Fondue or the Future of Tunnelling?

As I am a big fan of fondue, Swiss cheese, and the Haute-Savoie, I was intrigued by Belinda Smart's article titled "Future of Tunnelling | Key lessons from major tunnelling failures" on newcivilengineer.com. It highlighted a significant problem that resonated with me, even as a tunnel monitoring expert. Despite having the best monitoring systems in place and access to frequent data on material performance, ground conditions and structural parameters, reaching the right risk assessment and conclusion is not always straightforward.

The Challenge of Decision-Making:

One of the challenges we have observed is the complexity of decision-making processes. Often, no single person has the final say, and it is not always clear who is in charge. Decision-makers may not be familiar with the monitoring data until alarms are raised and issues escalate. They must rely on monitoring engineers, structural engineers, and geologists to piece together the full picture. However, when reality clashes with design, it can be difficult for decision-makers to believe a set of numbers or a graph indicating that something is wrong.

Lessons from Past Experiences:

Years ago, we encountered cases where monitoring data clearly showed a trend and allowed us to predict breach of red trigger limits days in advance. Yet, even when triggers were breached and cracks appeared in the tunnel lining, there were still debates among engineers, questioning whether this could be happening. It was not until the day sunlight penetrated the tunnel, making it impossible to ignore, that action was taken. Fortunately, no one was harmed, and safety measures had been implemented. Nevertheless, it was intriguing to witness the delays in decision-making due to various obligations and considerations of risk versus cost. Even with clear data, alarm plans were not always followed, as the red trigger was uncomfortably close to tunnel collapse conditions.

Trusting Sensor Data and Data Analytics:

One aspect that complicates decision-making is the inherent noise and environmental influences present in physical sensor data. Factors like temperature, water pressure, tides and weather can affect the data, creating a certain level of uncertainty. People are not accustomed to the inherent fuzziness of our world, where everything is in constant motion. This uncertainty makes it difficult for decision-makers to fully trust the power of data analytics, which can effectively differentiate between a signal and its noisy background.

Understanding Psychological Pressure:

It is crucial not to underestimate the psychological pressure decision-makers face in these challenging situations. The psychological factors at play can make it difficult for decision-makers to believe the facts presented to them. Moreover, the differing opinions of consultants and contractors, each with their own perspective on risk or gut feelings based on experience, contribute to a chorus of voices where the loudest is not necessarily the best singer.

The Importance of Preparedness and Testing:

To address these challenges, it is helpful to regularly review monitoring data and simulate trigger breaches as part of an exercise, similar to a fire drill. Nowadays, we can even simulate events like water ingress in immersive 3D virtual environments. By conducting drills, decision-makers become familiar with how the data might appear, who should be consulted and when, and what preventive measures can be taken. It is crucial to capture all these aspects in an alarm and mitigation plan, and if such a plan already exists, it should be thoroughly tested.

Conclusion:

In the realm of tunnel monitoring, decision-making can be a complex and psychologically challenging task. By acknowledging the intricacies involved and the factors that influence decision-makers, we can better support them in making informed choices. Regular data review, simulated drills, and comprehensive alarm and mitigation plans contribute to effective risk management and the successful integration of monitoring sensor data into the Swiss Cheese Model. Ultimately, by understanding the human element and leveraging advanced technologies, we can enhance tunnel safety and ensure the success of projects like HS2.

Read more about Murphy Geospatial’s monitoring solutions: https://murphygs.com/monitoring/