Learning from data is key to reducing error through technology

Capturing data through technology has great potential to reduce error but only if it is embedded in the process and is available to users when they need it, Steve Green from Bowmer & Kirkland told the recent GIRI technology webinar.

Steve took a measured view of the potential of technology to reduce error, highlighting the fact that simply collecting data and lessons learned has no benefit unless you do something with it.

Steve reviewed three types of technology, looking not just at their potential for error reduction, but also their shortcomings. He began with BIM, which he noted is effective at preventing designs reaching site with inherent clashes, and encourages collaboration through a federated design space, but nevertheless requires someone to redesign projects to overcome the clashes, and won’t reduce site errors. “Just because the model is right, it doesn’t mean that what we build is right. But BIM can improve productivity by removing initial clashes, and it can help clients visualise the end product, which probably helps overcome late variations.”

He then moved onto field-based applications – digital solutions that replace pens and clipboards – recounting his experience of using Field View. Steve was responsible for introducing Field View into Bouygues in 2014. “Although there was some initial resistance from users, the fact that you could complete forms in a tenth of the time meant the software was rapidly embraced. Instead of doing site admin, staff could be out on site or planning ahead, doing more value-adding tasks.”

Again, however, there are limitations, said Steve. “Field-based solutions can record error and are good at managing rectifications, but they won’t stop design errors. And while they can improve productivity, they are inspection tools and you cannot inspect quality into projects.”

These applications also generate a lot of data, which led into Steve’s final point about knowledge management and databases. He stressed that making use of the data collected is the key to knowledge management. “Digital technologies mean we can quickly record information, but what do we do with it? Lessons learned are of limited benefit unless you make these part of the process.”

As an example, he highlighted the potential of videos and how these were used by Bouygues. “We realised we didn’t know what good looked like. I went out with one of our post-completion managers, who used to be a bricklayer, and we recorded 15-second snippets of how to inspect brickwork – what a clean cavity should look like, or how DPCs should be lapped – and linked these to the inspection sheet. So, if you were going out to inspect brickwork, you could watch the video showing you what you needed to do.”

Shortly after this, Bouygnes’s parent company introduced a similar initiative, which saw staff on site record videos of up to two minutes on how to install or inspect particular elements of work. “However, rather than linking these videos to the inspection sheet they created posters with QR codes that went up on site. So, if you were installing a fire door, you could scan the QR code on a fire door poster and watch a video showing how the installation should be carried out.”

Capturing this digital information – from clashes in BIM to quality errors in field-based applications – has real potential to reduce errors, said Steve, but it is important to learn from them. “I think we can reduce site errors and improve productivity as long as the knowledge is received by the user in the right format at the right time. They do not always know when they need it, so the real challenge is understanding how and when to get that knowledge to the user.”

Read Melanie Dawson's webinar presentation.

Read Abhishek Srivastava's webinar presentation.

Sign up to our newsletter to hear about future events.