but that option raised structural issues related
to supporting walls and the second foor area of the plant.
Both the architectural team and the manufacturing
team worked the options over, searching for the answer to
what had become the key question: How do you design an
addition to the building in the area of the compactor?
Finally someone asked, Can we cut a hole in the
ceiling? Tat question unleashed more questions and
the ultimate answers. As it turned out, the foor above
the compactor had been abandoned some time ago. Tere
was nothing up there. A vertical expansion could include
a tall lif system that would raise a product-flled tote and
dispense it directly into the compactor feed tube.
Te lif could be designed to eliminate the time-
consuming docking and maneuvering that the current
compactor assembly required of the forklifs. Te forklifs
could feed more powder to a larger compactor, which
FACI LI TY DESI GN
Finding the Flow
Designing a new pharmaceutical manufacturing facility requires that architects and manufacturers
collaborate to improve process flow.
32 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
BY ALAN A. LIDDY, AIA, NCARB, PMP, SSOE GROUP
WHEN ARCHITECTS AND PROCESS
%&7&-01&34803,50(&5)&3
INGENUITY RESULTS.
could, in turn, feed enough ribbon to the fnal production
line to increase production as necessary.
Te concept posed several challenges. Te most
signifcant involved removing a portion of the existing
roof, reinforcing the roof framing and constructing a
roof penthouse to accommodate the new equipment.
Lesser but still signifcant challenges included
determining the most cost-efective approach to the
suite renovation that would still accommodate the
equipment; coordinating the equipment suppliers
during design as equipment design and assembly
occurred simultaneously; and maintaining production
operations during the renovation, which would have to
be carried out in an adjacent space.
Tese challenges all required close coordination between
the construction phasing and production phasing. At
the same time, the architects worked to keep the owners
fully informed about the complex design solution. Tis
was essential because the owner could not be expected to
visualize what the renovation would look like and how it
would function from the construction documents. Te
architects continually elicited questions from the owner
and provided explanations to ensure that the owner
understood and approved of what was being built.
ADDING A NEW AND POTENT
POWDER PRODUCT, SAFELY
Regulations also wield great infuence over the design of
a plant. FDA, for instance, requires that manufacturing
lines producing injectables meet an ISO-5 cleanliness
standard, which afects the HVAC design by requiring
more air changes than normal. ISO-5 also requires spe-
cialized packaging procedures handled by people wearing
personal protective equipment (PPE).
One manufacturer recently satisfed these regulations
while incorporating a new, potentially dangerous
product into an existing manufacturing facility. Te
simplest solution, from a logistical point of view, was to
expand through the back wall of the plant. Behind the
wall, however, were ofces and a primary connecting
corridor, which could not be moved efciently. Te
plant manufactures a pharmaceutical product by
formulating and batching ingredients. So the plant
managers and employees were familiar with batching
work. With the back wall of the plant eliminated as an
option, the architects had to fgure out how to ft the
new process into a limited space on the already crowded
manufacturing plant foor.
A downdraf booth seemed like the solution, but that
would require coming to terms with three challenges.
First and most importantly, the design would have to
control the cost of constructiondowndraf booths can
be prohibitively expensive to purchase and install.
Once again, the concept required a lot of discussion to
ensure that the owner had a complete understanding of a
complex concept that design drawings probably couldnt
convey in satisfactory detail to people unaccustomed
to reading plans. By the time construction began, the
architects and the owners had been through many more
review meetings than would be expected for a relatively
small project. Te architects were also careful to involve the
owners and other stakeholders as construction proceeded.
Second, the design would have to control the cost of
heating, ventilating and cooling the spacedowndraf
booths require 100 percent air changes all the time.
Tird came the problem of where to put the booth
within the existing facility. Te new facility had to be
big enough to do its job and small enough to hold down
costs and stay out of the way of existing processes. Te
plant already contained a half-dozen dispensing suites,
and one or two could be converted to a suite for potent
drug dispensing without requiring major modifcations
to the other suites to maintain the original production
lines. Still, the booth would have to be small yet allow
sufcient space for dressing rooms and airlock entries for
employees plus the airlock intake areas for the product.
Te architects whittled the booth itself down to an
8- x 8- x 10-foot box. Te small booth made it possible
to accommodate the support spaces for employees and
materials, while limiting the expense of the HVAC system
and its operating costs. As with any plant renovation, the
work had to be carried out without interrupting plant
operations. In the end, simplicity and efciency were the
keys to the downdraf booth design.
Indeed, simplicity and efciency are always the keys to
designing successful pharmaceutical plants, which number
among the most sophisticated manufacturing facilities in
the world. Tey cost thousands of dollars per square foot
to construct. Te simpler the process, the smaller the plant
and the lower the cost. While pharmaceutical plants must
embrace practical design considerations over aesthetic
desires, they are, in the truest architectural sense, the
result of form following function or process fow, and
then going with the fow.
About the Author
Alan A. Liddy, AIA, NCARB, PMP, is a Senior Project Manager at
SSOE Group (www.ssoe.com), an international engineering, procure-
ment, and construction management frm. With 23 years of experi-
ence, Alan specializes in pharmaceutical and nutraceutical projects.
He can be reached in SSOEs Raleigh-Durham, North Carolina ofce
at 919.361.9606 or Alan.Liddy@ssoe.com.
FACI LI TY DESI GN
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM ANUA|Y 2012 33
TECHNOLOGY MOVES fast, and it seems even
subject matter experts, aka SMEs, can have trouble
keeping up with their subject matter. Machine vision
is one such technology that keeps moving aheadin-
creasing the applications to which it can be applied in
pharma, for instance. Tose trying to follow along may
wish to visit LinkedIn and join in the Machine Vision
in Pharmaceutical Industry group. Its small by Internet
standardsjust 176 members at last countbut has some
good discussions, updates on new products, and can be
a helpful forum to get your questions answered by, yes,
SMEs in pharma machine vision.
The group is managed by Kasra Ravanbakhsh, who
specializes in machine vision and industrial automation.
A couple other LinkedIn groupsMachine Vision and
Machine Vision Groupare oriented towards various
industries but are larger and should also be interesting for
vision enthusiasts.
What are drug industry end users looking for in
their machine vision systems today? One thing is user
traceability, says Bob Ochiai of the machine vision group
at Keyence. That is, whos using the system, when are they
doing it, and who is authorized to make changes? Keyence
has upgraded user functions related to permissions
and changes for its XG Series (bottom, left) with more
sophisticated User Accounts and Modification Log
options, he says. These changes reflect the increased
importance of overall corporate accountability for all
systems and processes, Ochiai says.
Speed and resolution are also critical needs for end
users, he says. Just as consumers of digital cameras and
computers are demanding greater camera and processor
speeds and higher resolutions, so are drug manufacturers
demanding more robust machine vision systems.
Festo is focusing on simplicity and affordability,
and yet enhanced functionalities, says Frank Langro,
a manager for Marketing and Product Management.
Traditional systems with a master controller, sensor
interfaces, and various drives often result in complicated
and costly systems that place heavy demands
on operators, he says. In response, Festos
compact vision system SBOC-Q
utilizes familiar vision tools
for parts inspection. These
include Blob, ROI, and
Ray tools. In addition, the
vision system supports
OCR (optical character
recognition) and code
scanning for barcodes and
data matrixfor monitoring
blister packs, bottle caps, and
label coding, for example.
In addition to its inspection capabilities, the SBOC-Q
has an embedded CoDeSys PLC plus a CANopen
master interface to increase potential applications. The
system can inspect, but also control actuators to reject
failing packages, reorient items such as catheters, or take
other physical actions. All necessary information from
code reading to OCR can be passed back to a supervisory
system via an embedded Ethernet port.
TECHNOLOGY ROUNDUP
34 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
Pharmaceutical machine vision technologies are getting simpler
and cheaper, and yet tackling more varied applications.
By Paul Thomas, Senior Editor
End users have used SBOC-Q to measure the diameter
of glass ampoules for conformance to specifications,
Langro says. In the same application, the system used
backlighting principles to ascertain the presence of
printing. For blister checking, the vision system was
used to count pills and to assure that there were no
empty spots in the pack. If there were empty blisters,
the vision system identified the coordinates, made the
information accessible to operators, and then withdrew
the pack from the line.
In a syringe inspection application, the system was
used to identify missing or broken parts such as rubber
plunger head, auger flange, thumb rest, and needle hub. A
personal care application required the SBOC-Q to check
dental floss containers for the presence of the internal roll
of dental floss, threads coming out from the correct hole,
and threads correctly located on the cutter.
Banner Engineering has added the Q26 Series (below)
to its line of clear object detection sensors. While Banner
offers several sensors for reliable clear object detection,
the Q26 Series is optimized for the task with its polarized
retro coaxial design. The Q26 delivers both the sensitivity
required for reliable clear object detection and the robust
rejection of light from mirror like objects to prevent a
false detection. The coaxial optical design delivers the
additional benefit of precise leading edge detection, making
it useful for many high speed applications in bottling,
pharmaceutical and biopharmaceutical industries.
Primary applications are for pharmaceutical vials or
bag filling machines, Banner says.
In-Sight Track & Trace is a new software package
for Cognex In-Sight vision systems. The software
makes it easy for manufacturers to set up a complete
identification and data verification solution for
package labels, says John Lewis, the companys market
development manager. Customers choose the In-
Sight model that best meets the
performance and price
requirements of their
application, and then
add software, he says.
One key application is
reading an ID code and
verifying the accuracy of
printed textfor example,
date/lot code information.
If the label contains
GS1 data, In-Sight Track
& Trace can verify that the data is
formatted correctly, and that it matches the
printed text. Mismatched or incorrectly
formatted data would indicate a process or
database problem demanding immediate attention.
Another key application is that the software can
assess the quality of the Data Matrix codes on labels
while the system is operating in production mode. This
process check helps to make sure that the print quality
hasnt degraded in a way that will cause readability
issues downstream, Lewis explains. Equipment can
be mounted anywhere on the packaging lineat
the item level, carton level, case level, or at the pallet
level. According to Lewis, the most strategic area for
deployment is actually at the printing or marking
station. If a manufacturer is marking bad codes or
poor quality textor worst of all, printing the wrong
informationit needs to catch it immediately, before it
causes disruptions downstream in the supply chain.
PPT Vision recently agreed to be purchased by data
capture and industrial automation firm Datalogic
S.p.A., says PPTs president and CEO Bob Heller. PPT
will continue to operate as a separate company, with
headquarters in Minneapolis and a center of excellence
there for the Italian parent company. The partnership
provides both organizations with stronger global sales
channels, additional product
development resources and
industry leadership positions
in automation and machine
vision, Heller says.
Finally, in other news,
Begapt Solutions-India is
partnering with Microscan-
USA, with an emphasis on
solutions such as its Online 2D
barcode reading system.
TECHNOLOGY ROUNDUP
1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
Paperless Calibration Improves
Quality and Cuts costs
www.beamex.com
info@beamex.com
Portable calibrators
Workstations
Calibration software
Professional services
Industry solutions
Beamex

