Continued process verification (CPV) in commercial biomanufacturing

To ensure that a commercial biomanufacturing process is in a state of control, life science firms must develop and successfully implement projects that meet continued process verification (CPV) and other monitoring guidelines.

Managers at biotech, pharmaceutical, and medical device companies routinely receive data monitoring directives. There are many challenges in creating and maintaining an effective global monitoring program. Many companies thus develop non-scalable, non-sustainable data monitoring processes.

To achieve predetermined quality monitoring targets, company executives struggle to find the best ways to acquire, deploy, and scale monitoring systems to reach defined quality monitoring goals. This article provides a maturity model to guide companies through the major steps of implementing a global monitoring plan for continued process verification.

Continued process verification (CPV) in commercial biomanufacturing

Image Credit: BIOVIA, Dassault Systèmes

Life science companies are motivated to successfully adopt a monitoring program for several reasons, including reducing process variability, adhering to regulatory requirements, enhancing process predictability and consistency, deepening process understanding, and increasing efficiency and profits.

However, literature concerning specific techniques for an efficient monitoring program in the variable life science manufacturing environment is limited. One notable reference work is “Continued Process Verification: An Industry Position Paper With Example Plan” by the Biophorum Operations Group.1

Throughout BIOVIA’s history of aiding life science industrial organizations in developing and implementing monitoring systems, the team at BIOVIA has assisted companies in all stages of continuing process verification (CPV) implementation.

Each phase includes key tasks and factors required for successful implementation and continuous improvement. This white paper outlines the exact steps for deploying continued process verification. This is a collection of best practices based on BIOVIA’s experience developing CPV for various companies in the life science industry.

The challenges of maintaining a state of control

Over the years, the FDA has started a variety of programs. One of the most recent is "Guidance for Industry — Process Validation: General Principles and Practices."2 Experts and regulatory organizations have consistently emphasized the need for companies to be aware of, understand, and monitor their manufacturing processes. European authorities and the International Conference on Harmonization (ICH) also provide guidelines.

These guidelines share the same themes and give the same advice: to understand the process, reduce variability, be proactive, validate the tools, boost predictability, employ statistically suitable tools, and that team collaboration is necessary. The monitoring rules are not prescriptive, a bonus since they enable companies to devise techniques tailored to their specific organization and process.

However, this can also lead to misunderstandings about definitions and the precise tools, techniques, and resources required. There are also a number of complications that businesses must face when adopting CPV programs.

Being proactive is difficult. In production, investigations are launched retrospectively to determine why batches fail.

Businesses must take time to alter their approach from looking at what went wrong to a pre-emptive, proactive environment. However, it is challenging to find the resources to do this. In reality, both must take place. Companies must investigate a failure if a batch fails. In addition to this, they should also identify the root cause and prevent failures in the future.

When initiating corrective and preventative actions (CAPA), time must be allocated to the preventive and the corrective. The global monitoring maturity model addresses this issue. Businesses are required to aggregate their data to monitor it and must do so in a scalable and automated manner.

After data is gathered, experts analyze it by creating statistically valid outputs and reporting on them. To do this effectively, they should strip the overhead out of the process by automating report and output creation before scaling the process. Professionals in the life sciences industry routinely devote up to 80% of their time to data collection, cleaning, and organization and just 20% to data analysis. The ratio should be reversed so that data analysis takes up 80% of the time.

Life science firms may need to analyze tens of thousands of parameters during the manufacturing process of a single biological drug. These parameters must be organized and easily accessible to scale to that level. To add to this challenge, the life science manufacturing environment is increasingly complex as the use of contract manufacturing organizations (CMOs) expands.

Regulatory agencies have increasingly emphasized the obligations of both sponsor companies and production facilities to follow monitoring requirements.3 The significant risks of a successful deployment mean that management at life science manufacturing companies have a desire to develop cost-effective, global monitoring programs.

FDA guidelines3 state clearly that to achieve success, global monitoring necessitates the collaboration of team members with expertise in quality, manufacturing, and statistical analysis. However, in a global environment, resourcing a collaborative team is a difficult task.

