December 6th, 2005

New Walt Boyes

ISA PAT Conference

About 35 of us have gathered at ISA Headquarters in North Carolina to discuss PAT.

Mike Nager started off by talking about the Food and Pharmaceuticals Division, which has (suddenly, so far as I know) become one of the largest ISA divisions (technical interest groups) with over 1600 members.

Nager pointed out that the world Pharma market is huge, with over $170 billion in NA, and $101Billion in Western Europe.

What is needed, according to Nager, is to reduce time-to-market, reduce scrap and waste of product, increase overall utilization rate, increase manufacturing efficiencies, and improve compliance to regulation.

Enter the PAT initiative. Why Process Analytical Technology? The goal of PAT is, according to Nager and the FDA, to understand and control the manufacturing process, which is consistent with our current drug quality system: quality cannot be tested into products; it should be built-in or should be by design.
Here's the FDA resource website!

Nager introduced Alex Habib.
Walt Boyes
BlackBerry service provided by Nextel
New Walt Boyes

The Four Tools of PAT

Process Analyzers and sensors to provide process "signatures"

Process Control used to 'steer' final productd toward their desirable end-points

Multivariate mathematical tools for simulation, statistical design, pattern recognition

Knowledge Management for continuous process improvement.

Walt Boyes
BlackBerry service provided by Nextel
New Walt Boyes

Habib on PAT

Alex Habib has been working on process improvement "since God was a small boy," as they say down here in the South.

He showed how the PAT tools can be used in a fermenter using on-line analyzers & SPC software.

Then he pointed out that ISA is extremely involved in PAT:
Books, training, standards, certification and, of course, trade shows....
Walt Boyes
BlackBerry service provided by Nextel
New Walt Boyes


Now, Bob Sherman, author of several books on analyzers and sampling systems, is taking us through the state of ANSI-ISA76- the New Sampling Systems Initiative, or NeSSI. ISA/ANSI-76 is titled, "Sample Conditioning Systems for Process Control Analyzers & PAT/Continuous Improvement Processes." He's provided comparisons (in pictures) of legacy sampling systems and what an at-line process control analyzer using S-76 designs.
Walt Boyes
BlackBerry service provided by Nextel
New Walt Boyes

More from ISA's PAT Conference...

Thank God, ISA has wireless Internet in this conference room! My thumbs were beginning to die from doing the "Blackberry Prayer" too long.

Sherman on NeSSI, continuing:

Bob showed picures of actual products illustrating the principles of the S76 sample conditioning system...a modular system designed to maximuze efficiency and minimize costs, developed specifically for the process analytical instrumentation market, incorporating a single block and tubeset architechture for has become the perfect solution for most gaseous and many liquid sample conditioning systems.

The idea is to provide a standard sampling substrate with easy maintenance, lower internal volumne and surface area, external space savings and on-board microprocessors and ASICs (at least in the future).

Sherman went over the NeSSI roadmap and showed what will be coming. The current NeSSI specification is "mechanical: standard format control components." Next will be GenII: "Electrical/transducers/bus protocols, IS Ex). GenIII will provide for standardized microsystems including complete analytical devices and advanced control components.

Although the S-76 standard is new, (2002), there are already substantial numbers of S-76 compliant components:

-flow metres
-switches, sensors and guauges

SP76-based Analyzers:
Spectrophotometer Cell (UV/NIR fiberoptic)
Analyzers on a block: UV/VIS, NDIR, RI

SP-76-based Controllers
Gas Flow (mass basis, all 2-block format, except HoribaSTEC[1 block])
Porter Instrument
Flow Matrix
Honeywell (sensor)

bypass and particulate
cylindrical format membrane separator
TFE membrane

Rotameters (armored and light duty)
Emerson/Brooks makes both, and has 6-12 week or worse delivery
Buehler makes light duty
use common adaptors
King (best delivery, light duty, 2 weeks)
Kobold (both)
Krohne (both

ADI diaphragm pumps
HNP Mikrosysteme GmbH positive displacement (gear pumps)

Circor Tech

These same three companies are the only vendors making NeSSI footprint substrates.

Flow Switches
Delta M (the best, according to Sherman)

Moisture, Water switches


Sensors, Temperature


NeSSI compliant ball, 2 and 3 way; check and relief valves are available

The challenge is to reduce Life-cycle cost of sampling sywtems 35-70%


Integrating analyzers and sample systems
Field mounted and self contained
Smart incorporating diagnostics and control
Open-architecture data systems
Plug and play maintenance
Standard designs for generic applications

Enabling Technologies

Modular sample system components developed in the semicon industry, and standardized by ANSI/ISA-76
Application Specific Integrated Circuits (ASICs) provide the smart components
Open communication architectures: bus protocols provide data "transnipulation."

Waltnote: Whoa! There it is again, folks, end users need open architectures!

Field-mounted and self-contained
--can work anywhere from the poles to the tropics
No analyzer houses
No long sample transport lines
Minimum utilities' usage
Reduced design time, build cost, life-cycle cost

Smart with Diagnostics + Control

--full system information available
--remote modificatio of parameters
--control of P, T F.

