All Search Results matching “christie deitz”

Emerson's Christie Deitz and Bristol-Meyers Squibb's Joe Maguire are presenting, Electronic Workflow for a Bioreactor, at the 2010 WBF North American Conference this week in Austin. Their abstract:

Automating workflow and eliminating paper batch records can provide many benefits, including reducing deviations, expediting batch review and release, improving real-time inventory management, and utilizing industry and corporate standards. BMS is currently in the commissioning stage of its new state-of-the-art facility in Devens, Massachusetts. BMS' objective was to create a paperless manufacturing environment. To meet this objective, automation for the facility includes a process control system (PCS) and a manufacturing execution system (MES) system. The project is unique because, to date, it is BMS' most extensive automation of workflow; that is, the manual instructions that might be traditionally done using paper.

The project team learned some valuable lessons with regard to team organization and approach to testing. They also made some key technical decisions around prompting, phase boundaries and recipe design. This paper will explain many of the lessons learned using the bioreactor area of the project as an example.

Joe described the vision to implement a fully automated, paperless solution that supports release by exception utilizing S95/S88 integrated recipe. The systems included a DeltaV control system, Syncade manufacturing execution software (MES), SAP, SmartLab, and Maximo software for enterprise, lab management, and maintenance planning.

To BMS, release by exception means that all exceptions dealt with in real-time and that the batch release needs to verify all exceptions investigated and signed off. Accomplishing this means the process operations are compliant, more efficient, and more visible.

Joe showed workflows around equipment qualification & facilities, materials verification, batch execution, operations review, QA review, and batch disposition. Moving away from manual, paper-based workflows to electronic workflows removed time and mistakes throughout the workflow.

Christie described the how project team executed this electronic workflow effort. She showed the interaction of the MES workflows with the automatic operations driven from the automation system.

On lessons learned for such a large project, overall it went smoothly, considering the large scope of manual workflows. The team took a bottom up approach. For future projects, the team would do electronic workflow and traditional automation more integrally in design and project execution.

On future projects, the team would finalize vision and requirements early and bring in plant operations and manufacturing personnel early.

If your current work processes are causing delays in getting your manufactured products released for sale, you may want to catch Christie Deitz', Electronic Workflow for a Bioreactor presentation. She'll be co-presenting with an automation engineer from a leading pharmaceutical manufacturer at the 2010 World Batch Forum North American Conference, which will be held May 24-26 in Austin, Texas. You may recall Christie from earlier Life Sciences-related posts.

The presentation describes a large, complex project with a complete paperless manufacturing goal that supports release by exception. For those not familiar with regulated industries, regulatory bodies like the U.S. Food and Drug Administration (FDA) require that all exceptions in the manufacturing process be investigated and signed off before product can be released for sale.

Release-by-exception implies that the exceptions can be reviewed and addressed in real-time, not after the fact, which is typical for paper-based workflows. By moving to a paperless workflow, the number of exceptions can be reduced as the manufacturing rules are enforced in real-time. The business benefit is to reduce the time finished products are warehoused waiting for the exceptions to be addressed. From a financial perspective, this reduces inventory costs and increases inventory turnover.

The project included a DeltaV automation system, Syncade manufacturing execution system, SAP enterprise resource planning system, SmartLab laboratory information management system, and Maximo instrument asset management system. All were connected with the Syncade software to drive an electronic workflow process and final electronic batch record.

Christie and her co-presenter offer examples of manufacturing steps, role of electronic workflow, and how it ensures quality and/or expedites the review and release process. An example is the quality assurance (QA) review process. The electronic workflow enables presentation of the electronic batch record (EBR) in a checklist form for QA review. It ensures quality and/or expedites the review process by launching the notification for QA review in real time and providing a view that displays the relevant information in the familiar QA checklist format.

The presentation also shares the S88 (ISA-88) approach taken, specifically the bioreactor process control at the control system level and workflow connections at the MES (manufacturing execution system) level between the software applications mentioned above.

I won't steal the thunder from the best part, lessons learned, other than to share one example of the importance of finalizing the vision and requirements early and documenting them. It's critical for the success of large, complex projects to establish and connect this common vision with the project team and workflow stakeholders to move the project along and minimize rework.

If you're looking for ways to optimize your process manufacturing workflows, you'll want to attend this session on May 26 at 2:15pm.

MP3 | iTunes

Every industry has its special jargon that is like a secret handshake. If you're an insider, you can quickly spot the outsiders based upon their understanding of your industry's jargon and acronyms. For instance, my background was in the offshore oil and gas business back in the mid '80s. We had jargon like pigging (to clean out pipes) and Christmas trees (fittings and valves on the top of well casing which control the well production rate) to name a few.

