What 24 Lifts in Hong Kong Reveal About Why Buildings Need an Operational Intelligence Layer

Modern residential buildings depend on complex mechanical systems to function reliably, but those systems rarely fail in isolation or without warning signals. Instead, they degrade gradually, creating operational strain long before visible breakdowns occur.

A detailed case study of 24 lifts across eight high-rise residential buildings in Hong Kong exposes a deeper issue in building operations:

residential assets are not lacking data. They are lacking operational visibility.

This gap is exactly where a new generation of building intelligence is emerging, not as another layer of software, but as an operational intelligence layer that helps people run buildings better every day.

This is the shift represented by Twinview.

The hidden reality: buildings operate reactively, not intelligently

The study of 24 lifts combined maintenance records, facility management interviews, and on-site observations. What it revealed was not just technical degradation, but a structural operating model problem.

A peer-reviewed study on lift maintenance in high-rise residential estates (Building and Environment, ScienceDirect) found that operational performance is strongly shaped by usage patterns, system configuration, and maintenance response cycles.

Key findings included:

• Maintenance demand spikes during peak traffic periods (especially non-weekdays)
• A significant number of “faults” were false or non-critical calls
• Older lifts generated increasing corrective maintenance workload
• Service disruptions were frequent and operationally disruptive to residents

But the most important insight was not the faults themselves, it was the lack of system-wide awareness of why they were happening.

The core issue: operational signals are fragmented and delayed

Even though building operators had access to maintenance logs and service records, the information was:

• Retrospective (what already failed)
• Fragmented (split across systems and contractors)
• Non-contextual (not linked to usage or demand patterns)

This created a consistent operational pattern:

Buildings only become “visible” at the point of failure.

That means facilities teams are consistently operating without a live understanding of system health or emerging risk.

What gets measured today is not what matters in real time

The study benchmarks traditional performance indicators:

• Response time
• Repair time
• Downtime

These metrics are useful for reporting, but fundamentally they are lagging indicators of operational health.

They describe performance after disruption, not conditions before it.

As a result:

• Teams react to breakdowns rather than anticipating them
• Root causes are harder to identify
• Maintenance becomes repetitive rather than preventive

The deeper structural problem: buildings are not connected systems

A key insight from the research is that lift performance is shaped by interacting factors:

• Usage patterns (peak vs off-peak load)
• Configuration design (floor zoning strategies)
• System ageing and component degradation
• Operational environment and demand variability

Yet these factors are rarely analysed together in real time.

Instead, they exist as disconnected data points across:

• maintenance logs
• contractor reports
• building management systems
• resident complaints

This fragmentation creates a fundamental blind spot in how buildings are operated.

From maintenance systems to operational intelligence

The study itself points toward a future direction:

“Automatic data collection and tracking in the industry 4.0 era would enhance performance evaluation.”

But the implication goes further than automation.

The real shift is from data collection to decision intelligence:

• Connecting systems that currently operate independently
• Surfacing patterns that are otherwise invisible
• Translating raw building data into actionable operational insight

This is where the role of Twinview becomes distinct.

Twinview: from digital twin concept to operational intelligence layer

Historically, “digital twin” approaches have focused on modelling buildings as virtual representations.

But in practice, operators are not trying to manage models, they are trying to:

• reduce downtime
• improve resident experience
• control operational costs
• ensure safety and compliance

That is why Twinview is positioned differently:

not as another system to manage buildings, but as the intelligence layer that helps run them better every day.

Instead of focusing on simulation or abstraction, the emphasis shifts to:

• connecting existing building systems
• surfacing trusted, real-time operational information
• enabling confident, faster decisions
• turning complex building data into measurable performance outcomes

What this changes for residential operations

In this model, lift systems are no longer treated as isolated maintenance assets. They become part of a connected operational environment where:

• usage patterns inform maintenance planning
• faults are contextualised in system behaviour
• performance trends are visible before failure occurs
• operational decisions are based on live intelligence, not historical reporting

Final thoughts

The Hong Kong lift study makes one thing clear:

Residential buildings are already highly instrumented, but still fundamentally unobserved in real time.

The gap is not technology availability. It is operational visibility across fragmented systems.

And closing that gap is what defines the next generation of building performance.

Not digital twins as models, but intelligence layers that help buildings actually run better.

References

Li, H., Yang, J., et al. (2023). Assessing lift maintenance performance in high-rise residential buildings: A case study of 24 lifts in Hong Kong. Building and Environment. Available at: https://www.sciencedirect.com/science/article/abs/pii/S2352710223003819 (Accessed: 18 June 2026).