Mark Featherston is director, quality, strategic programs and global lab services at Avantor.

Mark Featherston

Innovative biopharmaceutical treatments change the lives of patients every day. As biopharma manufacturers bring transformational treatments to market, they face a confluence of challenges, from speed-to-market pressure and higher demand for novel therapeutics to regulatory compliance and increasing focus on sustainable practices.

In the face of these complex challenges, new solutions can help biopharmaceutical manufacturers find new opportunities. Digital transformation allows businesses to use smart, connected devices and then harness the collected intelligence to optimize manufacturing, improve and streamline quality control (QC) and compliance, and reduce resource use.

High-volume, high-data industries, such as oil and consumer goods, were early adopters of digitalization solutions because most sensors were already in place, lowering their initial investment cost. Finding, for example, small savings of $0.05 (€0.05) per unit across 10 million units would result in savings of $500,000 (€455,250) enabling a clear path to a favourable return. Annual savings of 5–30% have commonly been reported. 

Digital transformation in the biopharma industry can streamline manufacturing and help shorten time-to-market. However, adoption of digital technologies in the biopharma industry has been slow. Below are a few “best practice” recommendations to help biopharma companies overcome stumbling blocks to digitalization technology adoption:

Ensure data fusion: from electronic batch records in production to laboratory information management systems (LIMS) in the laboratory to enterprise resource planning (ERP) for inventory, it is essential to ensure an open architecture system that allows the seamless sharing of data between every technology. This will create a data ecosystem with the potential to serve as a digital treasure map of information.

Do not get lost in data, but do not lose what you do not use: biopharma manufacturing can produce an overwhelming amount of data, and it can be tempting to analyze all of these data to guide decision making. Smart analytics and cloud computing provide the capability to determine which data are relevant for a specific process and where there is opportunity for improvement. Consider also that data which are unused now could potentially play a role in future optimization or innovation.

Outline and understand security limitations and data privacy: avoid potentially significant disruptions or compliance issues by spending time early in the adoption process to mitigate security limitations and data privacy issues. A first step is ensuring all relevant information technology team members are involved in new digitalization technology adoption and integration.

Collaborate with experienced digitalization partners: partners, either a business or individuals who excel in their industry or subject matter area, can help provide guidance with their expertise and their technology.

The biopharma industry, in general, has been slower to adopt because biopharma companies did not have the sheer number of data-collecting sensors and instruments that higher-volume goods manufacturers have. Nor did biopharma companies produce the type of high-volume outputs where a savings of $0.05 (€0.05) per unit could scale to yield major savings. In addition, early digitalization technologies lacked the data control and integrity required for life sciences products. In 2019, 40% of biopharma labs in one study had not begun to use digital solutions, and just 37% were piloting new offerings (1).

The COVID-19 pandemic—as well as other challenges, such as supply chain and labour issues—combined with improvement in data integrity helped to drive the move toward digitalization. In 2022, 91% of biopharma labs reported acting toward digital transformation, and 69% reported they were piloting, scaling up, or in a phase of wide-scale adoption (1). While in some labs, manual or paper-based processes remain, the industry is taking significant steps toward capitalizing on the opportunities digitalization offers. The following are some ways digitalization can help biopharma companies improve business operations.

Biopharma companies run very complex processes. As a result, it can be very difficult to examine every single variable in a particular process. Digitalization enables companies to collect large datasets, then use technology to analyze that data, calculate conclusions, and recommend improvements. Digitalization technologies also greatly reduce reliance on manual record keeping. By digitizing, and even automating, information such as batch records or equipment use logs, companies can reduce manual data transfers and decrease the risk of human error.

Smart technology’s ability to provide real-time data also increases efficiency by identifying processes that have drifted outside set parameters. For example, if a certain pH is required for a specific chemistry, an alarm can sound on a lab worker’s mobile phone or personal computer (PC) to alert them that a correction is needed before the batch is put at risk.

