“Industry 4.0” burst onto the scene around 2015. The fourth industrial revolution. Smart factories. Connected everything. Manufacturing would never be the same.

It’s been a decade now. Time for an honest assessment: what’s actually happened?

What the vision promised

The original Industry 4.0 vision, as articulated at Hannover Messe and in countless presentations since, promised:

Fully connected factories: Every machine, sensor, and system communicating in real-time.

Cyber-physical systems: Digital models linked to physical equipment for monitoring, simulation, and optimisation.

Autonomous decision-making: AI systems making production decisions without human intervention.

Mass customisation: Producing individualised products at mass-production costs.

Transparent supply chains: End-to-end visibility from raw material to customer delivery.

Self-optimising production: Systems that continuously improve without human intervention.

The imagery was compelling: lights-out factories humming along, robots and AI orchestrating everything, humans only needed for strategic oversight.

What actually happened

The reality in 2025 is more modest, though still meaningful.

Connectivity improved substantially

Most new industrial equipment comes network-ready. Retrofitting older equipment with IoT sensors is now commonplace. The ability to collect and centralise production data has improved dramatically.

This is genuinely valuable. Visibility that required walking the floor now comes through dashboards. Problems that went unnoticed for shifts now trigger immediate alerts. Data for analysis that didn’t exist is now abundant.

But “everything connected” remains aspirational for most operations. Legacy equipment, competing standards, and integration complexity mean many factories still have significant blind spots.

AI found practical niches

AI applications in manufacturing have progressed, as I’ve written about extensively. Predictive maintenance, quality inspection, process optimisation, demand forecasting—these work and deliver value.

But “autonomous decision-making” overstates current capabilities. AI advises; humans still decide. AI handles specific tasks well; general-purpose factory AI doesn’t exist.

The AI that’s working is narrower and more focused than the vision suggested—but that’s not a bad thing. Practical, bounded AI solving specific problems is more useful than theoretical general AI.

Digital twins are partially delivered

Digital representations of physical systems are increasingly common, especially for complex or valuable equipment. Simulation capabilities have improved.

But the vision of complete digital twins—perfect real-time mirrors of entire factories—remains rare. Most implementations are partial: a twin of critical equipment, not the whole operation.

The value has come from focused applications (training, simulation, remote monitoring) rather than comprehensive digital replication.

Mass customisation is limited

Some industries have made progress on flexible manufacturing that handles product variation efficiently. Automotive paint shops can handle any colour without setup time. Some consumer products are customisable at scale.

But true mass customisation—economically producing one-of-a-kind products—remains elusive for most manufacturing. Batch sizes have decreased in many industries, but not to batch-of-one for complex products.

The economics still favour standardisation in most cases.

Supply chain transparency improved modestly

Visibility into suppliers has improved, driven partly by technology and partly by shock events (pandemic, natural disasters) that highlighted supply chain risks.

But end-to-end transparency remains limited. First-tier supplier visibility is achievable; deeper visibility is hard. Many supply chains still have significant blind spots.

Blockchain, once touted as the transparency solution, found limited practical application.

Why the gap between vision and reality?

Several factors explain why Industry 4.0 hasn’t fully materialised:

The installed base problem

Factories don’t start from scratch. They have equipment that’s 20, 30, sometimes 50 years old. Making everything connected and intelligent requires either replacing or retrofitting this equipment—both expensive.

New factories can be built closer to the Industry 4.0 vision. But most of Australian manufacturing is brownfield, not greenfield.

Standards fragmentation

Despite standardisation efforts, industrial connectivity remains fragmented. Different vendors, different protocols, different data formats. Getting systems to talk to each other requires significant integration effort.

This is slowly improving but remains a barrier.

Skills and culture

Technology can be bought. Skills and culture must be developed. Many manufacturing organisations lack the digital skills to implement and operate Industry 4.0 technologies.

Change management is hard. People resist new ways of working. Organisations that underinvest in the human side of digital transformation struggle to capture technology value.

ROI realism

Early Industry 4.0 business cases often oversold benefits and undersold costs. When projects didn’t deliver promised returns, enthusiasm waned.

More realistic expectations have emerged. Projects are more focused, ROI calculations more grounded. This means fewer grand transformation programs and more incremental improvement.

Economic conditions

Manufacturing investment competes for capital. In uncertain economic conditions, large technology investments are harder to justify. Many manufacturers prioritise operational efficiency over transformation.

What’s worked in Australia

Australian manufacturers have made progress, even if not at the pace the vision suggested.

Operational visibility: Many have dashboards providing real-time production information that didn’t exist five years ago.

Focused AI applications: Predictive maintenance, quality, scheduling—practical applications that deliver measurable value.

Digital skills development: A generation of manufacturing professionals now expects to work with data and digital tools.

Remote capabilities: The pandemic forced rapid development of remote monitoring and management capabilities.

Automation of specific tasks: Not full automation, but automation of specific operations where the business case is clear.

What makes sense now?

Given where we are, what should Australian manufacturers prioritise?

Finish the foundation

Many organisations started Industry 4.0 initiatives but didn’t complete foundational work. Connectivity infrastructure, data management, basic analytics—finish these before chasing advanced capabilities.

Prioritise specific applications

Rather than “digital transformation” as a goal, identify specific problems where technology can help. Predictive maintenance for critical equipment. Quality improvement for problematic processes. Scheduling optimisation for complex operations.

Build incrementally

Large transformation programs have poor track records. Incremental improvement—small projects that deliver value, learn, expand—is more reliable.

Invest in skills

Technology without skills wastes money. Training existing staff, hiring digital-savvy workers, and building partnerships with skilled providers all matter more than technology choices.

Stay patient

Industry 4.0 is a journey, not a destination. The factories of 2030 will look different from today. But getting there requires steady progress, not revolutionary leaps.

A realistic assessment

Industry 4.0 hasn’t delivered on its most ambitious promises. Autonomous factories, perfect digital twins, mass customisation—these remain aspirations, not reality.

But significant progress has occurred. Factories are more connected, more data-driven, more capable than a decade ago. AI is solving real problems. Digital skills are spreading.

The vision was probably always 20-30 years ahead of achievable reality—that’s how visions work. What matters is making progress, one practical step at a time.

Australian manufacturing won’t become fully Industry 4.0 overnight. But it can become more competitive, more efficient, and more capable by thoughtfully adopting what’s practical today while preparing for what’s coming tomorrow.