If team members are sourced from internal resources, they are typically occupied with ongoing responsibilities. If hired out of school, they may not have acquired a full range of technical expertise.

The firms' goals may diverge when a cooperative monitoring team includes CMOs in its process. A group of responsible individuals who can spend the necessary time to execute the program must be assembled. They must be able to work together and possess the appropriate technical proficiency and expertise.

It is not enough to maintain a state of control. Life science businesses must show regulatory agencies proof of relevant activities. However, several companies lack proof that they are implementing and upholding CPV. They can create good programs, but if appropriate confirmation that they have achieved key milestones is not provided, regulatory authorities may not have the necessary insight into the process.

These challenges demonstrate the need for a scalable solution that addresses continued process verification.

CPV maturity model: Deploying for success

Figure 1 illustrates a maturity model that can help companies tackle the challenges mentioned above. It is designed to help companies understand their major milestones, objectives, and the required tasks to successfully implement CPV and prove that they have done so.

The maturity model depicted in Figure 1 contains seven stages. Companies may move up and down the working model’s phases as they implement and develop their monitoring initiatives. They will work on many phases simultaneously, and they may overlap.

Seven phases for CPV maturity.

Figure 1. Seven phases for CPV maturity. Image Credit: BIOVIA, Dassault Systèmes

Companies progress through the framework, returning to earlier phases as required to align, scale, and globalize their initiatives. Each phase’s description will emphasize decisions that may accelerate long-term scalability and management supervision.

Phase 1: Develop a CPV initiative

Several companies start their CPV journey with isolated and manual monitoring projects. However, most life science companies desire a fully developed global program with standardized deployment and tools that give management visibility. A change in company culture is often necessary to develop and maintain a global CPV program.

Companies either develop a new CPV initiative or incorporate CPV monitoring techniques into current initiatives, such as quality by design (QbD), quality monitoring, and product robustness. The initiative's determination is specific to the company and largely dependent on internal dynamics.

The CPV initiative should summarize the general rationale, expectations, and procedures. Establishing a company culture that can support the new endeavor is essential. A lack of time for sites to implement CPV, a lack of understanding and acceptance of the need to monitor, and a lack of accountability or clearly defined roles are common difficulties.

Most businesses adopt a top-down strategy, with management establishing expectations of accountability for CPV-related operations to avoid these issues. Leaders in IT, quality, manufacturing, and statistics should be identified by management, along with their roles in the initiative. These leaders may aid everyone in comprehending the significance of creating a CPV program and the necessity of ongoing process verification.

A substantial organizational adjustment is required for a successful initiative. It is vital to offer evidence of major milestones in every phase of the maturity model.

For the initiative and culture phase, key evidence may include developing and amending an initiative document to include CPV monitoring with clearly defined roles and responsibilities, as well as buy-in and support from key stakeholders and executive management. The focus of efforts should be the development of an internal culture that supports change and implementation of a global CPV program.

Phase 2: Technology solutions

Life science firms must make the critical decision on what technical solutions they require to facilitate CPV adoption. It is critical to employ technology solutions to increase the maintenance and scalability of data collection, computation of data analysis and visualization, as well as reporting and dashboards for the results. GMP requirements must be followed while selecting technology to conduct a CPV program.

There are numerous software options available and several homegrown systems. Potential technological solutions should be examined using the following criteria:

  • Usability - How many people will be able to use it? How long does it take to learn something? Is it necessary to have specialized skills?’
  • The capacity to automate data aggregation and cleansing to reduce the time spent compiling it
  • The hardware and maintenance related to maintaining the technology solution
  • Automation of analysis and reporting to decrease the need for manual calculations 
  • The capacity, if necessary, to convert paper data to electronic data in a compatible manner.
  • Using “monitoring by exception” to receive automated notifications rather than manually reviewing each chart
  • Appropriate statistical tools for accounting for diverse processes
  • Dashboards (or control panels) and reporting capabilities to improve management understanding and save overhead for CPV-related data

Diverse technical preferences and budget allocations mean that many organizations struggle to deploy a single global technology solution. They should implement solutions that are in line with their long-term requirements. To keep expenses low, software tools such as Microsoft Excel® are frequently initially used until their feasibility is limited due to a lack of harmonization and scalability.

