9th June 2026
Artificial intelligence is no longer a distant concept for manufacturers. It is already being explored across quotation, production planning, engineering, quality, supply chain and customer service functions. Yet, for many organisations, the gap between experimentation and meaningful operational impact remains difficult to close.
That was the focus of a headline panel discussion on AI and data in manufacturing, curated and organised by manufacturing journalist James Devonshire at this year’s Smart Manufacturing Week, held on 3–4 June 2026 at the NEC in Birmingham.
The session brought together PP Control & Automation CEO Pinaki Banerjee, Sid Sethi from Specscart, John Cook from Groupe Atlantic and Chris Dungey from the High Value Manufacturing Catapult, who moderated the discussion.
The panel explored where AI is currently delivering the greatest impact across manufacturing operations, how businesses can build the foundations for scalable deployment, and what separates successful adopters from those still stuck in pilot mode.
For PP Control & Automation, the discussion reflected a very practical view: AI will only create lasting value when it is connected to real operational constraints, trusted by the people expected to use it, and measured against business outcomes rather than technology ambition alone.
Across manufacturing, there is no shortage of AI pilots. The challenge is that many of them are built in clean, controlled environments that do not reflect the complexity of day-to-day industrial operations.
Manufacturing is rarely neat. Data is often fragmented. Customer drawings can be inconsistent. Specifications change. Legacy systems remain in place. Quality requirements are strict. Supplier information can be difficult to interpret. And, in many areas, human judgement still plays a vital role.
This is where pilots often break down. The technology may work in isolation, but the workflow around it has not been redesigned. Data is not structured enough. Ownership is unclear between IT, operations, engineering and leadership. People do not yet trust the outputs. And too often, pilots are built around tasks rather than business outcomes.
For manufacturers, this creates an important lesson: AI should not be treated as a bolt-on to existing inefficiency. If the process itself is unclear, inconsistent or poorly owned, AI may simply expose those weaknesses faster.
For AI to work at scale, manufacturers need the right operational foundations, even more so than enthusiasm or experimentation.
That means clean, structured and accessible data. It means connected systems. It means clear ownership of processes. It also means governance around risk, quality, customer data and decision-making.
Just as importantly, it requires human verification where judgement is important. AI can support faster interpretation, better consistency and improved decision-making, but in complex manufacturing environments, trust has to be earned.
PP C&A’s own work around AI-enabled BOM extraction and interpretation from complex electrical PDFs is one example of this thinking in practice. The starting point is not “how can we use AI?” The starting point is a very real operational bottleneck: skilled people spending too much time finding, checking and structuring information before value-added work can begin.
That is where AI has immediate potential. It can support BOM extraction, drawing interpretation, quotation support, engineering checks, supplier and component intelligence, quality documentation, production planning and knowledge capture.
These are areas where friction is already visible and where better data, faster interpretation and decision support can release real value.
AI can look attractive at board level and investor level, but adoption is ultimately won or lost with the people who are asked to use it.
For PP C&A, this means involving engineers and operational teams directly in testing, challenging and improving AI-enabled workflows. Engineers have been encouraged to test models until they break, helping the business understand where the technology adds value, where it falls short and where human judgement remains essential.
This is an important cultural point, as AI should not be positioned as a replacement for expertise. Instead, it should be framed as a way to protect, scale and actually make better use of expertise.
In manufacturing, a significant amount of value sits in experience, judgement and “tribal knowledge”. AI can help capture that knowledge, make good practice more repeatable and give skilled people more time to focus on problem-solving, customer value and improvement.
However, there are still gaps to close. Many non-technical teams need more confidence in data. Businesses need to get better at translating operational problems into AI use cases. Leaders need the capability to support adoption. And teams need to understand where AI helps, where it should be questioned, and where human expertise remains critical.
The most successful manufacturers will be those that build trust, involve their people early and make AI part of how work improves.
One of the biggest challenges with AI is measuring return on investment credibly.
For manufacturers, the answer is to move away from vague technology metrics and measure AI in operational terms.
That could mean faster quotation turnaround. Less manual interpretation of documents. Improved quote accuracy. Reduced engineering rework. Better first-time-right performance. Shorter onboarding time for new people. Increased capacity without simply adding headcount. Faster customer response. More consistent decision-making. Stronger knowledge retention.
These are the measures that connect AI directly to operational performance, customer experience and commercial advantage.
For PP C&A, the advantage comes from combining AI with the company’s engineering knowledge, manufacturing experience, supplier intelligence, process discipline and customer understanding.
In that context, AI becomes an enabler of stronger outsourcing partnerships. It can help improve responsiveness, consistency and scalability, as well as support the expertise that customers rely on.
The practical takeaway from Pinaki’s contribution to the discussion is clear: manufacturers should stop starting with the technology and start with the operational constraint.
Businesses should ask where the friction is: Where are skilled people spending too much time on manual interpretation? Where is information difficult to access? Where do delays, errors or rework occur? Where would better decision support release measurable value?
From there, manufacturers can choose one high-friction, high-value workflow and redesign it properly around AI, data, people and measurable outcomes.
That could be quoting. It could be BOM validation. It could be drawing interpretation, engineering change control, supplier risk, quality documentation, production planning or customer onboarding.
The key is to be specific. Scalable AI is built by solving real operational problems, proving value, earning trust and then expanding from a stronger foundation.
As the Smart Manufacturing Week panel made clear, AI is already creating opportunities across manufacturing. But the manufacturers that gain the most will be those that treat it not as a technology project, but as an operational transformation challenge. For PP C&A, that means using AI to strengthen the way knowledge is captured, decisions are supported and customer value is delivered – helping the business become an even stronger strategic outsourcing partner to machine builders and OEMs.
A leading aquaculture specialist – that is changing the way fish are vaccinated safely – has signed a major manufacturing deal.
On paper, defined discipline-specific suppliers can look organised. However, every additional supplier introduces another handoff, and every handoff creates another point where time, quality, communication and accountability can be lost.
Very rarely does growth not surface because an OEM lacks ambition. Shortcomings arise because operating models built to support such ambition don’t evolve quickly enough.
Recent weeks have brought two important industry moments into sharp focus, concluding that demand for AI and automation is rising, but investment, skills, and long-term thinking must follow.
For decades, one question has sat at the heart of operational strategy for machine builders and OEMs: make or buy? It’s a familiar debate and it isn’t the wrong question by any means, but perhaps it is an incomplete one.
In most machine building businesses, change is still treated as an exception. A late-stage drawing revision, component substitution, or wiring tweak discovered during build. Each one is handled, resolved, and signed off. And then everyone moves on. But what if that’s the wrong way to think about it? What if change isn’t the disruption to the system but the system itself?
Our ideas. Now in audio. This series turns our latest opinion pieces into easy-to-access insights, exploring the real-world challenges facing machine builders and OEMs today. Simple. Practical. On demand. EXPLORE THE SERIES ON SPOTIFY In this feature, Tony Hague, Director at PP Control & Automation draws on decades of experience …
Global expansion is on the agenda for PP Control & Automation after it announced a major acquisition drive.
