Predictive Maintenance for Tunnel Boring Machines – The world’s population is projected to reach up to 9.7 billion in 2050 and 11 billion by 2100. As the world population continues to grow in size, cities have become more crowded and land resources have also become scarcer. Therefore, it has become a need to go underground for further development and expansion. This need has also given rise to modern tunnelling technologies that are simpler, easier and more efficient than ever before.
Egypt’s city planners have acknowledged that there are crowding and congestion problems as the population grows. Singapore as a small country also faces problems such as limited land, thus making it important for us to tap into the underground space for future development.
“To solve the problem of soul-destroying traffic, roads must go 3D, which means either flying cars or tunnels. Unlike flying cars, tunnels are weatherproof, out of sight and won’t fall on your head. A large network of tunnels many levels deep would help alleviate congestion in any city, no matter how large it grew (just keep adding levels).”
As digging underground is the future, the tunnel boring machine (TBM) market is anticipated to grow at a CAGR of 4.2% by 2022. Technological advancements will grow alongside it as well. Artificial Intelligence (AI) and Machine Learning applications are some of the significant transformations that can benefit the tunneling industry.
In this article, let us look at how AI, machine learning and predictive analytics can come together to help fuel tunneling projects.
What is Predictive Maintenance?
It is a maintenance strategy that helps to monitor machine performance and predict machine failures. Allowing the operations team to optimize maintenance activities. Predictive maintenance uses AI and Machine Learning to identify the trends and patterns of machine performance. Bringing the data together with predictive analytics, the system can accurately predict the likelihood of machine breakdowns.
The predictive maintenance strategy aims to reduce unplanned downtime by effectively predicting potential breakdowns. As such, allowing the operations team to know exactly when to schedule maintenance works. In simpler words, they can fix the right thing at the right time. Furthermore, the system will also provide comprehensive knowledge of the equipment and its capabilities.
Most businesses are currently using a preventive maintenance strategy to prevent machine failures, where inspection and maintenance are carried out according to a planned schedule, regardless of whether maintenance is really needed. Moreover, technicians are required to head down to the machine to inspect the machine parts and operation has to be paused for maintenance works.
On the other hand, predictive maintenance allows remote monitoring of your machine conditions and performance even during normal operation. As such, you only need to carry out maintenance work when your machine needs it.
Most importantly, predictive maintenance is a non-intrusive maintenance strategy. This is because only condition-monitoring sensors are placed close to the equipment, And it will not meddle with the machine parts.
Why adopt predictive maintenance for Tunnel Boring Machines?
TBM is such a huge machine and breaking down all of a sudden can cause many problems, such as costly repairs and delays. Therefore, by using predictive maintenance on TBM, the system can predict when the machine may experience trouble and prompt for maintenance before it fails.
Bertha, the world’s largest TBM in Seattle was brought to a halt for 2 years after working for only 3 days. The machine failed abruptly, causing the $3.2 billion project to experience a 2 year delay and faced a $57.2 million fine. It was later revealed that the TBM’s bearings and seals were damaged from overheating. Such incidents could have been avoided if the poor machine conditions were identified earlier. This is something that predictive maintenance can achieve.
With that said, companies simply cannot afford to allow such things to happen as the impact after can be extremely damaging to a company’s image and financials.
The value of predictive maintenance
1. Better management of cutter tool
To excavate the ground, the TBM pushes the rotating cutting wheel against the tunnel face. This cutting wheel is fitted with different cutter tools e.g. buckets, scrapers, discs as per the surface conditions. Therefore, the operators must determine the surface types. To do this, we can use computer vision and predictive analytics to acquire information about the surrounding rock. This helps you to ‘look-ahead’ into the material that the TBM is cutting into. And the proper cutting tool can be used.
Additionally, the cutter is a wear-intensive tool as it comes into direct contact with the tunnel face. As a result, it has to be inspected and replaced regularly. The operation has to put on hold for the inspection and replacement works. And it also takes a team of workers a couple of hours to do so.
Currently, most replacement works are performed based on a scheduled basis or when the tool fails. To improve project efficiency, predictive maintenance should be used instead. This is because the system can monitor the cutter’s performance and condition. Hence, there is no need for regular inspections and operation can continue as per normal. Not only that, but it will also provide an optimized replacement schedule where the ordering of parts and manpower plans can be done in advance.
2. Improve efficiency
On top of cutter tool management, it is also important for you to optimize the performance of your machine. A TBM ‘Godavari’ in India was damaged as the cutter heads were broken after running into hard rock. As a result, the project was delayed, and costs were stacking up as it took months to replace the cutter heads. This is an undeniable fact that unplanned downtime can cause a heavy strain on the TBM productivity and the overall tunnel construction progress.
Moreover, the duration of the tunneling projects may exceed some of the machine parts’ useful life and new replacements are required. Hence, you should leverage predictive maintenance to monitor your TBM health. At the same time, you can also understand the machine’s Remaining Useful Life (RUL) and determine the replacement work intervals before any failure.
In all, predictive maintenance can improve operational uptime by minimizing unplanned downtime and delays, as well as to keep the machinery performance at its peak.
3. Better cost control
Other than disruptions to the operation, monetary losses can be very high due to breakdowns too. Unplanned downtime costs businesses an average of $260,000 per hour and each downtime lasts an average of four hours. This equates to over $1.04 million for each unplanned downtime. As mentioned in the Bertha incident, the company had to pay a $57.2 million fine for the two-year delay. This is something that all TBM owners will want to avoid.
Hence, it is crucial to prevent unplanned downtime and avoid unnecessary delays. This is especially important for tunneling works that can cost millions in the event of disruptions.
4. Minimize risks at work
It is of higher risk to work underground. Therefore, it is necessary to create a safe working environment for the TBM operators. With predictive maintenance, you can avoid unplanned machine failures that may bring potential danger to the working crews. Furthermore, it will also help to eliminate project delays.
Sometimes, the TBM operator may also believe his judgment that nothing is wrong with the machine even if there is something unusual. Referring back to Bertha’s incident, the operator assumed the machine can go higher speed although there were some signals of poor performance. It eventually led to the failure of Bertha.
Hence, by using predictive maintenance, it will minimize human errors when it comes to determining the machinery condition. It can also free up the operators from machine health monitoring and they can focus on other crucial parts that may require manual intervention.
It is time to move into predictive maintenance
Predictive maintenance delivers extensive benefits in terms of better management of cutter tools, optimized maintenance schedule, cost efficiency and lower project risks. Technology is transforming the tunnelling industry for the better, and predictive maintenance on TBM could be a real game-changer for the industry.
Here at groundup.ai, we have extensive expertise in the field of AI and predictive analytics. These technologies are highly capable of supporting construction companies with the implementation of predictive maintenance for TBM. Have a chat with us to find out more details.