It is crucial to choose the appropriate solution given the size of the business and its long-term CPV objectives. Companies should provide proper training for employees who will use the technology. The implementation, validation, and training related to CPV technology solutions are common sources of evidence used to show the completion of this phase.

Phase 3: Guidelines/standard operating procedures

Although regulations can be developed before a technology solution is adopted, they are usually changed based on the solution. Life science companies typically vary in the volume of procedural documentation produced and who owns each document.

Four general documents are usually used. The most general is a high-level initiative guideline that outlines standards and protocols for the entire organization.

A monitoring procedures guideline document will usually explain the expected procedures for each site and product. It is not prescriptive, allowing sites to develop their own techniques. A technical statistical document may be written independently from the monitoring methods to demonstrate statistical techniques and discuss statistical process control ideas.

Monitoring plans for specific sites and goods are prepared at the most detailed level to identify the precise procedures, resources, and individuals responsible for each site and product.

The site-level and product-level monitoring plans should detail how the sites and products will implement these procedures. The following major areas should be included in monitoring methods and site/product-level monitoring plans:

  • What parameters (critical quality attributes [CQAs], critical process parameters, business parameters, etc.) will be monitored?
  • What risk-adjustment techniques should be followed?4
  • Which statistical tools should be employed?
  • How will risk levels be established?
  • When should a control chart, run chart, cumulative sum (CUSUM) chart, etc., be used?
  • How should a unique cause variation be dealt with?
  • How should dirty data be adjusted (non-normal, autocorrelation, outliers, small sample sizes, etc.)?
  • Which run rules should be used? When should alerts be expected to be received?
  • How frequently will new data be added to the system? How often will the user perform analysis and examine data visualizations?
  • Who is in charge of configuring and automating monitoring?
  • What steps should be carried out in response to out-of-trend (OOT) alerts?
    • Distinguishing between out of specification (OOS) and out of time (OOT) experimental techniques
    • Documenting actions. Consider the audience receiving the material.

It is crucial to find a balance between giving adequate technical information and not overcomplicating the documents.

The strategies listed below can help make documents more useful:

  1. Employing flow diagrams
  2. Restricting the document’s length
  3. Placing equations in appendices rather than the body of the text
  4. Giving training on statistical process control and monitoring approaches (more than standard statistical training).

Key milestones for providing evidence of usable and appropriate company-level guidelines include:

  1. Annual reviews and updates to guidelines
  2. Ensuring that guidelines/SOPs are detailed and understandable
  3. Informing responsible parties on the site/product monitoring methods.

Phase 4: Standard monitoring

Statistical methods for common monitoring should be developed after the CPV initiative is established, a technology solution is adopted, and guidelines are created. Standard monitoring procedures are typically simple techniques for critical measures like CQAs. These offer quick “wins” that can help businesses understand their process.

Standard monitoring procedures involve setting up trends (run and control charts) and estimates of process capacity or performance for all CQAs. Stakeholder presentations and the setting up of simple small-scale dashboards are frequently included in this phase.

The proper control chart application requires various decisions and may require statistical assistance. For instance, users must perform assumption testing (distribution testing, autocorrelation), choose the best control chart based on the intended distribution, and establish baseline control limits using historical data.5

The proper use of monitoring procedures correlates to how prescriptive and usable the guidelines are and the availability of supportive personnel such as subject-matter experts and statistical support personnel. The incorrect setup of standard monitoring tools usually results in an inappropriate alert or signal, which leads to a lack of system utilization.

Once the fundamental outputs are configured — typically 20 parameters per product/process — they are shown on a simple organizational dashboard. At this stage, these dashboards are rarely large-scale and are often made for a targeted audience to deliver crucial insights.

However, they offer valuable data and permit site-level staff members to understand the monitoring program's worth, which improves their process knowledge. It is also an excellent point to provide information to management and other stakeholders. To gain buy-in and enable a larger group of people to realize the value of the CPV initiative, it is crucial to define stages for a monitoring setup.

