BIM4Housing Advisory Working Group Meeting-20220713_110352
RICHARD beings by saying that the first BIM4housing steering group meeting took place last week. Participants include the leaders of all the working groups and the workstreams, meeting approximately every three months, and working on deciding how strategically BIM4housing moves forward. The first meeting looked at what’s been done so far, the methodologies being used (which are evolving all the time). It was agreed unanimously that what we are doing is broadly speaking right, consequently we’ll carry on like that with an ethos to improve as we go along.
GEORGE believes that the knowledge is growing incrementally by working together, significant things are being identified that, when tested in discussions, can be recalibrated or added to. A challenge is to keep everybody involved aware of those new ideas. The meetings are documented and recorded, but nobody really has the time to review those things. Therefore, we’re trying to instil them into some examples that we can then work with going forward. Considering that everyone in this group works in an advisory capacity (to other stakeholder groups) it would be useful if this advisory working group becomes knowledgable across the board about what is needed.
GEORGE aims in this meeting to instil down what’s been learnt up until now and to them agree how to go forward. DEBBIE suggests it would be a good idea to engage with ACE (Association of Consulting Engineers) as their members would be people worth targeting with this information. She will send ACE contact information to Richard.
GEORGE says that one of the challenges is that there are so many different institutions/membership organisations, all of which have their own perspective. That’s both helpful and a barrier because they look at things very much from their members perspective rather than holistically. NICHOLAS NISBET says many such organisations have pretty much declined to engage with the digital agenda in a shared way. DEBBIE says that ACE did a consultancy called ‘Consultancy 4.0’ which is very much about how they need to work with the clients and change the clients mindset to understand this.
GEORGE, beginning his presentation by sharing the Buildings Safeties Act 2022 on screen, states that we are now in a regulated industry. A further 30 or so regulations will be coming through on the back of the Act which is challenging. He would like everyone in the group to think about ‘Is my project in scope?’ - do I have to comply with the Gateways? He gives the Context of a project which has secured planning approval, started demolition work (June 2022) and is still at the beginning of work stage 4, Bsria feasible generic design. Consequently, the design may not complete until Q4 2022 or Q1 2023. handover will be 2024 or 2025.
The exam question is will it be exempt from gateway 2 and will it be exempt from gateway 3? He asks group members’ opinions. JIM CREAK and IHSAN HOQUE say yes. The question is clarified: will they (or will they not) have to do the gateway 2 review? NICHOLAS NISBET states that 18 months after royal assent Gateways 2 and 3 come into force - that’s not yet as royal assent only happened in April, it will be October 2023.
The reason GEORGE is asking the question is that he’s asked subject matter experts about this matter and he can’t get a common answer. Despite this being a regulation, somehow it is not clear. JIM CREAK is not stating they have to do it as a fact, but in an advisory capacity he would suggest that they carry out the review (despite it not currently being the law to do so). NICHOLAS NISBET absolutely thinks that they would have to go through the regulator in 2024/25. There’s no suggestion that you can claim that you first thought of your scheme a year ago and therefore you don’t have to do any of the transferring to occupancies Gateway 3. By October 2023 you’ll have to have your golden thread in order..
GEORGE notes that big law firms disagree about this matter. he believes this advisory group has something to contribute here because it’s not clear. The Act talks about ‘prescribed information’ - there is no indication that we’ll get anything more detailed than that. George and Nick’s work on the Golden Thread Initiative has been looking at what information is actually needed. Out of that they’ve produced the asset and survey information schedule of the documents/machine-readable data that people that are producing safety case say need to be in place.
NICHOLAS thinks the ‘prescribed information’ is a list of 18 fairly random bits of information, the preliminary list has been published. It asks a lot of information about external cladding systems and not a lot of information about anything else. It’s hardly a digital record of any kind. The digital element of the expectations seems to be being watered down.
It’s necessary to go back to the Hackett Report and the published definition of the golden thread and to hold the Department to account as they pass the secondary documents because at the moment it’s the Health & safety executive taking it’s conventional passive-aggressive position of following around and saying ‘impress us’, rather than saying anything other than there has to be a safety case.
If they get away with that (starting with their refusal to define digital) the industry will have to make its own business case for anything digital.
NICHOLAS thinks that one of the things this group might want to do is to connect up with the Golden Thread initiative to independently start to think about the business case for doing it anyway because it will have to br ready so that at least the industry can make its own choices. GEORGE is taking this approach and thinks ‘we can’t look for legislation to force us to do things’. Digital will mean digital because if it’s going to be analog then it’s going to be possible for anybody to deal with all of the documentation. We have to collectively come up with a code of practice/playbook to say this is how we are actually going to address it.
DEBBIE notes that other sectors golden threads are digital, so why should construction be any different? GEORGE says that what’s resonated with people so far is putting together scenarios including the information that needs to be digital. One of the things that came out from the GTI was that the design/specification group determined that we need to switch to a prescriptive specification at the end of workstage 4/gateway 2. Therefore the products need to be identified at that stage. George worked with Nick on a consolidation of the different working groups and they were consistent in saying that products and systems should be identified at Gateway 2, Some of the work that George will be highlighting that needs to be done is a combination of a data workstream and also fire safety and sustainability. The HSE said to concentrate on looking at things from a risk perspective. At the moment, the publications that we have created are just documents - what we’ve got to do is to turn them into proper information. Therefore, as an advisory group we need to put together a playbook with what information is required, how much can be documents, what must be machine-readable, which types of assets are relevant to, which systems, how to organise models.
