Unsupported browser

For a better experience, please update your browser to its latest version.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Data sharing: Please feed the Dingo

fugro assembly of o cell instruments into reinforcing cage

A project to gather pile load testing data and make it freely available, needs the industry to help feed it. Michaila Hancock reports.

Sharing of pile load test results is about to become reality with launch of new database. But it needs industry input to maintain its relevance.

The Databases to Interrogate Geotechnical Observations (Dingo) was born out of one researcher’s desire to take up a challenge he heard set in the 2003 Rankine Lecture. In his lecture that year, speaker Mark Randolph had said that it might not be possible for the industry to estimate the axial capacity in most soil types by plus or minus 30%.

It was this that inspired Paul Vardanega, Dingo’s principal investigator and a civil engineering lecturer at the University of Bristol, to develop the database.

The Dingo database, led by Vardanega and partnered with Arup, Atkins and Fugro, will provide a free database of pile load tests in UK soil deposits, open to everyone to use for their own research.

“While Randolph might be correct, it’s an interesting challenge to take up as a researcher,” says Vardanega, who believes that the database would be useful for designers and researchers to access datasets and calibrate their own models for pile behaviour.

“There are many uncertainties in pile design such as soil properties, spatial variation, properties in the concrete, types of pile construction and the approximations inherent to how we analyse the structures while we are designing them.”

We are often told there is not enough data, but what do we actually already know

In Vardanega’s opinion the geotechnical community is accustomed to using databases, but being able to collate data together on this scale hasn’t been done before.

“Geotechnics is arguably, for the first time in a long time, starting to grapple with the idea that we are no longer necessarily data poor. We might be data rich and not know it?

“We are often told there is not enough data, but what do we actually already know,” suggests Vardanega.

In late 2016, a grant proposal was submitted to Engineering and Physical Sciences Research Council and £96,500 was awarded to Vardanega for the 18-month project. This allows for a post-doctoral researcher to work on the project with two PhD students from the University of Bristol.

Six months down the line and, with support from industry partners, others are taking notice and willing to hand over the data for the project.

“We’ve recently had very good support from Highways England,” explains Vardanega. “We are looking for historical data sets that we might not know exist.

“This is very much a data mining exercise. They know where testing was done. But you wouldn’t know that data exists if you didn’t work on the initial project. It’s detective work.”

The project aims to obtain data from every scheme in the UK that has carried out a full-scale pile load testing, whether this is single pile load test, or many on a site. But the data needs to useable. As the project progresses, decisions will be made on what data will be stored in the database.

“When we have the final set of everything, we will then make the decision on what is the most useful set of data. And it might be we put everything up and note what parts are unavailable. For example, in the future if we have a pile test missing the ground investigation data, someone might still be able to make use of it and add to it at a later date,” adds Vardanega.

box 3 pile testing

box 3 pile testing

The project aims to obtain data from every scheme in the UK that has carried out a full-scale pile load testing

But the project hasn’t been all plain sailing. The nature of the industry has made it difficult to identify that data exists and then to gather it. This will require more research.

“This is not blaming anybody,” stresses Vardanega, “but the eternal problem is that the industry is very fragmented. There are many agencies who have done this kind of work in the past, so the data, by definition, is fragmented. It requires more digging around to actually locate where the data is.

“We knew that would be main problem and we’re overcoming by talking to as many people in the industry as possible. The project is about finding the data and if we are able to just deliver ‘here is the data, here is what we have found and it’s all in the one place’ that means that someone else doesn’t have to the data collection bit.

“It is an advance to just put it all in one place. Even if we haven’t found everything, we don’t know the totality of the it, we don’t know how many piles tests have been conducted in the UK. We just don’t know that number.”

The industry response been overwhelmingly positive so far reports Vardanega, and although there has been scratching of heads to identity and locate the data, the response has been productive.

“The size of the dataset for a project doesn’t matter,” says Vardanega. “For a person who did one or two tests on a small project 20 years ago, that might not seem like a big deal, but for us it is a very valuable piece of information when aggregated with other pieces of information.”

“We want as much data as we can possibly use,” adds Vardanega.

Donating data to Dingo

Anybody who in interested in donating data, no matter how small, or becoming an industry partner should contact Vardanega directly, who will be able to talk you through the process and how some of the information can be anonymised if required.

“We need as many people to get in touch as possible, the more data we have, the more useful it will be” says Vardanega.

Paul Vardanega can be contacted on p.j.vardanega@bristol.ac.uk and 0117 331 5710

 

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.