Manufacturing Culture
Net zero's real bottleneck isn't technology, it's teamwork
Volvo, JLR and Dassault Systèmes panellists agree that organisational silos, not sensors or software, now pose the greatest threat to automotive manufacturing's decarbonisation targets.
Ask three manufacturing leaders where their biggest sustainability obstacle lies, and you might expect three different technical bottlenecks. Instead, when AMS put that question to Volvo Cars, JLR and Dassault Systèmes during a recent livestream on advanced manufacturing and sustainability, the panel converged on something far less technical. Not batteries. Not simulation software. People, and the organisational walls between them.
It is a striking admission given how far digital tools have advanced. Battery passports now trace materials from mine to recycling plant. Digital twins model entire factories before a single robot moves. Yet according to Magnus Olsson, Vice President and Plant Manager at Volvo Cars Torslanda, Jyothi Matam, Senior Technical Leader for Systems Simulation and AD/ADAS at Dassault Systèmes - the livestream sponsors, and Garrett Bell, who leads manufacturing transformation at JLR, none of that infrastructure solves the harder problem of getting different teams to actually work as one.
Technology itself isn't the difficult part. The greater challenge is bringing together different teams, different ways of thinking and different organisational cultures
Breaking down the silos
Matam put it most directly. Having worked with manufacturers across regions and industries, she has observed a consistent pattern. Design engineering optimises for product development. Manufacturing chases production efficiency. Sustainability teams chase carbon reduction. Each discipline makes genuine progress in isolation, yet the sum rarely equals what the parts could achieve together.
"Technology itself isn't the difficult part," she said. "The greater challenge is bringing together different teams, different ways of thinking and different organisational cultures."
Software defined vehicles bring the point into sharp focus, she argued, because they leave no alternative. Control systems, software, electronics, manufacturing and validation simply have to work together, and there is no siloed shortcut through that requirement.
If I can mobilise 6,500 employees to act as sensors, giving them the tools and responsibility to continuously improve the business, I believe that creates a greater opportunity than any single digital system. It's not a question of choosing between people and AI
Turning the workforce into a sensor network
Olsson answered the same question from an unexpected direction. Rather than naming a barrier, he reframed it as an opportunity, one rooted in Volvo's Lean manufacturing culture at Torslanda.
"If I can mobilise 6,500 employees to act as sensors, giving them the tools and responsibility to continuously improve the business, I believe that creates a greater opportunity than any single digital system," he said. "It's not a question of choosing between people and AI."
That philosophy already shapes how Torslanda deploys emerging technology. Its camera based image recognition system, now expanding across roughly 50 assembly facilities group wide, did not originate in a boardroom. It came from the production team itself. Management's role, Olsson explained, was simply to remove the barriers that innovation from the shop floor tends to hit, chiefly around privacy, ensuring faces were digitally obscured whenever footage was recorded.
It is a useful corrective to the assumption that vision systems and AI are principally sustainability tools. Olsson was candid that today they deliver quality assurance benefits rather than direct environmental ones, and that they require heavy retraining whenever a product changes.
Human operators, he noted, use multiple senses to judge quality, while a camera can only inspect what sits in its line of sight.
The technology complements people. It has not replaced the need for them.
No silver bullet, and no shortage of humility
Bell's account of JLR's own journey underlined why that patience matters. Working with Tata Consultancy Services, JLR installed around 800 additional energy meters at its Solihull plant, expanding monitoring from roughly 2,500 to 3,000 data points to more than 5,000 by the project's end. The ambition, in hindsight, ran ahead of what the legacy infrastructure could support.
"We probably achieved around 60 to 70% of what we originally intended," Bell admitted, adding that even comprehensive visibility did not automatically translate into control, since many older assets were never designed for that level of automation.
The lesson has reshaped JLR's approach. Rather than treating Solihull as the template, the company is now building outward from Nitra, its newest and most automated European plant, having improved that facility's energy performance by around 30 to 40% over two years while simultaneously increasing production.
When we began this journey, like many manufacturers, we were searching for a silver bullet, a single solution that would solve every problem. We've learned that there isn't one
Reflecting on the broader search for solutions, Bell was blunt about an instinct many manufacturers recognise in themselves. "When we began this journey, like many manufacturers, we were searching for a silver bullet, a single solution that would solve every problem. We've learned that there isn't one." Momentum across many smaller initiatives, spread across a workforce rather than concentrated among specialists, delivers more than any single fix.
That same instinct is reshaping how JLR uses its own data. The business currently spends roughly 60% of its effort analysing information and only 40% acting on it, a ratio Bell wants inverted so that 90% of effort goes into implementation. Underpinning that ambition is real urgency. Under Group Chief Strategy Officer Balajae Rajan, JLR has brought its net zero target forward from 2039 to 2030, compressing what was planned as two decades of decarbonisation into roughly four years.
The platform that makes collaboration possible
If organisational silos are the condition, Volvo's approach to software architecture offers a glimpse of the cure. Olsson recalled that engineers once calculated the Hardware-in-the-Loop rigs needed to validate the SPA platform underpinning today's XC60 and XC90, and the number was almost absurd.
"We calculated that we would require the equivalent of seven football fields filled with HiL rigs running 24 hours a day, seven days a week," he said.
The successor platforms, SPA2 and SPA3, share a single software architecture across vehicle lines instead. A quality improvement made on the EX60 today applies immediately to future programmes, cutting duplicated validation work and, by extension, the wasted engineering effort that duplication implies. Olsson's warning was equally direct. "The moment you begin branching the software into multiple independent versions, you've created a problem for yourself."
Matam, working from the digital side of things, saw the same shift reshaping engineering roles more broadly. Systems once developed independently, from braking to steering to suspension, now converge inside shared high performance computing architectures, and verification becomes a system level activity rather than a component level one.
What's changing is that systems which were previously developed independently, braking, steering, suspension and many others, are now becoming integrated within high-performance computing architectures. That means control engineers, mechatronics engineers and software engineers all need to work together as part of one complete system
"What's changing is that systems which were previously developed independently, braking, steering, suspension and many others, are now becoming integrated within high-performance computing architectures," she said. "That means control engineers, mechatronics engineers and software engineers all need to work together as part of one complete system."
That requirement, she added, only works if the teams behind each component are already talking to each other, which loops back to the same organisational challenge she raised earlier. Verification and validation therefore become a test not just of engineering, but of culture.
A different kind of sustainability metric
None of this diminishes the value of battery passports, simulation, or energy dashboards. But the panel's closing exchange suggests automotive manufacturing's next efficiency gains may owe less to sensors and more to structure. Silos cost time, duplicate effort, and quietly generate what Matam called digital waste - the unnecessary duplication of files, data and disconnected processes that no amount of hardware can fix on its own.
Measured that way, the most valuable sustainability investment a manufacturer can make might not appear on an emissions dashboard at all. It might simply be the decision to let a production line engineer, a software architect and a sustainability lead sit in the same room before a project begins, rather than after something goes wrong.