My friends in Emerson's Life Sciences/Food & Beverage industry center are insiders in their industry jargon. What sparked this rambling opening was when I read a piece in the Asia Food Journal, Tackling CIP Automation with S88. Written by Emerson's Christie Deitz and Yogesh Rathi, they did a great job, right from the start, defining CIP for us outsiders:

Clean-in-place (CIP) is a method of cleaning vessels and lines without disassembling them. It involves delivering solutions of chemical detergents and rinses at specified flow rates and temperatures. Typically, a CIP skid creates the cleaning solutions and routes them to a user skid that requires cleaning.

According to Christie and Yogesh, CIP began in the 1950s as a manual operation. Over time, this process has been largely automated by most food & beverage, pharmaceutical and cosmetic manufacturers. The authors note that many common challenges exist whether CIP is performed manually or automatically. They include:

• Performing similar actions (acid wash, alkali wash) in an efficient way
• Managing CIP timing of available resources like process skids, and supply/return piping paths
• Minimizing CIP time cycle
• Managing distribution headers or transfer panels to process resources requiring cleaning
• For processes with portable CIP skids used in multi-product facilities, providing local control and easy point-of-use connects

Christie and Yogesh describe how the ISA88 (often referred to as S88) standard and terminology for batch control are applied to CIP processes to address these common challenges. The ISA88 model is comprised of both a physical and procedural model.

On the physical model side, one recommendation was to make the distribution headers and/or transfer panels into equipment modules that are independent of the CIP skid and the equipment it connects to in order to clean and sterilize. The path can be managed as a resource, which allows CIP skids to operate in a similar fashion.

The authors pointed out that the physical model helps to drive the procedural model. They wrote:

The sequences that operate the CIP skid equipment become phases on the CIP skid unit, and the sequences that operate on the user skid become phases on the user skid unit. Similar sequences can be modularized or made into reusable, flexible phases. The differences are handled with recipe parameters.

A final key point is that the ISA88 procedural model can be optimized to run phases in parallel to reduce the overall CIP cycle time for processes with fixed CIP skids. Their example:

...while circulating the pre-rinse solution from one of the vessels on the CIP skid to the user, the other vessel on the CIP skid can prepare the acid rinse solution.

Read the article for specific examples of physical and procedural models for fixed and portable CIP skids.

Understanding the jargon around batch processes begins with understanding the ISA88 terminology. Finding and reading articles like this is a good start as is the ISA website's ISA88 page.

MP3 | iTunes

Pharmaceutical Technology magazine published an interesting article by Emerson's Bob Lenich and Christie Deitz. The article, A Look at 30 Years of Change in Pharmaceutical Automation, recounts the changes affecting Life Science manufacturers from the late 1970s though today. I joined the world of process automation in the early 80s as a summer systems engineering intern in offshore oil and gas production and this article brought back some memories of the amazing changes we've seen.

I'll highlight some items from the article to see if it generates any nostalgic thoughts for you.

Although the distributed control system came along in the mid-70s, Bob and Christie note that most life science companies used pneumatic and single-loop electronic controllers. Data was collected manually or with circular and strip charts.

With growing U.S. Food and Drug Administration (FDA) regulations through the late 70s and early 80s, the DCS began to be seen by life science manufacturers as a tool to help comply.

Batch-based automation systems, the first one being the PROVOX system, came out in the early-to-mid 80s to help with sequencing, failure handling, parallel unit operations, and the creation of recipes.

Just a few years before I recall a little collaborative effort between IBM and Microsoft being introduced to the market (wow a 4.77MHz CPU!) This would have some impact in our industry in the following decade as commercially available technologies (COTS) were incorporated.

Toward the later part of the 80s and into the 90s, standards began to play a larger role. ISA-88 (S88), a batch automation standard was important to life science manufacturers. The digital busses including Foundation fieldbus were developing, and Microsoft operating systems began to make their appearance in systems like the DeltaV system. For communications, the OLE for Process Control (OPC) standard became the way to connect Microsoft-based clients and servers--a big improvement over earlier generation DDE communications technologies.

Automation systems became increasingly modular with class-based configurations. These technologies would help the trend toward more modular construction techniques that brought production on-line quicker compared with prior construction and engineering methods.

Regulations continued to advance to try to address concerns around system, production and data management through the balance of the 90s. Efforts began on the ISA-95 (S95) standard to better define the integration of enterprise and control systems.

