Why connecting the shop floor starts long before the first connection

Shop floor integration projects often fail, but it's rarely the technology that is at fault. Failures occur at the outset because of factors such as unclear requirements, unknown assets, and no defined value goal.

This is despite equipment and automation being more digital than ever. The result is a persistent disconnect in life sciences manufacturing between the shop floor and your MES and wider enterprise architecture.

This matters, too, as manufacturing data is no longer just an operations concern. It is also essential for cybersecurity, data analytics, AI initiatives, and ongoing modernization efforts. In other words, a disconnected shop floor holds back more than just production reporting.

From our experience, the outcome of shop floor integration projects is usually decided long before the first connection is made.

What makes a shop floor  integration project succeed?

Successful shop floor integration projects start by answering critical questions. Examples include:

•  Are the requirements clearly defined?

• What is the value goal of the integration?

• Where will the data be contextualized and stored?

•  Is there a data standard, and is everyone aligned on it?

Success also involves focusing on business outcomes, remembering that connectivity and integration are enablers. Better assurance and decision-making should be the primary objectives.

Why does shop floor integration start with an asset?

While it hasn't always been the case, IT/OT asset management is now essential in life sciences given the critical nature of manufacturing data. For example, with pharmaceutical manufacturing now in the scope of the EU's NIS2 directive, you need to know what you have before you can secure it.

Identifying assets is just the starting point, as for each asset, the inventory should include:

• Connectivity

• Data standards

• Compliance status

• Ownership

• Lifecycle status

• Available interfaces

• The data or process context

The inventory should also identify and highlight cybersecurity gaps, such as assets connected to infrastructure that isn’t secure.

Focusing on the wider objectives is also important, as you are doing more than a one-off audit with an asset inventory. As systems become obsolete, the interfaces need to be managed. The goal is to develop a program that manages assets through their lifecycle taking into account the asset’s interfaces and downstream consumers.

The inventory also establishes a baseline where enhancements can be mapped and obsolescence identified.

How do system boundaries reduce validation effort?

Defined system boundaries contain the impact of change. When production, utilities, building systems, and business areas are logically segregated, with clear responsibilities, a new system will only impact its own area. As a result, well-defined system boundaries enable clearer impact assessment and risk-based validation.

How does standardization simplify shop floor integration?

Standardization is about harmonizing the landscape as much as possible to ensure you have a handful of connection types across ISA-95 levels, rather than 15, 20, or more. Achieving this involves standardizing across native drivers, using open industrial protocols such as OPC UA or MQTT and ensuring consistent data operations. Standardization should also consider interface patterns, information models, naming conventions, and governance.

If calculations are being performed on the data, where this takes place needs to be deliberately designed as part of the asset management framework.

And the overall architecture should be defined independently of any platform. Leading with a platform is a common early misstep that can mask what the project needs. A platform-agnostic design, on the other hand, avoids vendor lock-in and ensures a right fit for your requirements.

How do you overcome resistance to shop floor integration?

Resistance is a given, so change management must run for the whole project. It must also address actual objections. The objections we hear most often include:

“If we change our integration, the compliance effort will be too high. And what benefit are we getting anyway?”

“This is a huge body of work, and we don't have time because of other priorities.”

The answer to objections like these should involve demonstrating the gaps and risks in the current setup and linking them to organizational strategy. This demonstrates what the current reality is costing the business, helping to change the conversation.

Other steps to overcome resistance include:

• Involve all those affected by the changes, including exploring solutions collaboratively and ensuring there are feedback loops throughout the project.

• Involve quality and validation from the start of the project to facilitate the shaping of validation effort in parallel with design.

• Don't leave governance too late in the process. Governance, with clearly defined responsibilities, needs to be embedded with the team well before go-live.

How do you know shop floor integration is delivering value?

Delivering value is a different milestone from going live. And while establishing a single source of truth helps to deliver value, simply connecting equipment is not enough to achieve that goal.

The signs that your shop floor integration project is delivering value include:

• There is a demonstrable increase in visibility and access to the data.

• There is an understanding of where data comes from and an understanding of the context.

• There is trust in the data as it has been tested and is aligned with ALCOA+ data integrity principles.

• The data is used daily because it is visible, sufficiently responsive, and understood.

You will also know the shop floor integration is delivering value when there are clearly linked payoffs. For example, being able to combine data from several data sources with full traceability to support a decision where that wasn't possible before. Moving to review by exception and speeding up batch release processes are also examples of clearly linked payoffs.

Where should you start?

Without a business case, even the most valuable integration project won't get funded. A strong case starts with choosing projects against strategically aligned criteria. This helps win stakeholder buy-in and delivers early value.

NNIT supports life sciences manufacturers through every stage, from strategy and business case development to system integration and deployment to validation and compliant operations. We help organizations in the life sciences sector establish scalable manufacturing integration capability that covers architecture, asset understanding, reusable integration patterns, contextualized data, governance, validation, and compliance. Explore our Digital Manufacturing services or talk to our team about connecting your shop floor.