Bridging process reality and digital systems in pharma – A talk with automation and validation specialist, Smiriti Gupta

Across the pharma and biopharma sector, digital transformation has become essential. Manufacturing Execution Systems (MES), electronic Batch Manufacturing Records (eBMR), and process automation promise improved efficiency, better data integrity, and streamlined compliance. Yet many initiatives struggle to deliver value because they digitalise processes without truly understanding them.

We've caught up with Smiriti, Senior Consultant at NIRAS, who specialises in Process Automation and Computer System Validation (CSV), to explore how her hands-on approach helps clients bridge the gap between shop floor reality and digital systems.

Smiriti helps pharmaceutical manufacturers transform paper-based operations into compliant, efficient digital manufacturing environments. With approximately 13 years of hands-on experience from bioprocessing and automation, she brings a clear philosophy: Successful digitalisation isn't about implementing the latest technology - it's about understanding the process deeply enough to make that technology work in service of operators, quality, and patients.

What drew you to work in pharma manufacturing initially, and what made you stay in this field for over 13 years?

"As a Biotechnology engineer, pharma manufacturing was the closest way for me to work with real biological systems - and, honestly, the "special gowning" and the energy of the shop floor pulled me in more than any lab ever could. I loved the structured chaos of reactors humming, utilities running, and batches moving in their rhythm.

Very early, I realised something important: cGxP and batch discipline aren't constraints. They're the scaffolding that lets you build processes people's lives depend on. Over the years, I moved through upstream, downstream, sterile filling, utilities, and automation, and I saw how small design decisions ripple into operator workload, data integrity, and regulatory exposure.

Thirteen years later, I'm still here because the floor continually teaches me something new."

Can you describe a specific moment on a manufacturing floor that fundamentally changed how you approach digitalisation projects?

"I was supporting a fermentation line when an operator managed a deviation in a way the MES design hadn't anticipated. The paper BMR contained a conditional step tied to an equipment signal - a tiny nuance - but the draft MES workflow had simplified it into a generic task.

At that moment, I understood that you cannot digitalise what you haven't walked, observed, or questioned. Since then, my rule has been simple: Before digitalising, trace every step from equipment signal to operator action. Digital workflows must behave like the real plant, not like the idealised version on a whiteboard."

You mention that "paper processes are often digitalised without being understood." Can you walk us through a concrete example - and how you helped solve it?

"A client once handed us hundreds of pages of BMRs for a "one-to-one" eBMR conversion. The issue wasn't the pages - it was everything hidden between them: parallel tasks, exceptions, operator judgement calls, and informal QA pauses.

We paused and did a process-first redesign. We stripped the process down to its essence, rebuilt the workflow using ISA-88/95 models, clarified equipment interfaces, and wrote URS and FS that reflected how the batch actually flowed. Only then did we begin system configuration.

The result was a cleaner, standardised workflow with built-in review-by-exception and compliant data handling. Batch review time dropped drastically because the workflow finally matched operational reality."

When you first meet with a client who's struggling with their digital transformation, what are the first three things you look for on the shop floor?

First, I look for the true decision points - where operators hesitate, check an instrument, or call QA. These moments reveal the real control points in the process.

Second, I trace how data is generated and recorded. Who signs? Under what conditions? Which system state does it correspond to? This is where ALCOA+ either lives or breaks.

Third, I examine the handshake between equipment and systems: what signals exist, how they're structured, and whether the workflow makes use of them. If these three are clear, everything else becomes solvable.

What's the most common misconception pharma companies have about implementing eBMR systems?

The misconception is that eBMR is a software implementation. It's not - it's a process and compliance transformation that uses software.

When companies treat it like coding, complexity gets pushed downstream and shows up as rework, change controls, and operator frustration. When treated like manufacturing, understanding is built upstream - and the implementation becomes smoother and more compliant.

You work at the intersection of operators, QA teams, automation engineers, and IT departments. How do you navigate situations where these groups have conflicting priorities?

I translate every perspective back to the process. Operators need usability, QA needs traceability, automation needs accurate states, and IT needs stability.

When we put the batch path at the centre, decisions become simpler. Anything that improves product quality, operator clarity, and audit readiness wins. This shifts conversations from "my requirement" to "the best requirement for the batch."

 Based on what you're seeing right now in 2025–26, what's one area where pharma companies are still struggling with digitalisation – and what's your advice for addressing it?

One of the biggest struggles I see today is not a lack of technology, but a lack of cohesion between systems, data, and decision‑making. Many companies have invested heavily in automation, MES, historians, and analytics platforms, yet they still struggle to turn data into something that meaningfully supports day‑to‑day operations and quality decisions.

In practice, this often shows up as fragmented digital landscapes. Equipment generates vast amounts of data, systems collect it, but teams still rely on manual checks, spreadsheets, or parallel paper processes to understand what is actually happening in a batch. Digital tools exist, but they don’t consistently reduce uncertainty, effort, or review time.

My advice is to shift focus from “implementing systems” to “enabling decisions.” Digitalisation should help operators, QA, and engineers answer questions faster and with more confidence: Is the process in control? Do we need to intervene? Can we release this batch safely?

Companies that succeed are the ones that treat digitalisation as a way to improve visibility, consistency, and responsiveness — not just compliance or efficiency. When digital systems support how people actually work and decide, value follows naturally.

If you could give one piece of advice to a pharma company just starting their digital transformation journey, what would it be?

Be very clear about why you are digitalising before deciding how.

Many organisations jump into digital transformation with ambitious roadmaps, multiple tools, and strong vendor involvement — but without a shared understanding of what success should look like for operators, quality teams, and the business. This often leads to complex solutions that are technically impressive but difficult to use, maintain, or scale.

My advice is to start by defining the outcomes you actually need: faster batch release, better data integrity, fewer deviations, easier inspections, or improved scalability across sites. Once those goals are clear, digitalisation becomes much more focused and pragmatic.

It also good to start small and learn. Pilot solutions, learn from early implementations, and build internal confidence before scaling. Digital transformation is not a one‑time project — it’s an ongoing capability.

How is it to work on a project like this in NIRAS - in terms of how we work together and our attitude towards projects?

Working directly with bioreactors, fermenters, and CIP/SIP systems taught me that biological systems don't behave in clean, linear lines, and operators rarely make robotic decisions. That lived experience became the lens through which I approach automation and digitalisation.

At NIRAS, this alignment between process truth and system logic is a principle I apply every day. The multidisciplinary teams here reinforce this process-led mindset. We're encouraged to dig deep into the shop floor reality before designing systems, and that's what makes our solutions work in practice, not just on paper.

Outside of work, what keeps you engaged and energised? Does anything from your personal interests inform how you approach professional challenges?

Curiosity and movement keep me energised. Living across India, Ireland, and now Denmark taught me to adapt quickly and embrace new environments. Travel, food, and exploring cultures have shaped how I handle challenges - more with openness than hesitation.

That adaptability translates directly into my work: Staying flexible, learning fast, and welcoming new ideas.