These regulations had a positive impact in building competency around data security, record security, lot tracking, and overall batch management. The downside was that it placed the focus of life science manufacturers on meeting regulations rather than continually improving their manufacturing operations compared with other industries.

The FDA's Process Analytical Technologies (PAT) initiative addressed this by changing the focus from meeting regulations to improving operation. The FDA's cGMPs for the 21^st Century added in using a risk-based approach to these improvements. As part of this initiative, they encouraged the use of innovative technologies. We've addressed a number of these innovations with respect to PAT in earlier posts.

Bob and Christie closed the article with a note of how flexibility and the integration of automation with the business-level systems is becoming increasingly important as life science manufacturers move from organic-based synthesis to biologics to continue to develop vaccines and medicines to address our health needs.

Update: Thank you Eric for pointing out the error of my ways! The link to the OPC Foundation has been corrected.

At the recent Interphex Pharmaceutical Manufacturing Conference, Emerson's Todd Ham presented on the subject of automating fermentation. Todd acknowledged that Christie Deitz, whom we've featured in several other posts, had a large hand in the development of this presentation and work on the project discussed.

The presentation discussed a recent project done on a large-scale, multi-product biopharmaceutical complex. This project was so successful it recently won the Facility of the Year Award Winner in Project Execution. One of the keys to success was a clear design philosophy established up front. Elements of this philosophy included:

• Fully automated
• Paperless, dock-to-dock using electronic records, operator handheld devices, and barcode scanning
• Consistency for operators based on industry standards like ISA-88 (S88), ISA-95 (S95), and digital bus technologies
• Focus on fermentation as a key process area for the project

A key to success in the project was the close working relationship between the manufacturer and the Emerson Life Sciences project team on the up front requirements and design, and the subsequent module-level and integration-level testing.

The upfront design considered not only the fermentation and recovery processes, but also the full automation required for paperless operations. This design included recipe-level batch control, warehouse management, electronic signatures, and a complete electronic batch record, including the manual processes. These manufacturing processes included material management, container management, filter management and sampling.

The project team applied the S88 standard to control modules looking to identify the common modules and instances for things like motors and valves. At the S88 equipment module level, the team created project wide module templates, area specific module templates, and unique, one-time use equipment modules.

The sampling system and sparger control are examples of project-wide templates. Fermentation agitator control and dissolved oxygen control are examples of area-specific equipment modules. Transfer panels and valve assemblies are examples of unique equipment modules.

At the S88 unit level, the team designed classes and instances based on physical similarity and phases that they use such as batch media, inoculate, ferment, etc. This led to various unit classes for fermentation vessels including seed fermenters, production fermenters, and feed vessels.

From a recipe standpoint, the design grouped phases into operations, then grouped operations into unit procedures, and finally grouped unit procedures into procedures, all again following the S88 standard.

Todd shared some lessons learned from the team. With regard to the modular design approach, the team learned to keep process units the same as much as possible. With similar units, it is also important to make sure the operations are also as uniform as possible. The team cautioned about the overuse of aliases, which reference pieces of physical equipment like valves and motors, in phase logic. By not overusing aliases, but rather relying on equipment modules to handle physical differences, the phase logic could be generically written to handle multiple pieces of similar equipment like process tanks.

Other lessons learned were to plan for the extra documentation required for high levels of modularity and dock-to-dock automation. Like other members of the Life Sciences team have counseled in earlier posts, time spent upfront in planning and testing saves a lot of project backend effort.

The benefits of a complete electronic batch record vs. a paper-based process in terms of faster release of products are pretty clear. It's important to assemble the project team and begin the planning and design early to prepare for the additional effort commensurate with the increased automation required for a successful project.

The growing conversation on the Food and Drug Administration's Process Analytical Technology (PAT) initiative continues. My persistent RSS search on PAT pointed to another great article, this time in Pharmaceutical Technology magazine. The article, The Five Steps to Starting PAT by Jacob Cook, discusses simplifying the process of getting started with a PAT initiative.

The five steps discussed were:

• Pick simple.
• Understand all the details and nuances.
• Evaluate the instrumentation you already have, and the information you can easily collect.
• Understand the appropriate intervals for collecting that data.
• Evaluate the tools available for reading and synchronizing the data.

Just last week we discussed the benefits of applying a structured approach to a PAT initiative to improve opportunities for initial success.

I passed this article by Christie Deitz, whom you may recall from earlier posts on PAT and ISA-88 (S88) projects. Like most initiatives, Christie believes having good data (step 3) is very important. The Life Sciences industry project teams use DeltaV Batch which integrates in a single location the data required for this analysis. This data includes: alarming, continuous and batch history, operator actions and other events. Having this information organized together around batches and campaigns can help identify PAT opportunities.

