Wouldn’t it be nice, if we had a fully digitised construction industry? If digital engineers could synchronise their systems to pull all the information they need from manufacturers straight into their modelling software? If you could see at a click of a button whether that product will work with the system, if it made it more efficient or if it will have unforeseen consequences? If all the performance and test data was expressed in the same way, users wouldn’t have to spend time trying to convert information or chase manufacturers for further details. The whole process would be easier, safer and more efficient.
It all sounds great, so the question becomes: why hasn’t it been done yet?
It comes down to how the industry works. The phrase ‘construction industry’ is not quite accurate. It would be better to call it ‘construction industries’ – as there are many. A building is a holistic system, but it remains so complicated that no one could be expected to be an expert in all its systems any more than we would expect a doctor to specialise in every part of a body. We have many professions, many approaches and many perspectives. The very nature of construction’s multiple industries brings many challenges, and when you add competition into the mix that compounds those difficulties. Dame Judith Hackitt called our industry ‘siloed’. She isn’t wrong.
Here’s the biggest test: we might all be talking about topics within construction, but the way we describe it can be fundamentally different. We don’t use English in the same way, it is often conflicting in either nuance or specificity. Even within our standards and regulations there are contradictions. That makes agreeing a digital language that computers can process and exchange a pretty large challenge. Thankfully, if you need to fix a digital problem you can use a digital solution: construction data dictionaries. Using a data dictionary, we can put language into a software platform making it easier to search, synchronise and agree, attach codes to that language, group and organise it so that computers can exchange it.
It’s an excellent premise and there’s been great progress. There are many construction data dictionaries in the UK. The catch is that they have been populated in many different ways. The information is structured differently in each of these cases, and the method of agreeing what goes in there also varies. Which means not everyone can trust it to deliver what is necessary for everyone, and we still are yet to have a ‘single source of the truth’.
To make the information in these data dictionaries trustworthy, we need to agree how it is all done. We have a good model for this in the UK. If you want to write a British Standard (BS), you have to follow the rules set out in BS 0:2016. It’s an industry agreed process, and as such it is respected and followed by the majority. The same needs to be true of our data dictionaries.
At the moment, there is no agreed process of how to agree what’s in a data dictionary. Thankfully, there are people working on it. CEN TC 442 are very close to setting the international principles, but it is down to each country to consider actionable steps to achieve them. That is what PAS 14191:2020 Built environment – Management and operation of interconnected construction data dictionaries – Specification intends to do. It is the rules for how to agree an industry digital language and group it up into structured information in the UK.
Once we all agree the best process, then dictionaries that sign up to this process can connect. In this method, if you add a change in one dictionary all the others will update automatically: safe in the knowledge that we are all following the same processes and fail safes. There will be no duplication of work, and we can expand a language that is a true representation of all the industries and their nuances, accessible and interoperable.
As lead sponsor of the PAS, we asked Carl Collins from the Chartered Institution of Building Services Engineers to write it for us. Well respected in the industry, Carl has pretty much been on every British and international standard there ever was to do with digital construction information. Along with a steering group representing all levels of construction, Carl has set out understandable rules about how to ensure data dictionaries are inclusive, rigorous, practical and secure. It talks about how to manage the consensus processes of how to agree properties and group them together into templates. It makes sure that nothing can be deleted or changed unless everything is agreed. And it talks about how this information is synchronized across all the dictionaries.
With this PAS, the UK will be leading the way in connecting our data dictionaries, and begin the process of agreeing a fully implementable, digitised industry agreed language. We will be able to structure our product information, the very building blocks (pun intended) of all construction. We will be taking the crucial step in making a truly non-siloed industry with efficient, effective systems that can automatically check human error, give better data about performance, sustainability and safety. We want every part of industry to give their feedback on PAS 14191:2020 to ensure we get it right, so we are asking for everyone to take the time to read and comment on it.
Do we think this PAS is important? Just a bit.
PAS 14191:2020 Built environment – Management and operation of interconnected construction data dictionaries – Specification is available for comment here by 11 February 2020.
Hanna Clarke is the CPA project lead on the LEXiCON project and PAS 14191:2020 Built environment – Operation of interconnected construction data dictionaries. She also sits on the Industry Response Group, the committee B/555 – Construction design, modelling and data exchange and is on the Steering Group member for BS 8464 Digital Management of Fire Safety Information.
By Hanna Clarke at 31 Jan 2020, 12:03 PM