From the Sensor to the Server to validate, produce, control, release ... The challenges of the evolution of the automation of the process to its control by its data

The purpose of this article is to highlight the increasingly widespread and elaborate use of computer resources for pharmaceutical production.

Historically, pharmaceutical production has grown from a small-scale production (from the lab work) characterized by a largely manual process to an automated production based on the measurement and monitoring / automated control equipment and control units.

Today, automated/computerized systems can be used to manage processes, collect information relating to process execution, and carry out quality checks, all in real-time.

Industrial development in pharma is evolving towards better control of interactions between products and manufacturing processes. "Demonstrating a better understanding of the pharmaceutical and production sciences can create the basis for a flexible approach to regulation. This degree of flexibility is tied to the level of scientific knowledge provided, "says ICH Q8, a tripartite guideline written in 2005 by ICH.

This is the Quality By Design which aims, in the development of a drug, to better explore the characteristics of molecules, to better control the manufacturing process by exploring the limits and identifying critical parameters. The purpose is to ensure that quality is no longer achieved by reaching a target value, but by a set of values ​​(Design Space), a design space in which production parameters can vary without altering the quality of the final product.

The requirements of production are therefore growing: from development to continuous production, to the integration of online analytical measuring instruments (PAT - Process Analytical Technology):
- Process control: Ergonomics of the operator interfaces, reliability, precision and fineness of the regulations, integration of the different parts of the process for a fluid and controlled operation, the management of the alarms,
- Data generation: relevance, precision and representativeness of process data for development and subsequent use in production
- Integration on the production line of quality control instruments: representativeness of the measurement, precision of the instrument and possibly feedback on process control.

At these production required, it should be added regulatory requirements concerning:
- Management of access certificates,
- Revenue management,
- Data protection,
- Editing and saving production data or production reports,
- The audit trail,
- Electronic signatures.

Et And user requirements for communication and exchange of information with:
- other production management systems,
- Electronic data management (batch reports, log books),
- Centralized management of user access,
- The display of trend curves and the statistical exploitation of data,
- Software maintenance and overall system.

To fulfil these different requirements, the computerised system must include:
- Specific applications: supervision and human / machine interfaces, PLC programs, specific regulation techniques, analysis of analysis from measuring instruments, etc.
- The processing and organization of the different databases,
- Maintenance and diagnostic tools of the production system,
- Communication capabilities and different protocols for communication between IT layers (PCs, servers) and industrial components specific to field activities: PLCs, customer industrial network, standalone equipment, input / output components or components engine control ...,
- Computing power for the smooth operation of applications, statistical processing of past data and management of exchanges between the various modules and system components,
- Storage capacity for data backup, the organization of summary information exchanges with third-party systems.

Systems become more sophisticated, the hardwares components are more and more, better and better, communicating.
Different applications need to coexist and work together through information exchange orchestrated in an order consistent walking, rhythmic and reliable.

So the challenges of validation are to take account of all these elements:
- Control-command systems - Supervision / PLC assembly: robustness of operation, process control, acquisition of production data, traceability of materials involved (continuous production)
- PAT: representativity of the production, analysis of the measurements and calculation of the critical parameters representative of the quality of the process, feedback on the regulation of the process (possible)
- Local databases: relevance of information, accuracy, frequency of acquisition
- System protection: access management, integrity of configuration data and collected on the basis of production activity, traceability of changes and audit trail 21CFR part11
- The interfaces between the different systems (server to server, PC / PLC / Networks, intelligent equipment, intelligent sensors, ...): the various information exchanges, the management of the CPU load
- Data exchange with the customer network: access management, data repatriation (integrity), MES integration
- The maintainability of the system
- Maintenance in operational conditions.

The trend is therefore not towards a reduction in information, nor to the simplification of communication sources and formats.

Data multiplies (equipment, instruments, open systems, link interfaces, servers, ...) or can also bring new parameters for existing processes.

This information provides a better understanding of process parameters for their definition, control of production and release in real-time, but entails the risk of drowning under an excess of information, of having poorly-defined links between the different systems, and ultimately a non-robust process.

The major challenge is to have a design for the different systems that is focused on the operation of the process, the information it generates and its processing.

The transition from an automated system towards data-driven process control brings a new level of complexity and new risks to be considered. The controlled production approach (QBD, continuous validation) will strengthen aspects of production quality assurance.

This development also has repercussions for qualification and validation exercises with the qualification of mechanical aspects, automated systems, information systems and the PAT analytical approach.

Coordination of engineering activities with validation activities becomes paramount to the successful launch into production equipment and process.

The pharma company is looking for a better risk control (reputation risk) via predictive analyzes and contextual proposals of actions.
To be effective in the long-term, the sector is also seeking relevant information in summarized form, particularly for gap analyses and to create complete automated reports (production and qualification of its systems). Traceability is becoming a commercial issue: the specifications are critical points (configuration), as are QA requirements (GAMP5 document/cat4). Traceability is also a determining factor in the global supply chain and a factor in combatting counterfeits

New technologies are indispensable tools for the companies producing tomorrow’s drugs and treatments, provided that these technologies are properly analysed, integrated and controlled while meeting the requirements of the IS, production, QA and maintenance departments:
- Flexible "plug & play" technology
- Wireless technologies
- Pilot Dashboard

- Geolocation of equipment, productions in transit
- Cloud

The pharmaceutical industry is headed towards new challenges in production control.
The industry and its partners will need to understand, share and master the different technologies and their development, whilst integrating the regulatory aspects applicable to each element.
The industry and its partners will need to understand, share and master the different technologies and their development, whilst integrating the regulatory aspects applicable to each element.

Wave50 1 Contributor1


Sales and Marketing Director of BiiON, specialist in the management of real time data in pharma production. After starting an academic career at the Catholic University of Louvain, Anne joined a start-up focused on the development of tools BMP (Business Process Model). Then she joins the BiiON team as Account Manager and Marketing Manager. In technical professions such as BiiON's, good technical knowledge, active listening and understanding of the right need are major assets that have made him progress to become today commercial director, member of the A3P Belgium committee and other networks active in life sciences.