MC5 Multifunction Calibrator

Beamex

CMX Calibration Software

The business benets are signicant for companies
that use software-based calibration systems. The whole
calibration process from initial recording of calibration
data through to historical trend analysis will take less
time, whilst mistakes and manual errors will be virtually
eliminated.
Beamex, Inc.
Phone: 800 888-9892
Toll free: 770-951-1927
E-mail: beamex.inc@beamex.com
BLENDI NG OPERATI ONS
BLENDING, ONE of the most basic of pharmaceuti-
cal unit operations, can also be one of the most chal-
lenging to control. Solid formulations contain mul-
tiple ingredients beyond the active pharmaceutical
ingredients: fllers, tabletting agents, disintegrants,
and absorption enhancers or agents that slow down
and control absorption. Ingredients from difer-
ent vendors may behave diferently due to their
particle size and shape and other factors, and
their tendency to form aggregates.
Tere are strong economic drivers
for optimizing blending. Reinforcing
these is the pharmaceutical
Quality by Design (QbD)
framework advanced by
FDA, which requires a
deeper understanding
of pharmaceutical
manufacturing processes,
how ingredients blend and
how blending progresses
through diferent stages.
Traditionally,
formulation scientists
and technologists
have used destructive
analytical methods, such
as dissolution followed
by HPLC or UV, to
optimize blending.
Tis requires running
the process, pulling
and analyzing samples,
which can lengthen
the time required for
development.
Recently, hyperspectral
imaging (HSI) has been
applied to study the
behavior of solid particles in
various unit processing steps
as well as during multistep continuous processes. Tis
nondestructive method generates thousands of spectra per
second, providing more compositional information than
conventional methods.
Tis article summarizes the results of blend monitoring
studies using a new HSI device, in situ without stopping
the blending and pulling samples. In this case, a batch-
type blender was equipped with a computer controlled
drive mechanism capable of imaging blending through
a window mounted on the blender. Te push-broom HSI
camera is timed synchronously with blender rotation.
Spectral information acquired at each rotation was used
to assess whether the nominal composition is achieved,
to reveal blend uniformity, and to avoid incomplete
blending, large aggregates or re-aggregation.
A COMBINATION OF SPECTROSCOPY
AND DIGITAL IMAGING
Hyperspectral imaging (HSI), or chemical imaging (CI),
is the combination of spectroscopy and digital imaging.
A hyperspectral image contains many spectra, one for
each individual point on the samples surface. Te image
contains information about the spatial distribution of the
materials within the sample.
A hyperspectral camera (Figure 1) integrates an
imaging spectrograph with a matrix array sensor. In these
studies, we applied near infrared (NIR) spectroscopy,
using NIR hyperspectral imaging to analyze the average
composition, and the distribution of ingredients.
Hyperspectral camera literature refers to the full 1000-
2500 nm range as short-wave infrared (SWIR).
A special lens images the sample onto a slit of a
transmission spectrograph. Te spectrograph produces
a spectrum imaged on a focal plane array detector,
preserving the location of respective points on the slit and
thus the points of the line on the sample.
In push-broom HSI, successive lines on the sample
measured over time form a complete HS dataset.
Tis data from a HS camera is called a hypercube,
containing information in two spatial dimensions and
one spectral dimension. Te hypercube is typically
Imaging the Blending Process
Hyperspectral imaging can be used to optimize blending, by monitoring the
distribution of excipients and APIs in formulation
BY GABOR KEMENY AND GINA STUESSY, MIDDLETON RESEARCH
PHARMACEUTICAL MANUFACTURING JANUARY 2012 37
ratioed with similar hypercube
measurements of a highly refective
white reference material and with the
residual background signal, the latter
of which is measured when no light is
falling on the focal plane array.
Te resultant corrected spectra
are produced in transmittance,
refectance, or absorbance similar
to traditional spectroscopic
measurements. Te results can be
further processed, scaled, smoothed,
and eventually compressed to
produce the information sought from
the measurements.
PUSH-BROOM IMAGING
One of the signifcant diferences be-
tween HSI and conventional single-
point spectroscopy is the very large
amount of data generated. Processing
sofware and hardware are necessary
to keep up with the data stream and
provide compressed and processed
data, thus producing composition
maps and other technologically
meaningful information.
Push-broom HS cameras gather
a complete spectrum of each point
on one spatial line at a time [1]. Te
area of the object is scanned, one
line at a time in rapid succession.
To image the whole sample, either
the sample or the camera must
move. Te hypercube is collected
by compiling the optical data from
each spatial line. Since push-broom
imaging detects one line at a time,
the spectral data in the hypercubes
correlate with the same sample point,
thus push-broom HSI cameras are
used with the samples moving, which
is the case in many pharmaceutical
manufacturing lines, schematically
shown in Figure 2.
OPTICS VARY WITH NEEDS
In pharmaceutical production, there
are many points where the increased
amount of spectral information
and the spatial information could
BLENDI NG OPERATI ONS
38 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
spectrograph
attening
lter
lens
order
blocking
lter
sensor
S
p
e
c
t
r
a
l

