How to improve TBM performance with data

Tunnel boring machine performance is often explained in simple terms, but the reality is more complex. As projects seek better predictability and productivity, understanding what truly drives advance rates requires looking beyond ground conditions and examining how time is used across the entire tunnelling system.

Geology is often the first explanation for slow tunnel boring machine (TBM) progress. When advance rates fall short, ground conditions are easy to blame because variable geology, groundwater, and abrasive earth are known risks in tunnelling.

We had a theory that predictable data would show a different picture, which proved correct. Our model replaces assumptions with evidence and turns TBM data into practical lessons for future tunnels. 

We drew on insights from multiple TBMs across major Canadian tunnelling projects and found that some of the biggest constraints lie behind the cutterhead, not ahead of it.

While geology matters, traditional metrics fail to explain its actual impact. We developed a new, data‑driven approach to measure TBM performance that goes beyond availability and utilization. Our model separates ground conditions from logistics and operations, and shows that planning, decision‑making, and site operations often matter more than geology in terms of how quickly a machine advances. 

Accounting for 100% of TBM calendar time

We developed a time usage model (TUM) to track every minute of time, from TBM launch to breakthrough. The model separates productive excavation time from all types of pauses, including operating delays, standby periods, maintenance, breakdowns, and work related to ground conditions. 

This approach allows project teams to see exactly where time goes during a TBM drive and why advance rates change. Owners, contractors, and designers can then compare performance across projects using a common basis rather than fragmented logs or summary metrics. 

That clarity is what makes the model innovative. Tunnelling projects have often struggled to compare performance across different TBMs, sites, and ground conditions. Delays get lumped together, making it hard to understand what really slowed production. But using a TUM to standardize how projects measure and classify time allows meaningful comparison across projects. It helps teams focus on changes that will actually improve performance. 

When projects capture time consistently and completely, patterns emerge quickly. Delays no longer blur together, and teams can see where performance is truly lost. Our study of multiple TBM drives found that standby time, material supply issues, segment delivery, muck removal, and coordination behind the TBM regularly outweigh geological delays.  

These factors also happen to be the most controllable. 

Early decisions and small delays amplify into production losses

Our model shows that early systems choices (often made during design and procurement) impact TBM performance throughout a drive. For example, using continuous tunnel conveyors can be 20% more efficient than using muck cars. 

The TUM distinguishes between operating delay (crew‑driven inefficiencies) and operating standby (logistics‑driven interruptions). This separation allows you to identify whether people, processes, or supply chains limit progress.

Individually, short delays may appear insignificant. Over weeks and months, however, they accumulate and materially reduce production. Teams that identify and quickly address these recurring inefficiencies often recover substantial working time. 

When you collect standardized data, analyze delays by cause, and act on what the data reveal, performance becomes measurable, comparable, and improvable. The TUM provides a practical framework to do exactly that. It helps teams move past assumptions, focus on the real drivers of delay, and make targeted changes that deliver results. 

The Kemano T2 tunnel makes this especially clear. There, data‑driven construction management improved advance rates by roughly 60%, without changing the TBM, the crew, or the geology. That outcome proves this approach is not just analytical. It directly improves productivity and reduces schedule risk. 

For owners, contractors, and designers, the implications are significant. Standardized metrics enable fair performance comparisons, more realistic schedules, better forecasts, and stronger claim assessments. Just as importantly, they support a culture of continuous improvement, where teams learn from data, adapt quickly, and consistently outperform expectations. 

In the end, the lesson is straightforward: better data leads to better tunnels. When project teams treat time with the same rigor they apply to design and construction, TBM performance becomes a competitive advantage. 

Join the conversation

These themes will be explored at the upcoming World Tunnel Congress in Montreal, May 15–21, 2026, where we will share practical insights into data-driven construction management for TBM tunnelling, including real project results. For owners, designers, and contractors, it is an opportunity to engage with emerging best practices that support better decision-making and performance. 

Interested in applying data-driven approaches on your next project? Connect with us to learn more.