Preventing sudden failures and unexpected downtime in life sciences organizations starts with promoting a data-driven culture across the enterprise.

Promoting a data-driven culture across the enterprise is what makes today’s most successful life sciences organizations stand out. Efficient operations are only possible when the processes underlying those activities are functional, and this type of reliable operation depends on acute visibility. That visibility is, in turn, reliant on clear, comprehensive maintenance and equipment reliability data. When equipment is down due to an unexpected failure, processes grind to a halt and costs skyrocket. 

For this reason, today’s most successful life sciences organizations are expanding their focus as they now place the same value on their maintenance and reliability data as their operational data. These teams are building a data-driven culture across their facilities as a foundation to a larger boundless automation vision, where democratized data moves seamlessly from the intelligent field through the edge and into the cloud. 

The risks of uninformed maintenance in life sciences organizations

In general, for equipment maintenance practices, life sciences companies have followed suppliers’ recommendations, sometimes more conservatively adjusting them based on the criticality of equipment. However, since these recommendations do not incorporate the unique nature of each plant with its specific operating scenarios (e.g., equipment cycling more or less frequently than the basis for recommended maintenance), these companies are often either over-maintaining equipment or running to failure. 

In one scenario, plants over-maintain when maintenance groups regularly overhaul equipment on time-based plans, replacing consumable parts, and finding and repairing the normal mechanical “wear and tear” issues to stay ahead of any potential problems. While such a system can be effective, it is also inefficient by itself. Performing regular preventive maintenance regardless of equipment condition means teams are often increasing costs by using more parts and materials than required, decreasing availability by taking equipment offline more than necessary, and creating labor issues by focusing key personnel on low-value-added tasks. 

Moving from time-based to data-driven strategies

Time-based plans can also result in running to failure if equipment is wearing out more frequently than expected. When discovered during production, these types of incidents often lead to unexpected downtime due to equipment failure, which means costly production halts while teams scramble to bring their equipment back online. Such failures also create a high risk of additional costs through the need to expedite parts or evaluate deviations which may result in loss of raw materials, intermediates, or even entire batches. 

In either case, time-based plans are not optimal, but when plants implement a data-driven culture to drive more informed maintenance, they can eliminate the need to over-maintain equipment or run to failure, decreasing the cost of parts, materials, and labor, while increasing process availability. Moreover, they also eliminate the worry of unexpected failures, decreasing deviations and their related investigations and product loss, and impact on production capacity. 

More, better data is the key

To improve complete and on time delivery of products from their facilities, today’s most successful organizations are evolving their preventive maintenance programs into a data-driven, holistic, enterprise-wide asset management solution. In fact, many of the elements that reliability teams need to drive such an evolution are already in place. 

Most life sciences plants already have smart instrumentation installed to monitor their operations, and many also have intelligent field devices to monitor reliability. For those that do not, the wide variety of intelligent wireless sensors available today make it easy to quickly and cost-effectively add points of measurement anywhere in the facility. 

Smart instrumentation is supported by the plant’s control system, which provides context to the critical data passing through it. When sensor data is coupled with situational information provided by the control architecture, both human technicians and high-level analytics software can make more sense of the available data. These systems work in tandem with other critical analytics tools such as edge analytics devices, device management software platforms, and machinery management software to provide a comprehensive, holistic view of the health of the plant and its assets.

Harnessing the power of integrated data

The most critical step, and the one most plants struggle to achieve, is bringing the data from all of these solutions together, with the right context, to a higher level where they can be used and analyzed, both by powerful software packages and cross-functional teams. The first part of this solution is organizational— moving from a vision of maintenance and reliability as a site function to a corporate function. At the enterprise level, organizations can more easily define the data-driven work culture needed to support more proactive solutions.

The second part of the solution is technical, grounded in a comprehensive data layer, built on an industrial data platform to standardize information from a wide array of sources, while also making it more easily accessible. Industrial information management systems connect to, collect, clean, and contextualize data. 

This data is then prepared for use in the machine learning tools that warn teams in real time when and how their equipment will fail, while providing actionable advice to remedy aberrations well before they cause unplanned downtime. Armed with data and the organizational support to act upon it, maintenance and reliability teams across the enterprise can stop performing reactive maintenance by heading off problems well before they become major issues (Figure 1).

Figure 1: When intelligent field devices and the control system pass data into an industrial data platform, teams can use machine learning tools to turn that data into actionable information and intervene before anomalies cause damage.

Taking the first steps toward data-driven culture in life sciences

In an industry focused on constant digital transformation to drive the newest, most cutting-edge treatments, the spreadsheets and cumbersome work order management systems of the past are no longer sufficient to maintain competitive advantage. Securing speed-to-market, reliable supply, and operational efficiency starts with a data-driven culture, and that culture is supported by the tools teams need to democratize data so plant personnel can make better, faster decisions. This type of improved data foundation is not out of reach, with the necessary tools and guidance available today to those organizations ready for the next step on their digital transformation journey.

About the Author

Kristel Biehler is vice president of life sciences for Emerson’s process systems and solutions business, where she leads the day-to-day business activities in sales, operations, and technology that serve the life science industries. In her previous role at Emerson for five years, she was the automation solutions vice president of sales for the western United States, where she led teams that helped customers identify, architect, and implement automation and digital strategies across a wide range of industries. Biehler started her career with Emerson in 1998, and she holds a bachelor’s degree in mechanical engineering from the University of Utah. Prior to Emerson, she worked for Sorex Medical as an automation engineer.