A
x
i
s

[

]
D
o
w
n

W
e
b

/

T
i
m
e
Across Web
Hypercube data collection along
moving manufacturing line
Different sizes and
types of blenders
SWIR camera source
and optics module
Motor, power supplies,
camera controller module
Line,
switch,
fuse
LVDS/Control
cables IN
Computer
interface
LVDS/Control cables OUT
Figure 3. imMix system block diagram
Figure 2. Push-broom HSI for continuous monitoring of manufacturing process
Figure 1. Simplied Hyperspectral (HS) camera components diagram.
provide additional insights, such as
during transdermal manufacture,
tablets, capsule flling, and blend-
ing [2-6]. For diferent magnifca-
tions, required by the diferent
pharmaceutical applications, the
same type of push-broom camera
can be used with diferent optics. In
this research project, for instance,
a larger magnifcation was used to
resolve the aggregates of the various
pharmaceutical ingredients found in
solid formulations.
For these studies we used the
inMix system, which consists of
a push-broom SWIR HS camera
(Specim Ltd., Oulu Finland),
which is positioned to view the
blend inside the rotating blender
(Figure 3). Te blender is rotated
by a computer-controlled motor.
For the collection of HS data on
all or certain specifed rotations,
the blender is slowed down and
the camera is programmed to scan
the blend covering the window. HS
data collected through the optical
window on the blender is turned
into composition maps (indicating
spatial dispersion), which are used
to predict ingredients throughout
the blending process. Te very
large amount of imaging data is
condensed to a limited number of
useful micromixing parameters.
Te camera is protected in a
stainless steel housing; the motor,
power supply, and camera controller
and other electronics are housed
in another stainless steel module.
Te instrument is compatible with
diferent sizes and diferent types
and sizes of blenders, as the camera
position is adjustable. To empty
the blender, the front housing
including the camera module
can be easily slid out from under
the blender. A white reference for
background measurements and
devices to establish proper focus and
help measure pure components or
smaller quantities of materials can be
attached to the blender.
Te SWIR (Short Wave Infrared)
HS camera, with a wavelength
range of 1000-2500 nm, is equipped
with an OLES Macro SWIR HS
lens (Specim Ltd., Oulu Finland),
viewing approximately a 1 cm line
with 30 m optical resolution.
A stationary directing mirror is
used to direct the refected light
90 degrees back to the camera
lens. Te blender with a window
is illuminated with two quartz
halogen lamps which are protected
by another glass window. Te height
of the camera and optical parts can
be adjusted. Tis is necessary to
bring the blend into sharp focus, as
well as adjusting for the diferences
among the diferent types and sizes
of blenders used.
Te blender is a one liter IBC
(intermediate bulk container)
blender, with flling and emptying
ports, and whose emptying port is
equipped with a removable 1-inch
diameter sapphire observation
window. Te blender is attached to
a rotating shaf, which is rotated by
a computer-controlled motor. For
frequent flling and emptying of
the blender, the camera and optical
parts of the blend monitor are
mounted on a base plate, attached
to a slide mechanism to move the
entire front housing out of the way
of the emptying port.
Te data collection sofware allows
the user to view live camera data,
store white and dark references,
and adjust camera settings such
as exposure time and frame rate.
Te blend speed and measurement
resolution can be selected, and blend
rotations, where the image should be
measured, can be specifed.
Te analysis sofware allows the
prediction of the composition of
any of the ingredients for any of the
blender rotations using the Science-
Based Calibration (SBC) [7] or
partial least squares (PLS) methods.
Te SBC method requires the input
of the pure analytes spectra which
can be collected with the system
or imported as a single spectrum
obtained from other instruments.
BLENDI NG OPERATI ONS
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM ANUA|Y 2012 3
Figure 4b. Mean lactose aggregate size for acetaminophen blend
(Outlier if > 1.25 x Prediction Image Nominal %)
Turn 5 Turn 50 Turn 100 Turn 200
Figure 4a. Lactose prediction images
(Rotations)
Te large amount of HS data is
automatically collected, sequentially
arranged by blender rotation,
and analyzed. Te composition
maps provide the frst level of data
compression, resulting in the images
of the predicted compositions of each
ingredient. In subsequent analysis,
the images are compressed into
parameters that are meaningful for
the blending process and displayed as
a function of blender rotation.
Various image analyses can be
performed with the prediction
images, including standard
statistical measures such as image
average, standard deviation,
relative standard deviation, and
the fraction of pixels above/below/
within a certain threshold, and
spatial uniformity measures such
as the distribution of aggregate
sizes. The analysis software allows
the prediction images for selected
rotations and components to be
viewed, compared, and saved.
In one blending experiment, 20%
acetaminophen was blended with 39%
methyl cellulose, 39% lactose, and 2%
magnesium stearate. Te blending
was monitored at every rotation of the
blender up to 200 rotations (Figure
4a). It can be observed that lactose is
evenly blended by about 30 rotations
of the blender, and that there is some
evidence of re-agglomeration in the
80-90 rotation range. Even though
there is some statistical variability
from turn to turn, in this experiment,
lactose seems to break up again over
the consequent hundred rotations.
From the various image-processing
options, the fraction of pixels that
are within range of the nominal
composition as shown in ranging
from zero to one (Figure 4b).
Each pixel is much smaller than
the unit dose and is even smaller
than the usual aggregate sizes,
thus it is a good metric to show the
progression of the blending. Of the
main ingredients, cellulose and
lactose reach their best uniformity
at around 25 turns, whereas the
acetaminophen breaks up more
slowly (Figures 5 and 6), improving
until about 160 turns. Tese
diferences would not have been
revealed using single-point near-
infrared monitoring, which obtains
one average spectrum for each
rotation of the blender. It can also
be observed that in this example the
cellulose and lactose are both slowly
getting less homogeneously blended,
although these changes probably do
not afect the quality of the blend
as much as the API getting more
uniformly distributed.
References
1. Hyvrinen, T., et al. (2007). High speed
hyperspectral chemical imaging.
2. Kemeny, G., et al. (2009). Hyperspectral
monitoring of moving process samples.
FACSS Poster.
3. Kemeny, G., et al. (2010). Hyperspectral
monitoring of continuous pharmaceutical
manufacturing. Transdermal Magazine.
4. Kemeny, G., et al. (2010). Pharmaceutical
blend homogeneity. AAPS Poster.
5. Kemeny, G., et al. (2011). Micromixing
analysis for formulation developers. AAPS
Poster.
6. Ma, H. & Anderson, C. (2007). Optimi-
sation of magnifcation levels for near
infrared chemical imaging of blending of
pharmaceutical powders. Journal of Near
Infrared Spectroscopy, 15, pp. 137-151.
7. Marbach, R. (2007). Multivariate Calibra-
tion: A Science-based Method. http://www.
pharmamanufacturing.com/Media/Media-
Manager/ralf-marbach_PATI061207.pdf
Acknowledgements
Tanks for helpful discussion and feedback
from Steve Hammond, Mojgan Moshgbar
and Jun Huang of Pfzers Process Analyti-
cal Sciences Group of Pfzer, Inc.; Juan G.
Osorio and Prof. Fernando J. Muzzio of
Rutgers University, and John Bobiak of BMS
Analytical and Bioanalytical Research and
Development Group.
BLENDI NG OPERATI ONS
40 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
Figure 5. The fraction of pixels within a threshold increases for acetaminophen as blending
progresses.
Figure 6. Acetylsalicylic acid blend with magnesium stearate added at turn 100.
Fraction of Pixels Within +/- 0.25 Prediction Image Nominal %
Median Aggregate Size (Outlier if > 125% x Prediction Image Nominal %)
(Rotations)
(Rotations)
UNEMPLOYMENT REMAINS high in the U.S. and
other countries, and yet manufacturers struggle to fnd
the skilled workers they need. Its a maddening paradox.
Theres a clear skills gap out there. Young people
especiallymillennialsarent gravitating towards
manufacturing careers and dont see the work as sexy.