Likewise, digitalization technologies can also improve scheduling. When systems for equipment, maintenance work orders, labs, employee schedules, and electronic batch records are digitally connected, an intelligent scheduling system optimizes a production schedule over a specified period, such as two weeks or a month. Likewise, a scheduling system can create better balance between manufacturing and QC. Because it typically takes less time to manufacture a product than it does to test one, the technology could identify optimum ways to move a sample and track all its inputs throughout the workflow, increasing efficiency and shortening test time.

A single batch record can be long and complex, involving significant paperwork and leaving opportunity for human error. Even transposing a number incorrectly or writing the wrong lot number in the wrong place can impact manufacturing. Digitizing these records virtually eliminates those quality concerns.

It also creates manufacturing efficiency because a worker can spend less time on documentation to capture each step. When it is time to review and release the batch record, it can be done electronically, validating and guaranteeing the record contains the correct data. Improving QC systems also supports the need to comply with strict regulatory requirements.

As with companies across many industries, biopharma manufacturers are challenging themselves to meet ambitious sustainability goals. Whether it is sensors that adjust equipment such as dehumidifiers for energy efficiency or digitizing records to reduce paper use, smart technologies can play a significant role in conserving resources. In addition, the ability to optimize processes through digitalization prevents the batch failures that result in scrapped product and wasted resources.

 Special PCs and smart programmable logic controllers (PLCs) are installed at the “edge” of the network to collect, process, and provide context to plant floor data, then send that data to the cloud for analytics. For a biopharma facility, this could be a device that sits in the purification area, connected to the PLC that controls the process. This edge device will collect data from the PLC and assign meaning to that data, including which plant, which area, which line, and which equipment the data are coming from. That contextualized data can be transmitted to the cloud for analytics leading to improvement of that purification process. Optimized recipe parameters could be sent to the edge device, then to the PLC for a better purification run on the next batch.

 Smart sensors can have a significant impact on a biopharma plant floor. For temperature-dependent process variables, smart temperature sensors can be connected directly to another device such as an anti-foam pump so that anti-foam is added proportional to the temperature profile that effects foaming, while also sending data to the cloud. Software can historize and analyze the data, recognizing patterns and issues within the datasets—and it does so significantly faster and with much more pattern-recognition capability than humans. The software could even order more anti-foam when needed. Sensor technology can be used to collect various data points, including pressure, turbidity, humidity, flow, and more.

Sensors connected to digitalization technologies can also provide essential real-time monitoring. This allows biopharma manufacturers to keep processes more effectively on track, especially when working with variables, such as total organic carbon, that are difficult-to-measure outside of the process in the lab. For inventory, smart shelves with embedded sensors can detect when materials on the shelves drop below a predefined number. When connected to an inventory management tool, smart shelves can initiate timely stocking and improve inventory visibility.

 This technology uses cameras or sensors to identify or inspect critical information. In recent years, vision systems have become affordable and much easier to use, providing biopharma manufacturers with significant opportunities to improve production.

For example, a vision system added to a filling line can inspect product labels to ensure lot numbers and expiration dates are correct. The system can also identify labels that are incorrectly placed or are marred by smears or smudges. A vision system can also be used to determine if a bottle cap is cross-threaded. For instance, one camera with two different inspection criteria (height of cap in view vs. expected height can detect front-to-back variance, levelness can detect left-to-right variance) can identify a cross-threaded cap and immediately reject what could become a leaky cap in the field.

Advanced analytic tools for digitalization

Connected devices, such as sensors and vision systems, generate massive datasets when combined with integration of data collected from laboratory information management systems (LIMS) and more. Advanced tools such as artificial intelligence (AI) and machine learning (ML) assess and analyze the data for deep-dive insights, finding patterns and interrelationships or detecting process deviations that would be virtually impossible for a human to detect. These predictive analytics allow manufacturers to make necessary corrections or adjustments before the point of product failure.

The technology can also optimize the manufacturing process to save time, energy, and resources. For example, AI and ML have the potential to optimize a specific process beyond the capabilities of base digitalization technology, such as a manufacturing execution system.