When a company uses a manual process (as indicated earlier), the user will arrive at phase 4 of this model and understand that they cannot scale with the guidelines and technology solution they created in the earlier phases. This makes the manual process even more time-consuming and only capable of monitoring a minimal number of parameters.

Companies in this situation usually realize that a spreadsheet-style solution cannot be scaled to incorporate all their products and parameters. They usually move to a technological approach that lowers administrative costs while boosting scalability.

Figure 2 illustrates an instance of when it is necessary to go back through the model to review earlier phases.

It may be necessary to revisit earlier phases in the maturity model

Figure 2. It may be necessary to revisit earlier phases in the maturity model. Image Credit: BIOVIA, Dassault Systèmes

A CQA dashboard, a training roadmap, reports showing the number/percent increase in parameters monitored, the number of OOT/OOS warnings, the decline in OOT/increase in process capabilities, and other statistics are crucial evidence of milestones related to standard monitoring techniques.

Phase 5: Advanced monitoring

When a business has successfully finished the essential procedures, it can go on to more sophisticated monitoring techniques for more technically focused individuals. More parameters can be monitored, including multivariate monitoring, techniques for time series data, and advanced dashboards and reporting.

The objective is to automate this procedure to provide the right information to the right consumer.

Different levels of reports should be delivered for different needs. The strategies listed below are made to help technically savvy people improve process comprehension. They are capable of setting up and automating intricate monitoring and inquiry procedures, including:

  • Monitoring of critical process parameters (CPPs) and other business variables
  • Interactions between CPPs and CQAs
  • Effective practices for dealing with assumption testing; many businesses revisit and change their standards to improve signal and process information.
  • “Golden batch” evaluation
  • Capabilities for continuous monitoring
  • Multivariate analysis software
  • CPV report generation is automated
  • Connect monitoring outputs to standardized OOT investigative analyses
  • Sophisticated dashboards with drill-down functionality
  • Comprehensive trending approaches for niche applications (control charts based on other distributions)

The following are the most common indicators of advanced monitoring milestones:

  1. An increase in the number of parameters and analyses performed
  2. A decrease in the time spent reporting and communicating CPV monitoring information
  3. A decrease in the time spent on OOT investigations
  4. An increase in analysis complexity
  5. An increase in the number of individuals who consume CPV information.

As illustrated in Figure 2, the urge to revisit previous phases frequently occurs at this stage. Companies discover they cannot do advanced process monitoring with their current technology and return to phase 2 to look for a better solution.

Phase 6: Expanded data access

To account for process changes and establish process knowledge, it is vital to decide how data will be introduced to the system after monitoring tools have been installed and automated. A CPV initiative is a dynamic process that will be continuously updated over time.

When a company evaluates the data that it has been processing, personnel will learn valuable information about the manufacturing process. As understanding increases, more questions will be asked. Other parameters can and should be monitored to resolve those issues. The CPV maturity model must be advanced to continuously improve and modify monitoring methods.

Businesses should prepare for growing data volumes after the first implementation and evaluate their average need for additional data to help with resource and budgetary planning.

Multiyear planning and budgets for increasing data demands, reports on changes to data access, and resources committed to sustaining data access and aggregation are typical enhanced data access milestones.

Phase 7: Meta-monitoring

A conclusive conclusion cannot be made based on monitoring a single parameter. Making informed, data-driven business decisions necessitates the integration of hundreds of thousands of statistical outputs and data visualizations to obtain a complete picture of a product, site, and overall company condition.

For example, if a product is manufactured in multiple sites across the world, numerous parameters and tools are monitored at each location that must be combined to determine the product’s status.

Meta-monitoring is the process of combining data to gain a broad understanding of the status of a process or product. Meta-monitoring is the act of monitoring other monitoring processes. For consumers to understand the results, meta-analysis, based on research language, combines different types of analyses. The idea behind meta-monitoring and CPV monitoring is the same.

Figure 3 demonstrates how monitoring for each individual metric must be combined to identify the status of a process step, and then combined again to assess the status of a product or site, and finally, combined again to assess the overall status of all company locations and products.

Meta-monitoring includes parameter step-level, process step-level, and site-level monitoring.