Manufacturing has a PR problem with young people,
said John Nesi, VP of Market Development for Rockwell
Automation, speaking recently during a panel discussion
at his companys annual Automation Fair. Millennials
dont have an appreciation for the high-tech nature and
diversity of careers in manufacturing today, he said.
Its not just in the U.S., said another panelist, Tom
Duesterberg, a manufacturing expert at The Aspen
Institute. In both India and China, the quality of
engineers and those going into jobs as line workers and so
on is not adequate. In the U.S., were not training enough
scientists and engineers. We are going to have to focus on
immigration as one solution. But also, basic literacy and
numeracy skills and the ability to be trained are missing.
Skill is a combination of education and experience,
said Mary Isbister, president of the metal fabrication
firm GenMet. Most younger workers she sees dont have
experience, and they dont have basic math and science,
and even the basic work-readiness skills of arriving to
work on time!
Irving McPhail, president of the National Action
Council for Minorities in Engineering, bemoaned the
Engineering Awareness Conundrum. There are just
not enough young people aware of the excitement in
STEM careers, he said. Not a lot of young people know
people who work in these fields. And we need K-12
educators who can impart enthusiasm in these fields.
The end result, said McPhail: We are not producing
the number of engineers required to give the U.S. the
ability to participate in the flat, global world.
Duesterberg said that the U.S. must, as Germany and
Japan have, place more emphasis on technical colleges
and technical tracks in high schools, with support from
government and industry. (North Carolina got this
message long ago. It is a model of collaboration between
educators, government, and industry. Other regions are
catching on as well.)
But two things stand in the way, Duesterberg said.
First, the U.S. as a society places an inordinate emphasis
on the entrepreneurial rather than the industrial.
(Are they mutually exclusive?)
And, despite evidence to the contrary, many of our
most influential thought leaders say the future of our
economy is in the service sector. As Duesterberg and
others have pointed out, successful economies need
diversityincluding a healthy manufacturing sector.
But making manufacturing sexy is an uphill climb. A
better approach would be to tap into a trait that all of us,
millennials included, have in spades: greed.
If sexy just wont work, maybe greedy will. In the
National Association of Colleges and Employers (NACE)
report on the best paying jobs for 2011 college graduates,
engineering dominated the list. Here are the first 10 (with
average starting salary in parentheses):
1. Chemical engineering ($66,886)
2. Computer science ($63,017)
3. Mechanical engineering ($60,739)
4. Electrical/electronics and communications engineer-
ing ($60,646)
5. Computer engineering ($60,112)
6. Industrial/manufacturing engineering ($58,549)
7. Systems engineering ($57,497)
8. Engineering technology ($57,176)
9. Information sciences & systems ($56,868)
10. Business systems networking/telecommunications
($56,808)
Tese jobs top even those in fnance, accounting, and
human resources. Te good salaries, of course, have
something to do with the fact that there are not a lot of
qualifed candidates out there. But one would think that a
starting salary of $60K and a wealth of open job positions
would be enough to raise the eyebrows of even the most
dour and discriminating millennials.
PHARMA VI EW
Can Manufacturing Be Sexy for Millennials?
If not, maybe greed can help overcome manufacturings PR problem
WITH YOUNG PROFESSIONALS.
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM !ANUA|Y 2012 41
BY PAUL THOMAS, SENIOR EDITOR
A STARTING SALARY OF $60K SHOULD
BE ENOUCH TO |AlSE THE EYEB|OWS O| THE
MOST DISCRIMINATING MILLENNIALS.
www.IBCLifeSciences.com/BDPWeek
February 27 March 2, 2012
IIItcr Sar OIeQc Eayfrcrt Sar OIeQc, CA
IBCs 28th International
Antibody Development & Production
Combining Technical and Scientifc Innovations with Industry
Collaboration to Overcome Bioprocessing Challenges
IBCs 3rd Annual
Recombinant Protein & Complex
Biologic Development & Production
Approaches to Overcome the Unique Technical and Scientifc
Challenges of Developing Novel Molecules
IBCs 2nd Annual
Analytical Technologies for
Biotherapeutic Development
Implementing Novel Analytical Approaches and
Technologies from Early to Late Stage Development
to Enable Project Progression
IBCs Inaugural
Biobetters & Biosimilars: Technologies
& Development Strategies
Turning Potential into Realistic Opportunities and
Commercial Viability
IBCs 17th International
Process & Product Validation
Moving Toward a Risk-based, Lifecycle Approach to Ensure
Continuous Validation
IBCs 8th International
Outsourcing Manufacturing
of Biopharmaceuticals
Risk Mitigating and Productivity Improvement through
Successful Partnering & Co-Development with CMOs
IBCs 7th International
Technology Transfer for
Biopharmaceuticals
Maximizing Facility Throughput, Accelerating Timelines and
Ensuring Partnership Success in Technology Transfer
IBCs Inaugural
Biopharmaceutical Raw Materials
throughout the Product Lifecycle
Strategies to Identify, Test, Control and Mitigate Risk of
Raw Material Variability in Upstream/Downstream
Development and Clinical/Commercial Manufacturing
IBCs 9th Annual
Viral Safety for Biologicals
Strategies, Technologies and Regulatory Perspectives for
Detection, Prevention and Remediation in Upstream,
Downstream and Manufacturing Operations
Receive up to $699
off the standard rates
when you register
UsIrQ PrIcrIty Cce
EOPJ2PIAO
Monday - Tuesday Wednesday - Friday
Scan this code with your
smart phone to check out
the Agenda-at-a-Glance
Join Us Follow Us Connect
Join the "Bioprocessing
Professionals" group
Follow @IBCUSA and use the
#bdpweek hashtag
Be sure to "like" IBC Life
Sciences on Facebook.
TABLETTI NG
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM ANUA|Y 2012 43
TABLET SCORING has long been used by oral solid
dosage manufacturersoriginally as a means to prevent
tablet stress fractures, but recently as more of a cosmetic
feature [1]. Bisects, of course, present manufacturing
challenges. Tey may be afected by the tablet cup depth,
band thickness and the intended tablet hardness. As the
tablet size increases or changes, so do the size and type of
bisects that may be placed [2,3].
A score is a useful feature for the patient who, for
instance, wants to switch from a name-brand to generic
product and may need to halve tablets to maintain a
consistent dosage regimen. By some accounts, this is
happening more ofen, as insurance companies and
doctors are increasingly recommending that patients split
tablets for proper dosing or even cost savings.
Scoring has also been an issue in determining
whether a generic drug is equivalent to its reference
product. Whether or not a tablet is scored (and scored
properly) can play a role in the cat-and-mouse game that
originating manufacturers play with would-be generic
competition. A reference drug may have a well-defned
score while a generic product may only have a cosmetic
breaklinein such a case, is the generic truly equivalent?
As one consultant recently stated on a LinkedIn
conversation, scoring is a door to patient non-
compliance. FDA has taken this to heart and has
conducted its own research, fnding that scoring can lead
to discrepancies in tablet content, weight, disintegration,
or dissolution. (If youve split tablets yourself at home,
how ofen have you gotten a good, clean split?)
Last fall, the Agency established draf guidance for
tablet scoring [4]. (See PharmaManufacturing.com for
a link.) Te guidance is intended to provide consistent
and meaningful criteria by which scored tablets can be
evaluated and labeled by: (1) providing a harmonized
approach to chemistry, manufacturing, and controls
(CMC) reviews of scored tablets; (2) ensuring consistency
in nomenclature (e.g., score versus bisect) and labeling;
and (3) providing information through product labeling
or other means to healthcare providers.
Te draf guidances fundamental guidelines and
criteria are:
1. Te dosage amount meant to be achieved afer splitting
the tablet should not be below the minimum therapeu-
tic dose indicated on the approved labeling.
2. Te scored dosage form should be safe to handle and
not pose risk of unintended drug exposure.
3. Modifed release products for which the control of drug