The rise of the single-use (SU) system has proven to be a game-changer in biopharma, and digitalization used in its production has demonstrated benefits to biopharma manufacturers as well as SU suppliers. In the past, bulk material—typically delivered to biopharma manufacturers in pails, drums, or super sacks—required quality assurance testing upon receipt before being subdivided and used in production. Now, materials are delivered in pre-weighed, single-use packages supplied with e-delivery of material documentation.

Because the material in that single-use product is only used for a given production run, the package’s traceable e-data can be automatically incorporated into upstream or downstream operations. The packaging also may allow for quick, nondestructive identification, such as Raman identification, which provides a rich dataset of raw material variability.

By working with SU suppliers, biopharma manufacturers can leverage a supplier’s available datasets without the need to build the database itself. The time savings can be significant. In one case study (conducted internally by Avantor), a manufacturer reduced the process of receiving and preparing production materials from 30 hours to nine.

Some SU suppliers are incorporating transformational technologies into their own processes. By moving beyond data aggregation to more connected operations, SU suppliers can automate data-driven decisions to efficiently advance their own production processes to meet biopharmamanufacturers’ needs.

SU technology production requires the assembly of multiple types of tubing, connectors, sensors, and other components. Other factors impact manufacturing too, including the need to maintain production in cleanroom conditions by technicians wearing personal protective equipment. On-time production and delivery rely on these environments being fully stocked.

Tools such as smart buttons provide a single point of contact for onsite support teams to provide QC inspections or engineering support from outside the cleanroom. Use of smart shelves minimize stockouts and inventory discrepancies. Combined with AI and ML systems, this transforms human-driven inventory management to a data-driven process that generates savings and efficiencies in SU production and QC.

Smart technologies will continue to create opportunities for the biopharma industry. Consider digital twins, a modelling technology that uses AI or ML to virtually test improvements and outcomes. More common in high-volume, high-data industries, digital twins can improve biopharma manufacturing processes. In addition, results from digital twin modelling can identify projects that should be prioritized or rank projects in the order of their ability to deliver the best overall quality, productivity, or customer satisfaction-related improvements. This would allow manufacturers to best determine where to invest capital.

As technology improves, ML has the potential to further optimize processes through prescriptive analytics that go beyond predicting what will happen and make recommendations, as well as predict their potential outcomes (see 

. For example, prescriptive analytics could identify whether or not an agitator that runs faster at the beginning of a batch will produce a specified improved yield. Furthermore, the system could then automatically issue a management of change order that, after human review and approval, would then change the agitator’s setting. 

 Three key stages moving from predictive to artificial intelligence (AI)-driven prescriptive decision making.

Augmented reality (AR) appeared several years ago with the creationof a mobile app that allows users to move around in the real world and interact with AR-generated content. This technology can help support biopharma manufacturer workforce development by being utilized to train operators. For steps that are easier to explain with a video rather than a batch record, the operator can point their phone camera or tablet at the equipment to overlay a graphic that demonstrates how to perform the next step. Thus, AR can serve as a 3D standard operating procedure that helps reduce the risk of human error and, ultimately, save time during production.

Digital transformation technologies allow manufacturers to ensure the product is the process. By optimizing manufacturing, improving and streamlining QC and compliance, and decreasing resource use, the end product will be better—and a life-changing treatment can get to the patient faster.

1. Elicker, J.; Maixner, D.; Fish, M.; Heavey, B. Driving Digitalization at Scale in the Lab. 

www.accenture.com/us-en/insights/life-sciences/digital-labs

Mark Featherston is director, quality, strategic programs and global lab services, and John D. Fisher is director, engineering, global ops; both at Avantor.


Vol. 35, No. 12
December 2023
Pages: 18–21

When referring to this article, please cite it as Featherston, M.; Fisher, J. D. Revolutionizing Biomanufacturing: The Digitalization Advantage. 

Articles

Digital transformation allows for smarter and connected biomanufacturing operations.