Figure 3. Meta-monitoring includes parameter step-level, process step-level, and site-level monitoring. Image Credit: BIOVIA, Dassault Systèmes

During this step, companies frequently discover a lack of coordination among locations that manufacture the same drug or use similar techniques. Businesses typically use harmonization techniques to better match parameters across sites and products.

Often, in the initial meta-monitoring attempt, only 10–20% of parameters can be compared across sites, so companies commonly return to the previous step to update or add comparison parameters.

Various statistical techniques must be utilized to ensure a true comparison is made while carrying out meta-monitoring processes (such as effect sizes). It is easy to use the simplest comparison methods (e.g., t-tests) and make incorrect conclusions based on false statistical comparisons, such as failing to account for sample size.

These pitfalls are standard and can discourage buy-in from stakeholders since the site/CMO shows poor performance resulting from a statistical artifact rather than an actual difference.

Meta-monitoring is a tool that may be used by commercial-scale manufacturing companies to determine which of their sites, products, and process steps are satisfactory and where changes should be made. Meta-monitoring evidence commonly includes annual company-level monitoring reports that address site, process, and meta-monitoring assessments and recommend data-driven decisions based on monitoring dashboards and reports.


To maintain control, life science companies should implement a CPV initiative. Companies will likely face several challenges in building and sustaining a scalable and sustainable monitoring program.

With the help of a maturity model, businesses may overcome these challenges and effectively complete the critical phases required to build a global monitoring plan for continuous process verification. A well-executed monitoring program helps life science organizations meet regulatory requirements, enhance process uniformity and predictability, and increase process understanding and variability.

Using this platform, companies can efficiently conduct continuing process verification.


  1. An industry position paper with example plan. Continued process verification. ©2014, BPOG – Biophorum Operations Group.
  2. FDA (CDER/CBER/CVM) Guidance for Industry. Process validation: general principles and practices. Jan. 2011, CGMP, Rev. 1. (Accessed 24 March 2015)
  3. FDA (CDER/CBER/CVM) Guidance for Industry. Contract manufacturing arrangements for drugs: quality agreements. May 2013, CGMP. (Accessed 24 March 2015)
  4. FDA. Pharmaceutical quality for the 21st century: a risk-based approach. May 2007. (Accessed 24 March 2015)
  5. Montgomery DC. Introduction to statistical quality control. © 2012, John Wiley & Sons, Inc.

About BIOVIA, Dassault Systèmes

BIOVIA™ provides global, collaborative product lifecycle experiences to transform scientific innovation. Our solutions create an unmatched scientific management environment that can help science-based organizations create and connect biological, chemical and material innovations to improve the way we live.

The industry-leading BIOVIA portfolio integrates the diversity of science, experimental processes and information requirements, end-to-end, across research, development, QA/QC and manufacturing. Capabilities include Scientific Informatics, Molecular Modeling/Simulation, Data Science, Laboratory Informatics, Formulation Design, BioPharma Quality & Compliance and Manufacturing Analytics.

BIOVIA is committed to enhancing and speeding innovation, increasing productivity, improving quality and compliance, reducing costs and accelerating product development for customers in multiple industries.

Sponsored Content Policy: publishes articles and related content that may be derived from sources where we have existing commercial relationships, provided such content adds value to the core editorial ethos of News-Medical.Net which is to educate and inform site visitors interested in medical research, science, medical devices and treatments.

Last updated: Oct 11, 2022 at 9:39 AM


Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    BIOVIA, Dassault Systèmes. (2022, October 11). Continued process verification (CPV) in commercial biomanufacturing. News-Medical. Retrieved on December 03, 2022 from

  • MLA

    BIOVIA, Dassault Systèmes. "Continued process verification (CPV) in commercial biomanufacturing". News-Medical. 03 December 2022. <>.

  • Chicago

    BIOVIA, Dassault Systèmes. "Continued process verification (CPV) in commercial biomanufacturing". News-Medical. (accessed December 03, 2022).

  • Harvard

    BIOVIA, Dassault Systèmes. 2022. Continued process verification (CPV) in commercial biomanufacturing. News-Medical, viewed 03 December 2022,

Other White Papers by this Supplier