release can be compromised by tablet splitting should
not have a scoring feature.
4. Te split tablet, when stored in standard high-density
polyethylene pharmacy bottles and caps (no seal),
FDA provides more
direction as questions
arise about dosage
consistency.
By Paul Thomas, Senior Editor
Split Decisions:
should meet established stability requirements for a pe-
riod of 90 days at 25 C, plus or minus 2 C/60 percent
Relative Humidity (RH), plus or minus 5 percent RH.
5. Te split tablet portions should meet the same fnished-
product testing requirements as for a whole-tablet
product with equivalent strength. A risk assessment
should be provided to justify the tests and criteria for
product with the proposed functional score.
6. Te scored tablet should be tested using the indicated
patient population to ensure patients can split the tablet
correctly, as labeled.
7. Te scoring confguration of generic drug products
should be the same as the reference drug.
8. New study data on tablet splitability should be pro-
vided during postapproval for any product changes per
FDAs SUPAC guidances.
For more clarity on this divisive topic, we sought
out Dale Natoli, president of Natoli Engineering, who
has 35 years of experience in tabletting and has written
extensively on oral solid dosage forms.
PhM: From an equipment standpoint, what are the lat-
est in improvements in tablet scoring or bisecting?
D.N.: Tere have been recent improvements to a bisect de-
sign commonly referred to as the Pressure Sensitive bisect
to help reduce edge chipping and edge attrition. Te new
design requires a facet or radius eliminating the sharp edge
on the tablet at the bisect and the beginning of the punch
cup. Te Pressure Sensitive bisect design is more common
to the European pharmaceutical industry due to an earlier
adoption of the European Pharmacopeia standards pertain-
ing to uniform dose of a split tablet presented in 2002.
PhM: FDA wants to ensure good splitability for scored
tablets. What are some of the keys or best practices for
manufacturers in this regard?
D.N.: A key practice to adopt is establishing good com-
munications regarding tablet requirements, powder and
compression idiosyncrasies with your tooling supplier.
When a tablet is required to be split providing equal dose,
then careful consideration should be given to the tablet
design. Proper tablet confguration, tablet thickness, hard-
ness, bisect type, and placement play a tremendous role
in achieving a uniform dose of a split tablet. Most tablet
designers have the capability to create a digital model of a
tablet with details of the bisect in relationship to the tablet
thickness. Take advantage of these services when develop-
ing new products or when redesigning an existing tablet.
PhM: How about the monitoring and Quality Control of
scored tablets? Have there been advances?
D.N.: I am not a tablet analytical expert but from my
point of view there have not been advances related to
equipment or testing protocol to assure that a bisected
tablet will yield a uniform dose. On the other hand,
from recent interviews with product development and
QA professionals, achieving a uniform dose of a split
tablet has recently gained more attention to assure com-
pliance to the EU Pharmacopeia and the new proposed
FDA Guidelines to comply with export regulations.
A recent development by Accu-Break
Pharmaceuticals is a unique patented process and
tablet design consisting of two layers. The first or
bottom layer is a non-drug layer and is compressed
f lat and is used primarily as a base for the second
layer consisting of the active powder. The top layer
is compressed using an upper punch with a bisect or
bisects engineered to precisely divide the active second
layer and compress until the bisect slightly penetrates
to the first non-drug layer, assuring a precise and
uniform dose with each segment of the split tablet.
PhM: Finally, one consultant has said, Scoring is just
a door to patient non-compliance and should be done
away with. Your thoughts?
D.N.: There is no question that taking a whole tablet
opposed to a split tablet provides the most accurate
dose of a prescribed medicine. As a patient, I would
feel more comfortable knowing that I am taking the
proper dose of medication by only consuming a whole
tablet. But unfortunately, I dont think we will see
tablets designed to be split leaving us any time soon. It
now becomes the responsibility of regulators to assure
that if a tablet is to be split that it splits evenly and
doses the proper amount of medication.
References
1. Rowley, F. Minimize Tablet Bisect Risk: Part I. http://
www.pharmamanufacturing.com/articles/2006/188.html
2. Rowley, F. Minimize Tablet Bisect Risk: Part II.
http://www.pharmamanufacturing.com/arti-
cles/2006/234.html
3. Tableting Specifications Manual, 7th edition, APhA
Tableting Specification Steering Committee, APhA
Publishing, 2005.
4. FDA. http://www.fda.gov/downloads/Drugs/Guid-
anceComplianceRegulatoryInformation/Guidances/
UCM269921.pdf
TABLETTI NG
44 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
OPERATI ONAL EXCELLENCE
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM ANUA|Y 2012 4S
DESPITE THE cautious optimism following the recent
economic crisis, people who design, manage and main-
tain production facilities are still facing tough decisions.
Competition from emerging economies such as the BRIC
countries (Brazil, Russia, India and China), slow consum-
er response, rising prices for raw materi-
als and uncertainty about the immediate
economic future, are just a few of the
issues that are continuing to challenge
production managers today.
In this context, what can managers do
to ensure the continued viability of their
operations? A certain number of companies have gone
for of-shoring and outsourcing to low-wage countries.
But with production and quality issues becoming more
and more of a concern, this option is beginning to lose
some of its shine, and some managers are hesitant to take
this route as it can complicate an already complex and
over stretched supply chain. Another popular option is
cost cutting, but it would appear that most of the low
hanging fruit has already been picked, with the danger
now being that companies will be
trimming muscle instead of fat.
For pharmaceutical frms the picture
might be a little more complicated in
that, for many, the primary concern is
the development of new products, which
soaks up huge amounts of research and
development. Te actual production is either the last step
in an intricate series of events or outsourced altogether.
An already complicated process would have an added
dimension of complexity in that one is ofen dealing
By Tom McNamara
and Sarah Hudson,
Rennes School of Business,
and Sabry Shaaban,
Groupe ESC La Rochelle
There might be another way to increase output, reduce
idle time or lower the average amount of material in a
given linked set of processes.
with heightened safety and quality
standards, far more stringent than
those found in most other industries.
Many drug manufacturing
procedures involve large amounts
of raw material passing through a
series of processes. Invariably done
as a batch operation, this activity
could also be thought of in terms
of a fowline. Te material fows in
one direction. Tere are a series
of precedent constraints. At each
step some work takes place until
the material in process has passed
through all stages.
In pharmaceutical production,
quite ofen the focus is on
scheduling and capacity planning,
with the goal being to get the most
output out of a given facility, given
a limited amount of resources. But
our research has shown that there
might be another way to increase
output, reduce idle time or lower
the average amount of material in a
given linked set of processes.
FROM STARVING TO BLOCKING
Te aim of this article is to high-
light possible ways for improving
the performance of facilities having
unpaced production lines, where
operators are allowed to work at
their own speed without the aid of
an automated moving belt. We also
take into account that these lines
will necessarily sufer breakdown at
times. Such lines usually have some
storage spaces allotted in between
the individual stations, referred to as
bufers, for storing work-in-progress
(WIP) units. Tis permits smoother
production by avoiding stoppage of
work due to an operator not having
a work piece available (known as
starving), or not being able to pass
an item on to the next station due to
lack of space (termed blocking).
Te importance of bufers should not
