In February, GE in association with Mabey hosted a round table debate about how instrumentation and monitoring could lead to better geotechnical design.
Managing the scale and timing of the failure of geotechnical assets is the ultimate goal forthe industry. GE asked a group of contractors, consultants and clients if a better understanding of the issues through instrumentation and monitoring could lead to better design.
“It’s about validation,” said AKTII technical director Dimitrios Selemetas. “We need good quality monitoring data because we want to be able to validate sophisticated numerical models.
“When we do a sophisticated analysis, and you are spending money analysing and modelling everything, you have to think how to validate the model.
“The client needs to accept the value of monitoring at an early stage. It is exactly the same as with ground investigation.”
Swanton Consulting director Rob Millard agreed, but thought that it was sometimes difficult for people to accept the data to justify the design.
“The client will often only accept the design if the advisors they employ will accept it,” explained Millard. “It is the same with data. You are trying to justify or substantiate your design and you are trying to prove it with observational techniques and methods, but this approach seems not to be appropriate to some engineers, so we end up not using this method.
The industry learns best from its mistakes, but these aren’t talked about
“It’s not too late, but it is having faith in your design substantiated by onsite monitoring with a back up plan if it doesn’t work. But people still do not accept it.”
Selemetas said he believed that innovation was being stifled by the current approach: “Innovation is seen as a risk rather than an opportunity and I think, as an industry, if we want to build a complex basement we cannot just use previous guidance based on case studies, we need to use a more performance-based design and validate that. We need more high-quality monitoring data.”
Transport for London head of earth structure and geotechnics Nader Saffari explained that for an intelligent client, innovation is a priority but for monitoring to be of any value it should to be owned, processed and managed appropriately, which has cost implications for long term monitoring.
Fellow client Network Rail route asset manager for geotechnical engineering Derek Butcher agreed: “The data should go from the instrumentation contractor, back to the consultant to optimise their characteristic values to save the client money, but this doesn’t happen.”
Using the construction of Crossrail as an example, Arup fellow and former director Duncan Nicholson explained how the observational method had been extended for use on subsequent projects.
“Tottenham Court Road ticket hall excavation started as a characteristic or conservative design. During excavation the monitoring data was back analysed and showed that the bottom strut leve could be omitted.
“We have subsequently back analysed Crossrail monitoring data and come up with the most probable parameters and reapplied them to the original design.
“This enabled the wall thickness to be reduced and increased cost saving could be made.
“You have to keep learning from recent experience and plough it into the next project.”
WSP associate for ground engineering and tunnelling Guy Swains added: “The more monitoring data you get from active projects, the more predictive you can be. If that data gets fed back to the industry then it can be used in design.
“The industry learns best from its mistakes, but these aren’t talked about.”
And does the cost of monitoring hinder its use? And who should, ultimately, pay for it?
Millard continued: “The contractor will say that the client has to pay; a consultant will say that a client has to pay, but the client will ask where the value is for them. No one will close the loop.
“Monitoring is cheaper than it has ever been before and you can create mountains of data, but what good is it until we accept that we can use that data to do something.”
Saffari added: “A lot of the time I don’t think people realise the benefit of the monitoring they do, as it is very often seen as a routine exercise – an insurance policy – to fulfill a contractual requirement.
“However, the real value and benefit of monitoring is realised when the results provide warnings against excessive deformation or onset of instability. So, because in many cases nothing happens, it doesn’t mean that monitoring is of no value.”
If the industry is finding it hard to use active monitoring now, how will it adapt to use monitoring for the whole life of projects to offer predictable design life guarantees, such HS2’s plans to make its infrastructure “smart”?
“Badly,” said Swains, who said he thought the industry would struggle to deal with forward thinking monitoring.
“We can’t solve the problems we have now, let alone what HS2 is trying to achieve.”
For the optimist the glass is half full, for the pessimist the glass is half empty but for the designer the glass is twice as big as it needs to be.
GE editor Claire Smith asked what data will the industry want in the future, based on the 120 year service life that HS2 is designing to.
“It’ll be smart asset that will be able to tell you how it is performing. It’ll make monitoring safer as it will take people off the tracks,” added Butcher.
Swains thought that it would still need a person to be able to interpret the data and set the trigger levels. “This will always be one of its limitations. Someone will need to be there to understand and manage the asset,” he said.
Saffari said: “In the case of a long term monitoring we need to ensure that the instrumentation remains functional or can be maintained or replaced. We need to ensure that the data produced and the processing and management will be compatible with the future advances in technology.”
Butcher said he thought that the inability to sign up for a long-term plan that cannot be committed to, especially around the funding model used by the government for large asset providers, is a limitation.
Tensar chief civil engineer Yuli Doulala-Rigby said that everybody in the industry should have the same objectives if real change was to be achieved.
“Everybody needs to talk the same language, then we will have real improvement. It won’t changeu ntil that happens,” she said. “All stakeholders need to sit down and talk at the beginning of each project to understand and share the risk.”
So, what if the industry can get over those problems and we are collecting and sharing the data? Could the sector use it to its advantage?
“We know we can design things conservatively, and ultimately thec onservativeness will decrease in the industry if we are collaborative,” said Swains.
Selemetas agreed: “It will allow us to design and build structures that at the moment seem to be impossible. Every time we try something innovative it is seen as a risk, rather than an opportunity. We need to continue prove design methods.“
Essentially the problem in the industry is that we look at innovation as a risk. For the optimist the glass is half full, for the pessimist the glass is half empty but for the designer the glass is twice as big as it needs to be.”
Mabey Hire head of structural monitoring Aris Nikolarakos agreed and added that the industry needed to work how it could share data and who could manage it.
“We could improve geotechnical design, but the industry needs to talk about it outside of the sector. We all agree here, but it is getting that message out.”
Around the table
Derek Butcher route asset manager - geotechnical engineering Network Rail
Yuli Doulala-Rigby chief civil engineer Tensar
Jake Kaufman business development manager Mabey Hire
Duncan Nicholson fellow and former director Arup
Aris Nikolarakos head of structural monitoring Mabey Hire
Rob Millard director Swanton Consulting
Nader Saffari head of earth structure and geotechnics TfL
Dimitrios Selemetas technical director AKTII
Claire Smith editor Ground Engineering
Guy Swains associate for ground engineering and tunnelling WSP