Where manufacturing execution systems (MES) like Compliance Suite are also used, exception-based reporting can also help with this process of analyzing the data. We discussed using XSL style sheets to do these reports in an earlier post. An example of this exception-based reporting is showing the batch reviewers only the alarm data that occurred during any particular batch run or campaign.

Christie also points out that where manufacturers have already implemented PAT analyzers, they can make decisions in electronic work instructions (EWIs) based on the analyzers' real-time data values to help verify its correct operation. For example, if a PAT analyzer is not reading the expected value based on other operating data, the work instruction can be to have the operator take a manual sample against which to compare the analyzer data value.

Whether you "pick simply" as a starting point or apply a structured methodology to assess the best opportunities to begin, analyzing your existing data is extremely important. The analysis process is less manually intensive when this data is either centralized or logically organized together in some manner to help better identify these opportunities.

Pharmaceutical Manufacturing magazine's On Pharma blog had an interesting post last week entitled Silencing Pharma's cGMP high priesthood. The post pointedly asks:

We all know that cGMPs are essential to safety. But has the industry gone too far with them? Has pharma, in effect, created a high priesthood of cGMP that stymies creativity, and even common sense?

This post references a Pharmaceutical Manufacturing editorial by Emil Ciurczak who makes the argument:

But, in some cases, cGMPs: have become laws to follow without question. Just as many religious people may not really know why they perform certain ceremonies, many scientists today don't know why they do many things. Ask them and you may hear an answer very close to "It is written!"

I ran this by Christie Deitz, a senior principal engineer in our Life Sciences organization that you may recall from an earlier post on executing S88 projects. Christie believes that this viewpoint has many subscribers and has some elements of truth to it.

She reminded me that this is on of the key drivers for the Process Analytical Technologies (PAT) initiative from the Food and Drug Administration. PAT, a risk-based regulatory framework, is one of the initiatives in the FDA's GMPs for the 21^st Century program.

The intent behind this initiative is to break down some of the barriers preventing pharmaceutical manufacturers from using current technologies to improve the quality of the products manufactured. An added benefit is that these technologies can also reduce the cost of manufacturing through improved cycle times and other process efficiencies. One example where this has happened is Baxter's use of model predictive control to improve solvent recovery.

Another example where today's technologies can help maintain high quality is Talecris Biotherapeutics' use of smart devices and digital bus technologies like Foundation Fieldbus to monitor on-line analyzers and generate "GMP critical alarms" in real-time for anything out of tolerance.

While there is some truth to the statement in the On Pharma blog, many in the industry, including the FDA, are working to change the "cGMP high priesthood" mentality. And, many Life Science manufacturers are beginning to take advantage of the risk-based regulatory framework and are using current technologies to improve their manufacturing quality and improve their overall operational performance.

S88, short for ANSI/ISA-88 is a standard for addressing batch process control. This design philosophy for software, equipment and procedures provides a consistent set of standards and terminology for a batch automation project.

I spoke with Christie Deitz who coauthored a paper entitled, Writing a Functional Specification for an S88 Batch Project.

Christie believes that S88 provides many benefits for the project team and project stakeholders. It starts with establishing common structure and terminology for clear communications between the automation, quality control, manufacturing, and the management teams. The nature of the modular standard facilitates object-oriented, class-based designs. This helps minimize documentation by defining requirements only once for the entire class. It also helps improve the consistency of the design. By streamlining many instances into one class it means that design, implementation and testing efforts are reduced which help the project stay on schedule.

She stresses that the key is to use S88 early during the requirements definition. According to GAMP (Good Automation Manufacturing Practice), the functional specification defines the process automation requirements and becomes the basis for the design specifications. The functional specification may include a process description, piping and instrumentation diagrams (P&IDs), process flow diagrams (PFDs), and an instrument list.

Christie encourages process manufacturers to make sure automation or other S88 knowledgeable people are involved early in the process design to make sure the advantages of a modular approach are built into the project. It's also a good idea to include stakeholders from automation, process engineering, production and quality into the creation of the functional specification. This front end work will minimize changes due to misunderstandings by the project stakeholders. These changes become more expensive the later they occur in the project schedule and can delay the startup date.

Process manufacturers have many choices in how to organize the specifications. Christie's experience is that functional specifications should be created for each area, which allows classes to be described within a single document. There are typically five to ten process areas within a process. This allows for a limited number of documents to manage.

By taking this approach Christie and the experts in our Life Sciences organization have helped deliver projects ahead of schedule, which means faster payback on the project.