be underestimated since they allow
workers to process items relatively in-
dependent of each other. Tis type of
production is called asynchronous.
Tere has been a considerable
amount of attention paid to
achieving a balanced line, where
every workstation along the line
completes its task at equal rates, and
passes the unfnished piece on to the
next station. Some frms have made
substantial investments in time,
money and efort to bring balance
into their lines, but is a balanced line
desirable, or even possible?
If we want to be realistic and
fexible in our line designs, it needs
to be recognized that diferent people
work at diferent speeds, so perfect
balance is a nigh on impossible
dream. Tere are several reasons
for this. One is the type of work
performedsometimes, because
of pre-existing technological or
precedence restrictions, it might
be impossible to break down an
individual job into a number of
simpler tasks where each one has
the same average completion time.
We can imagine, say, a process
where a series of chemical reactions
are followed by centrifuging,
drying, crystallization, and so on
until the fnal packaged product
is manufactured. We cant simply
run one of the reactions for half the
necessary time in order to maintain a
balanced line.
THE HUMAN ELEMENT
Another reason in these manual lines
is human nature. Diferent people
have diferent levels of training,
education, skill, ability and motiva-
tion. For these reasons, the amount
of time it takes for diferent opera-
tors to complete the same task will
never be equal. Contrary to popular
opinion, this is not necessarily a
bad thing. Studies have shown that
an unbalanced line where people
work at diferent speeds can actually
outperform a perfectly balanced line.
By deliberately having at least one
station slower than the rest (a bottle-
neck), higher production rates than
those attainable by a corresponding
balanced line were possible. An oper-
ations management approach known
as the Teory of Constraints views
bottlenecks as a normal everyday
occurrence. By identifying the con-
straining bottleneck station and pro-
viding it with additional resources,
better on-time delivery and overall
performance are possible.
OPERATI ONAL EXCELLENCE
46 JANUARY 2012 1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
Figure 1. The patterns of worker speeds
1. ASCENDING ORDER OF MEAN TIMES 3. INVERTED BOWL SHAPE
2. DECENDING ORDER 4. BOWL SHAPE
Slow
Slow
Slow
Slow
Slow
Fast
Fast
Fast
Fast Fast
Even the slightest improvements in efciency and
operating costs can result in sizeable savings when
considered over the anticipated useful life of a facility.
So how can we design our unpaced lines to take best
advantage of the natural working rhythms of our
employees in so-called unreliable lines that can break
down at any time?
SIMULATED CASE STUDY
We conducted a computer simulation investigation into
production lines having fve and eight stations. We in-
corporated station failure, to duplicate real-life operating
conditions as closely as possible. We assumed that indi-
vidual workers fell into one of three categories according
to their mean (average) service time; that is to say an op-
erator could be fast, medium or slow. Four confgurations
of employee arrangement were looked at (Figure 1):
1. the fastest worker is located at the front of the line,
followed by slower and slower operatorsan ascend-
ing order (/).
2. the fastest worker is placed at the end of the line, pre-
ceded by progressively slower workersa descending
order (\).
3. the slowest operator is positioned in the middle, with
workers getting faster as we go towards both the front
and back of the linean inverted bowl shape (/\).
4. the fastest worker is stationed in the middle, with
operators getting progressively slower as one moves
outward in both directionsa bowl shape (\/).
Another factor that we considered was the percentage
degree of imbalance, i.e. the diference in speed between
any two successive operators. Tree degrees of line
imbalance were viewed: 2% (slight), 5% (medium), and
12% (high). In addition, inter-station bufer capacities
were set at 1, 2, 3 and 6 units.
Our objective was to fnd out which pattern (if
any) of worker allocation would provide the greatest
enhancement in performance by way of: 1) an increase in
the throughput (output) rate; 2) a reduction in the overall
amount of worker idle time; and 3) a decrease in the
average bufer level.
Some of our results were surprising and might run
contrary to prevailing wisdom concerning production
line management (Figure 2). We found that the best
configuration resulting in increased throughput rates
and lower average worker idle times was a bowl-shaped
arrangement (fastest worker placed in the middle).
The biggest improvement, in certain cases, when
compared to a balanced line, was nearly a 2% increase
in output rate and an almost 6% reduction in average
idle time. These might not appear to be significant, but
when taken over the expected lifetime of an assembly
line, higher revenues and increased savings could
prove to be substantial. As for the average bufer
level (Figure 3), the best pattern turned out to be a
descending order (\) of mean service times (i.e., workers
get progressively faster). It was observed that, on certain
occasions, a reduction in average bufer level of almost
83% was possiblea huge amount of saving.
LESSONS FOR MANAGERS
Our findings offer some useful insights into how
managers can improve the efficiency of their facili-
ties. They indicate that a line manager who chooses to
reap the benefits of unbalancing his or her line is faced
with a decision making situation. Would he or she like
to increase output and lower idle times, or prefer to
reduce average buffer levels? This decision will most
likely depend on the type of operating environment
involved. Is product demand high? Is this a technical
product for which highly skilled (and paid) workers
are needed, where specific or complex scientific, legal
OPERATI ONAL EXCELLENCE
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM ANUA|Y 2012 47
Figure 2. Best Idle time and throughput results: An eight-station line with a slight (2%) bowl-shaped imbalance in mean time and a buffer capacity
of one unit
= Pattern of average service time imbalance
Slow Slow
Fast
or health and safety knowledge is required? Here we
may feel that throughput and idle time are of primary
concern. In these cases a bowl arrangement of workers
(\/) would likely provide the best benefits.
Alternatively, managers might fnd themselves
operating in an industry with extremely short product life
cycles, or chemical compounds that are not very stable
at certain steps of the process. Here the goal might be to
keep as little material waiting in bufers as possible, and
correspondingly, a descending order of workers (\)i.e.
workers get faster as you move down the linewould be
the preferred allocation
It should be noted that only a few of the almost
unlimited number of possible alternatives for
unbalancing a line were examined. Caution should be
taken in that if a line is imbalanced in the wrong way,
adverse performance could be the result. In spite of
the caution needed, this is undoubtedly an attractive
proposition to production managers, since the line
can be redesigned simply by reassigning workers along
the line at no extra cost. Te fact that the diversity in
working speeds can actually be taken advantage of
means that every line worker can ofer something to the
whole process and can feel of value to the companya
motivating exercise indeed.
About the Authors
Tom McNamara and Sarah Hudson are professors at
the Rennes School of Business, Rennes, France, in the
Department of Finance and Operations. They can be
reached at the following email addresses, respectively:
tom.mcnamara@esc-rennes.fr and sarah.hudson@esc-
rennes.fr. Sabry Shaaban works within the Department of
Economics Strategy and Organization at Groupe ESC La
Rochelle, Cerege, France, and can be reached via email at
shaabans@esc-larochelle.fr.
OPERATI ONAL EXCELLENCE
1)"3."$&65*$"-."/6'"$563*/(t8881)"3."."/6'"$563*/($0.
Figure 3. Best average buffer level results: A five-station line with a high (12%) descending mean-time order imbalance and a buffer capacity of six
units
Slow
Fast
MEAN TIME DIFFERENCE 12%
= Pattern of avereage service time imbalance
For additional information, please contact Foster Printing Service,
the ofcial reprint provider for Pharmaceutical Manufacturing.
Call 866.879.9144 or
sales@fosterprinting.com
REPRINTS ARE IDEAL FOR:
Q New Product Announcements
