Moving toward a real-time manufacturing enterprise

March 15, 2022

A real-time enterprise integrates business and management processes with manufacturing processes in real time.

Skkynet's Xavier Mesrobian

Who could have guessed? Did they see it coming?

Back on April 2, 2019, a global pandemic was far from the minds of most of us. But on that day the Norwood plant of Moderna, Inc. received the Facility of the Year Award from the International Society for Pharmaceutical Engineering. Who knew at that moment how critical this facility would become for the rapid development and production of a vaccine that would soon be saving millions of lives?

Before the COVID-19 pandemic, it typically took 10 years or more to create a vaccine. But the combination of urgent need, global scientific collaboration, and new technologies led to an unprecedented initiation of clinical trials, followed by a production rollout of millions of doses—all in less than a year. Although many factors contributed to this success, Moderna can take pride in their award-winning facility that played a significant role.

Integration in real time

What can we learn from this success? For one thing, the Norwood plant stands out as an example of a real-time enterprise, one that integrates business and management processes with manufacturing processes in real time. According to the award description, the “enterprise and process control systems are integrated in a manner that enables flexibility and rapid new product introduction in a highly-automated landscape.”

This integration of real-time process-control data across the enterprise makes possible a digital architecture that is real-time, synchronized, and can be optimized through bi-directional, closed-loop connectivity between IT departments and OT groups. Data coming from equipment or process operations is streamed in real time to business systems like digital-twin models, analytical tools, and artificial intelligence (AI) engines. These systems can then send control commands back to the OT systems in a closed loop, all in real time.

This is a vision of the future for most companies since the norm is to rely on historical data. They might use enterprise-resource planning (ERP) systems, which are not directly synchronized with operations, so the data can be hours to months old. Many also employ a manufacturing-execution system (MES) that is quicker, but adds a layer of cost, complexity and fragility. Typical MES systems only collect data, which then needs to be analyzed separately using other tools. Any feedback loop is commonly a manual process. In contrast, at Moderna’s Norwood plant they have integrated digital technology site-wide, tying the ERP system and production records in with the process-control and lab-management systems.

Is it practical?

Real-time, enterprise-wide performance meets the needs of the times. Enhancing and expanding automation systems keeps costs down as economies shift from labor-rich to labor-scarce. Smarter and more versatile production lines help meet consumer demand for more customized products and sustainable use of resources. And enabling more dynamic and flexible supply chains supports just-in-time sourcing of raw materials and parts.

There are challenges, of course. The three biggest are network security, managing complexity, and providing closed-loop control. Addressing these challenges can go a long way toward becoming a real-time enterprise.

Network security is paramount for accessing data from any kind of industrial system. In the past, using a VPN was considered adequate, but more companies are discovering (the hard way) that VPNs offer little protection from ransomware attacks. One successful exploit exposes the whole network. Instead, security firms and government agencies alike advocate isolating OT and IT networks through DMZs. Critically, all inbound firewall ports on the production system must be kept closed.

At the same time, as the system grows more complex, high performance can be difficult to maintain. IoT projects may look great in pilot but can fail miserably when scaled up to full production. The devil is often in the details. Each of the protocols commonly used for OT/IT and IIoT has its advantages and drawbacks for large-scale projects. Systems should be designed from the beginning with these realities in mind, and thoroughly tested at each stage of implementation.

Finally, for a real-time enterprise to be fully effective, closed-loop control of some kind is expected. To connect suppliers, power AI systems, and fully integrate real-time analytics with production, you'll need to be able to feed supervisory control data back into the system.

A worthwhile endeavor

Your company may not be building a new facility like the Norwood plant right now, but you can still work toward becoming a real-time enterprise. The key is to implement a secure digital architecture that is real-time, synchronized, and can be optimized through closed-loop connectivity between IT and OT groups.

As Moderna knows, integrating business and manufacturing processes in real time offers significant benefits.

Xavier Mesrobian is vice president of sales and marketing with Skkynet Cloud Systems, Inc.