Rafe Swan/Cultura/getty images
As pharmaceutical manufacturers implement automated electronic systems, such as a manufacturing execution system (MES) and electronic batch records (EBRs), there is an increasing need to simplify communication between process systems to improve manufacturing efficiency. The International Society of Automation’s (ISA) procedural model for batch automation, ISA-88 (1), is a common framework of standards that can be used to develop batch automation software. ISA-88 can act as a common language to enable people and systems to work collaboratively to implement automated batch systems. Pharmaceutical Technology interviewed Jonathan Lustri, life sciences industry consultant for Process Systems & Solutions at Emerson Process Management, about recent advances in using ISA-88 to coordinate MES and distributed control system (DCS) (i.e., the system that controls the processing equipment and the process) platforms.
Design standardPharmTech: What is an ISA-88 design?
Lustri: ISA-88 is a standard for implementing batch process automation and includes a procedural model and an equipment model. The procedural model defines the hierarchy of objects used to configure batch processing logic. This model includes a master recipe, procedures, unit procedures, operations, and phases that are configured into reusable software components (i.e., class-based objects) for common actions, such as ‘heat a tank.’ This way, whenever a tank must be heated, regardless of the type of process involved, the code can be used without having to be revalidated.
Coordinating MES and DCSPharmTech: What are some of the challenges in implementing MES and how do ISA-88 standards help?
Lustri: One of the challenges is the management of the ISA-88 procedural model across two different software systems. MES and DCS systems both use this model. The design, implementation, and operation of two different systems, both designed around the ISA-88 model, can be a challenge to interface and to keep coordinated and synchronized during batch execution. It can also be a challenge to decide in which system manufacturing activities should be implemented: the MES or DCS. In the pharmaceutical industry, the MES system is typically used to implement electronic batch records. The effort to transform the existing paper batch record documents into an ISA-88 procedural model and integrating this with the ISA-88 batch model (executing within the DCS) requires significant analysis. To make things more of a challenge, it can’t be assumed that everything currently in the paper batch record document should be implemented within the MES.
The DCS is good at interfacing with instrumentation and values and at performing automatic sequencing and control actions. The DCS is also good at capturing data from process instrumentation. The MES performs operations management. The responsibility of the MES during manufacturing is handling material transactions and managing manual activities that must be done by operators, such as equipment preparation, taking samples, or charging materials into a vessel. These manual activities, which have traditionally been tracked by personnel with paper records, are now often performed electronically using an MES ‘workflow.’ An operator still physically takes a sample, for example, but the MES makes the request to the operator to take the sample and may provide instructions on where to take the sample from, print the label, and require an electronic signature. The challenge is how to coordinate the automated equipment activities controlled by the DCS and manual activities executed by the MES. For example, an operator shouldn’t take a sample until the equipment cool-down phase is completed. Both MES and DCS levels use ISA-88 models for common procedural methods. For a complete system, however, both process control automation and MES workflow activity must be integrated. Another challenge is implementing the systems in a simple way that minimizes change-management requirements.
PharmTech: How can these issues be resolved?
Lustri: Opinions differ on the most rational, cost-effective way to implement both MES and DCS. Over the past few years, I have concluded that the master recipe (procedural model) should be in the DCS, and we are seeing more customers go in this direction. This approach starts with a philosophy to maximize the use of instrumentation and valves to automate the process with the DCS and minimize the manual activities to the extent practical; the MES looks after the remaining manual activities. The master procedural model is designed into the DCS, and the DCS is the ‘quarterback’, requesting the MES execute the manual transactions when they are required. The MES workflows are designed to be product and equipment independent, and the product- and equipment-specific data are passed to the MES workflow as parameters when the workflow is requested by the DCS. Once the MES workflow is completed, the MES simply waits until the DCS requests another manual workflow to be performed. The value in this strategy is to minimize or eliminate the need to coordinate between the MES workflow and the DCS batch. Defining small, activity-based workflows that only are instantiated when needed reduces or eliminates the need for this coordination.
Future advancesPharmTech: What do you foresee as changes or advances in MES and DCS in the near future?
Lustri: One advancement we see is manufacturing companies investing in improving the efficiency of technical transfer from process development to manufacturing. This transfer requires creation of new master recipes and coordination with other manufacturing information technology (IT) systems, such as lab information systems and enterprise resource planning systems. Today, a process development team typically generates a stack of paper and gives it to IT and automation to implement without regard to the impact on system configuration. In the future, we see manufacturing companies implementing standardized data models across process development and manufacturing departments to speed technical transfer. Once standardized data models are agreed to within a company, it becomes possible to implement product lifecycle management systems with data structures that can facilitate transfer from process development into manufacturing faster, with fewer errors.
Reference
1. ISA, ANSI/ISA-88.01-1995 Batch Control Models and Terminology (Research Triangle Park, NC, 1995).
Article DetailsPharmaceutical Technology
Vol. 39, No. 10
Pages: 70–71
Citation:
When referring to this article, please cite it as J. Markarian, "Connecting MES to Process Control," Pharmaceutical Technology 39 (10) 2015.
Drug Solutions Podcast: Gliding Through the Ins and Outs of the Pharma Supply Chain
November 14th 2023In this episode of the Drug Solutions podcast, Jill Murphy, former editor, speaks with Bourji Mourad, partnership director at ThermoSafe, about the supply chain in the pharmaceutical industry, specifically related to packaging, pharma air freight, and the pressure on suppliers with post-COVID-19 changes on delivery.