Q Sales Aid For Your Field Force
Q PR Materials & Media Kits
Q Direct Mail Enclosures
Q Customer & Prospect
Communications/Presentations
Q Trade Shows/Promotional Events
Q Conferences & Speaking Engagements
Q Recruitment & Training Packages
LET YOUR POSITIVE
EXPOSURE STAND OUT.
MARKETPLACE
CLASSI FI ED
PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM ANUA|Y 2012 4
FOR ADVERTI SI NG SALES OR
PRODUCTI ON I NFORMATI ON,
CONTACT THE FOLLOWI NG
FOR SUBSCRIPTION INFORMATION CALL
888-644-1803 or 847-559-7360
SALES
TONIA BECKER PUBLISHER
tbecker@putman.net
MATT OSHEA ASSOCIATE PUBLISHER
moshea@putman.net
GREG ZAMIN REGIONAL SALES MANAGER
gzamin@putman.net
JEANNE FREEDLAND DIGITAL SALES SPECIALIST
Jfreedland@putman.net
POLLY DICKSON CLASSIFIED SALES/
pdickson@putman.net ACCOUNT MANAGER
HILLARY FREELEY CLIENT SUCCESS MANAGER
hfreeley@putman.net
CARMELA KAPPEL ADVERTISING COORDINATOR
ckappel@putman.net
JILL KALETHA REPRINTS MANAGER
jillk@fosterprinting.com
866-879-9144 ext. 168
Fax 219-561-2033
PRODUCTION
RITA FITZGERALD PRODUCTION MANAGER
rtzgerald@putman.net
630-467-1300, ext. 468
ADMINISTRATION
CARMELA KAPPEL ASSISTANT TO THE PUBLISHER
ckappel@putman.net
630-467-1300, ext. 314
ADVERTISER INDEX
ADVERTISER PG #
3M Purication 2
Beamex 36
Biologics Modular LLC 13
BW Technologies 11
DSM Pharmaceutical Products 3
Emerson Process Management 52
Federal Equipment 29
Festo 25
Gamajet Cleaning Systems 26
Gemu Valves 6
Hamilton 23
IBC Life Sciences 42
Natoli Engineering 4
PDA 51
Sabin Metal 8
Sensient Phamaceutical Coating Systems 17
Viega LLC 49
As a service to our readers, Pharmaceutical Manufacturing
publishes an advertiser directory. Every effort has been
made to be accurate in this listing. However, the
publisher assumes no responsibility or liability for
errors or omissions in the ad index.
THE TORCH IS PAST
(AND SO ARE WELDING AND GROOVING)
Put those welding torches away and pack up that
old roll groover, too. Now, theres a better way to
connect stainless steel pipe with Viega Pro-
Press. The system is faster, safer and more con-
sistent when compared to traditional methods.
Pressing saves labor time and eliminates wasteful
downtime. Plus, Viega offers the industrys wid-
est selection of press fitting sizes (1/2" up to 4"). For more information
or a demonstration, call 866.766.7805 or visit TheTorchIsPast.com.
Viega
Phone. 866.766.7805 EmaiI. insidesaIesviega.com
EQUIPMENT & CUSTOM SERVICES
CENTRIFUGES
Horizontal Peeler Centrifuges
Vertical Basket Centrifuges
856-467-3399
www.heinkelusa.com
Inverting Filter Centrifuges
Newly acquired Comber Nutsche
Filter-Dryers available
ESTABLI SHED 1952
sales@electrosteam.com
Toll Free: 1-866-617-0764
Electric Steam Generators for a
Wide Range of Pharmaceutical
Clean Steam Applications.
VISIT US ONLINE
t%*(*5"-*446&4
t&/&84-&55&34
t$"3&&3$&/5&3
t8)*5&1"1&34
t8&#$"454
t$0..6/*5:-*/,4
Get Reprints of
Pharmaceutical
Manufacturing
from
Foster Printing
Service
Call
866.879.9144
or email
sales@fosterprinting.com
LAST YEAR marked the release (escape?) of FDAs Pro-
cess Validation guidance, which outlines what is needed
for proper manufacture of a pharmaceutical product.
Many manufacturers have been saying, Oh sure, we
would love to do PAT/QbD, but how? Guess the whist-
ling past the graveyard is over now. Te essence of the
Guidance is simply stated (the italics are mine):
A successful validation program depends upon
information and knowledge from product and process
development. Tis knowledge and understanding is
the basis for establishing an approach to control of the
manufacturing process that results in products with the
desired quality attributes. Manufacturers should:
t Understand the sources of variation
t%FUFDUUIFQSFTFODFBOEEFHSFFPGWBSJBUJPO
t6OEFSTUBOEUIFJNQBDUPGWBSJBUJPOPOUIFQSPDFTTBOE
VMUJNBUFMZPOQSPEVDUBUUSJCVUFT
t$POUSPMUIFWBSJBUJPOJOBNBOOFSDPNNFOTVSBUFXJUI
UIFSJTLJUSFQSFTFOUTUPUIFQSPDFTTBOEQSPEVDU
Each manufacturer should judge whether it has
gained sufcient understanding to provide a high degree
PGBTTVSBODFJOJUTNBOVGBDUVSJOHQSPDFTT to justify
commercial distribution . . . Afer establishing and
confrming the process, manufacturers must maintain
UIFQSPDFTTJOBTUBUFPGDPOUSPMPWFSUIFMJGFPGUIFQSPDFTT
FWFOBTNBUFSJBMTFRVJQNFOUQSPEVDUJPOFOWJSPONFOU
QFSTPOOFMBOENBOVGBDUVSJOHQSPDFEVSFTDIBOHF
Currently, companies without a working PAT (QbD)
program are relying on the 1960s approach to product
control, namely, testing a 10 or 20 sample set and either
selling or discarding the product. Afer-the-fact product
testing, of course, is inefective in managing the process.
Numerous people have suggested more samples (cGMP
does call for a statistically signifcant number, afer all),
but the question of how many is signifcant has been up
in the air. However, that may have changed.
Recently, I was made aware of a new(ish) ASTM
standard, E2790-10, for calculating the proper sample
sizes of lots of materials. While I was waiting for a
greater mind (Howard Mark) to examine the standard,
I considered an older, simpler approach: the n for a
batch of 10
6
would be 10
3
and be quite expensive to
perform under the current paradigm of all HPLC, all the
time. My friend and super-statistician, Howard Mark,
explained the application, and I will paraphrase:
Tree key parameters are needed: the fraction of
samples OOS, the fraction of OOS tablets captured,
and the confdence level. One million tablets is a good
approximation of infnity, allowing assumptions
about Normality and other properties (as the number
approaches infnity, tolerance and confdence limits
become the same). Confdence intervals for the normal
distribution for large numbers of degrees of freedom
allows an approximation to the much more complicated
formulas otherwise needed. It shows that 99.99% of the
readings are within 3.7 standard deviations and 99.999%
of the samples are within 4.2 standard deviations. Take
that number as 0.1%.
Tat means, in a batch of one million tablets, 1,000 of
them would be OOS. To a frst approximation, one tablet
per thousand would be at or beyond the 0.001 probability
point of the Normal distribution, corresponding to
3.15 standard deviations. Tat corresponds to (0.001
x 1,000,000) = 1,000 tablets. Measuring 1,000 tablets
allows a chance to capture one of the OOS tabs. Since
the actual number of tablets that would be beyond the
3.15 std. dev. value are distributed according to a Poisson
distribution, measuring only 1,000 tablets gives only a
50% chance that one of the OOS tabs will be found. To
increase the probability of including at least one OOS
tablet among those measured, more than 1,000 samples
need be measured. Tat means a true statistician would
laugh hysterically if told ten or twenty samples are
representative of a 1,000,000-5,000,000-unit batch!
Unfortunately, business as usual would dictate using
HPLC. Changing from the status quo to meaningful
(non-destructive, information-rich) testing will satisfy
both FDA and ASTM. Therefore, the good news is
that we now have blueprints for QbD; the bad news
is that manufacturers have run out of excuses for not
running QbD.
THERAPEUTI C DOSE
Sampling: Good News, Bad News
Changing from the status quo to meaningful testing will satisfy both FDA and ASTM.
50 JANUARY 2012 PHA|MACEUTlCAL MANU|ACTU|lNC WWW.PHA|MAMANU|ACTU|lNC.COM
BY EMIL W. CIURCZAK, CONTRIBUTING EDITOR
A TRUE STATISTICIAN WOULD LAUGH
HYSTE|lCALLY l| TOLD THAT 10 O| 20 SAMPLES
REPRESENT A MILLION-UNIT BATCH.
April 16-18, 2012
JW MARRIOTT DESERT RIDGE RESORT PHOENIX, ARIZONA
The 2012 PDA Annual Meeting is the meeting place this April. The distinguished Program Planning Committee,
made up of your peers, is hard at work to bring you the best content in the industry. They know what you are
concerned about, what you want to hear and who you want to hear it from.
The Best Content in the Industry
Conference Highlights Include:
Two Great Opening Plenary Topics:
Future Benets for Patients: From Discovery
to Commercial Products, Cellular and Gene
Therapies, David Shanahan, President,
Mary Crowley Research Center and President,
CEO and Founder, Gradalis
The Future of Personalized Medicine
Challenges Ahead, Ted Love, MD, Executive
Vice President, R&D and Technical Operations,
Onyx Pharmaceuticals
Plenary Session Two:
The Future of the Biopharmaceutical Industry,
David Urdal, Chief Scientic Ofcer, Dendreon
Financial Analyst Perspective on the
Pharmaceutical Industry, Barbara A. Ryan,
Managing Director, Research Analyst,
Deutsche Bank Securities, Inc. (invited)
Student Call for Posters Abstracts Due
February 6, 2012