Reduced Life-cycle costs
New Walt Boyes

More from ISA's PAT Conference...

Joydeep Ganguly and Real-time Bioprocess Monitoring and Control

"It isn't enough to have a more sophisticated way of telling you that you have a product that is crappy...that's not PAT.
PAT is about consistently creating product that is NOT crappy."

CBER monitors the biologicals industry.

Talecris is a newly-formed company that inherited the assets of Bayer's plasma business.

Automation needs to be put in where it is needed, Ganguly says.

What is the automation pyramid?

---Batch Control and Reporting-------
---Historian, Sequencing---------------------
-------------Monitoring, Alarming-------------------
------------------------Batch Control-----------------------

Field devices(sensors, actuators, and local control)

If you work with PAT concepts you can get far better quality.

As an industry we are very good at finding out what went wrong after we lose a batch. What we need to do is get as good with keeping batches from going bad.

Don't start PAT too big.

Here's a case study, Talecris' Water For Injection (WFI) system
---an online PAT monitoring and automation case study

Before PAT:
Individual skid basis
no historians or local historians with different formats
look and feel widely different between skids
no handshake between skids
manual actuators,local gauges, bench process

After PAT:
Batch control and reporting for entire process
Central data access, one format
local or central control, same look and feel
communication accross skids,
connected and operated by the control system

The bandwidth of PAT includes:
--Monitoring via inline analytics
--collection of data
--analysis and correlation of data (continuous and batch)
--control and mitigative actions in place

We decided to start small...instead of trying to do the entire plant we just decided to do the WFI system...


WFI used 45% of the time in the process
Request by production to use the WFI system
Call QC, take sample from the distribution line
Ensure no other user wants WFI
Check TOC, conductivity and Temperature
Water released at risk
If water experiences high TOC in the interim, the batch is lost

Communication at systems level
Production control system communicates with Utilities control system
online TOC, conductivity, and temperature on utilities control system confirms WFI good
online totalizer verifies usage
messages back and forth logged
realtime quality assurance, valves to production close on "bad" WFI
correlation on information online
entire gamut of PAT deployed

PAT is not, according to the FDA, implementation of process analyzers.

What we got from WFI case study:
Real time quality assurance
WFI management electronic
electronic reporting allows for 'realtime' release of WFI
Implemented and fully validated with 1+ yerar of peroation experience
Reduced manual sampling

Bottom line: $500,000 per year, not including cost of batches saved ($3-4 million a year)

So very simple can have a huge increase for the bottom line.
New Walt Boyes

Joydeep continues his talk...

Batch "fingerprinting"

Having a good batch structure in place is essential for PAT
It helps to identify opportunities by allowing online feedback of "atypical process performance" A 1% increase in yield, identified, is worth $2.7 million per year.

In the absence of a batch structure process does not know where it is--
It is difficult to determine what the "golden batch" is
If we do determine the "golden batch" almost impossible to determine deviations from it...

Safeguards revert to relying on operator knowledge of process

For realtime QA, the DCS is able to detect deviations from golden batch and contorl (administratively or feedback) the process to prevent danger of bad product/yield loss.

If the batch is bad, we used to send it to drain. Now we wait and stop the batch, check to see what the problem is, and if it can be corrected, and if not, then we dump the batch. Just doing this affords us another 1% of increase, or about $2 million per year.

Conclusion and Outlook

Utilization of PAT based on an automation concept and infrastructure
Robust application of inline analysis and controls for a WFI system
Development of an effective online analysis and reporting tool that allows
for batch 'fingerprinting."
Reduced manual testing
Realtime quality assurmance
increased process understanding
reduced cycle time
eventually parametric release of final product.
New Walt Boyes

Finally ISA fills the big job!

From Don Zee

Dear ISA Member,

I am pleased to announce the appointment of Patrick Gouhin as ISA's Executive Director. At a special meeting of ISA's Executive Board today, the Board accepted Pat's nomination from the Search Committee and appointed him to the position. Pat will join the ISA staff on 3 January 2006.

Pat currently is the Chief Operating Officer of the American Institute of Aeronautics and Astronautics. In his 15-year tenure with AIAA, he was responsible for developing organizational strategies; launching numerous new ventures; enhancing the effectiveness of publishing, training, and student programs; and conceiving strategic marketing and communications programs.

Pat has a bachelor of science degree in aeronautical and astronautical engineering from The Ohio State University and a master's in engineering management from George Washington University.

Pat's extensive experience and successes in engineering association leadership, coupled with his educational background, made him a clear choice for this position. His work as a "change agent" at AIAA in a variety of situations is ideal for what we face in our exciting initiatives to re-establish our position in the automation community. Your Board and I sincerely believe that Pat can work effectively with our leadership, members, customers, and staff to lead the organization to new successes in setting the standard for automation.

I want to thank the Search Committee members for their contributions and commitment that resulted in the selection of Pat.

These are exciting times for our organization and I hope you'll join me in welcoming Pat, his wife Terry, and children Nicole and Michelle into our ISA family.

Don W. Zee
ISA President

Congratulations to ISA, and here's hoping Pat can act as the "CHANGE AGENT" ISA needs.

What do YOU think?