Closing Plenary Topics:
Manufacturing
Opportunities and
Challenges in the
Next 10-20 Years,
Matt Croughan, Professor, Keck
Graduate Institute of Applied Life Sciences
Emerging Regulatory Expectations,
Emily Shacter, PhD, Chief, Laboratory
of Biochemistry, CDER, FDA
New this Year: A breakfast Session on Career
Development Strategies
Networking Receptions & Events like the 6th
Annual PDA Golf Tournament at the Wildre Golf
Club & the PDA 6th Annual Walk/Run (beneting
the Phoenix Childrens Hospital)
Post-Conference Workshop: PDA Single Use Systems
Workshop on April 18-19
PDAs Training and Research Institute (PDA TRI)
will be offering eight courses on April 19-20
Hotel activities for the entire family!
www.pda.org/annual2012
EXHIBITION: April 16-17 | CAREER FAIR: April 16-17
POST-CONFERENCE WORKSHOP: April 18-19 | COURSES: April 19-20
The Parenteral Drug Association presents...
Also:
Post-Conference
Workshop
on Single Use
Systems!
C
o
n
f
e
r
e
n
c
e

B
r
o
c
h
u
r
e

J
u
s
t

P
o
s
t
e
d

O
n
l
i
n
e
!
Empower your people and accelerate your business with the Syncade

Suite.
By simplifying routine, non-value-added tasks, Syncade helps your people be more effective and your plant be
more efficient. You can integrate work activities with real-time information, assuring consistent production is
performed with best practices and approved, up-to-date procedures. And the modular, scalable nature of
Syncade lets you add functionality as you need it. Its time to put the focus back where it belongs on your
business scan the code below or go to EmersonProcess.com/Syncade
The Emerson logo is a trademark and a service mark of Emerson Electric Co. 2012 Emerson Electric Co.
My people are forced to waste their
time on monotonous procedural tasks.
I need them focused on right-first-time
production instead.
YOU CAN DO THAT