Additionally, what information is it reasonable to expect from manufacturers, how is their IP protected, what information should we expect from installers, how do we assess and verify competence and how is the safety case demonstrated. if we can draw some of these things together we’ll have something we can use to advise the other working groups and also our customers in the market place.
DEBBIE thinks that in terms of defining machine-readable it’s necessary to be aware that the BSI has a program called Smart Standards where they are defining machine readable. NICHOLAS thinks this is a good agenda and what needs to be added to this is a business case best practice pre-amble to say what we’re recommending will meet the regulatory requirements that will be recognisable as quality by business leads in an organisation and insurers (irrespective of whether the HSC is going to sign up to it).
He would like to change the order of what George read out from the Golden Thread Playbook powerpoint and the last layer of the cake is evidence and that’s where documents come in. maintaining lists of documents and contacts is where the virtual world anchors itself. Nick’s Risk, Information and Asset Management paper (co-published with Chris Lees) goes quite a long way to answer quite a lot of these questions: https://www.ukbimalliance.org/wp-content/uploads/2022/02/Information Risk and Built Asset Management 1.pdf
NICHOLAS would like to move the question about the safety case up to near the top. And then supplementary questions like the role of individual participants, manufacturers, installers, building managers can be described. As for IP he thinks it’s a load of nonsense. There is no difference to the IP in a BIM model to the IP in a drawing.
GEORGE interrupts to explain the context. There’s a lot of uncertainty as to what is a tested and approved product e.g. the fire door example used at DCW. All asset managers want to know about a fire door is that it’s certified. He question this because, from an asset management perspective, what happens in the future when a hinge or closer needs replacing, then we’d need to know what the product was. We also need to know the various different test criteria. Some Door manufacturers are not willing to share this because its commercially sensitive information and they wouldn’t want their competitors to know how their product was tested. PRP are even, when they are asking for detailed product information, being asked to sign NDAs. Safety critical information should not be secret.
IHSAN has said he’s had to sign NDAs in his work at Guinness. It’s the manufacturers who have primary test evidence who do not want to give the information over easily as (he believes) it’s not a regulated industry as you can go and get a global assessment (desktop study by IFCC etc.) done. Primary tested evidence means they’ve paid a lot of money to burn/test their doors, hence they don’t want to give out the information as someone else can then use it. It’s commercially sensitive because of the large amount of money they spend on it. GEORGE says it’s about field of application, a critical context in which the particular item was tested.
GEORGE says another challenge for the industry is that there is probably too much data, standards and guidance. Building regs refers to 800 standards. What we need to do is make it all machine readable. Fire rating information is presented in different ways in the data sheets of different door manufacturers. Manually checking this information is not practical. There is a perception that all of this data is in the BIM model (which is incorrect). There are many different data dictionaries (IFC, Uniclass, Sibsey BImhawk, BS8644 etc) in which terminology varies - consequently there are many different ways of describing the same thing.
A solution to this is to use the means of capturing the information that we would normally want for asset management and then just add to that the safety information. If we do that we’ll then have the opportunity to just add the embodied carbon. One way of doing that is by using the Templater. We have an instance of the Templater that we can connect into all these different data dictionaries and therefore make it so that that information is consumable in many different ways. For a door you can have different classifications according to the context that it’s being used.
Also, you can have groups of properties (e.g. IFCs) that are held against that item and you can drill into that for further info. You can define when you want that information and also who needs it. Particular information can be accessed at the specific workstage the information needed to perform that function. The problem is it becomes locked up in BIM objects when actually its fantastic data that can be used in lots of different meaningful ways. EPD information was populated into Templater as an exercise which means you can look at the groups of properties e.g. climate change and we can hold the information against that particular asset/type of product and all of its other information as well. It can then be used to generate a data template/set that can then be consumed.
NICHOLAS says it’s important to distinguish between the dictionary initiatives to define the terms that we need to use, it’s a separate question about the actual values for particular products and how they are delivered. Anything that can handle synonyms is going to be vital for the industry for the foreseeable future to have friction-free information logistics. Only then can we think about the data libraries that we want.
JIM CREAK Regulation 38 asks who’s job is it to dissect all of the information from the drawings and only make that which is being used in the project pertinent to the building manager because the last thing he would want as a building manager is to find the door set that he has in amongst 150 other door sets with different attributes and to have to go back to the actual door to understand which one his is.
GEORGE replies that’s a critical part of the BIM process. You can define at the beginning of the project a unique identifier for every asset type. You can determine that from the beginning (before the design work has been done) and it becomes that data key which flows through the whole process. JIM, when talking about BIM, has always been shown huge amounts of information but the Regulation 38 ends up by being an analog amalgam of manufacturers information.
NICHOLAS replies that the direction of travel for the industry has been to move from a push model which manufacturers are wedded to (they choose what information they think we can be trusted with) to a pull information where client and operators actually give expression to what information they need and then the industry can respond. That’s the lean and efficient way of doing it and its the way that those with the needs, requirements and knowledge as to what information is critical can influence the industry. Jim, with his team, needs to work out what it is that he needs and the industry is increasingly equipped to deliver it.
GEORGE, continuing his presentation, says we’ve got lots of different Stakeholders that need different information for different purposes. The evolving methodology is saying we’ve got lot of data. The process we’re operating is to add structure to this data which then allows you to apply experitse to it and then used that in a particular context to then capture that in reusable information. Then, we can use reusable libraries that can be used for particular functions.
The DCW context was a building in scope, we used that to apply a particular scenario that a fire breaks out in one of the kitchens. Then looking at how does the building enable the building to escape. We examined all the different elements of compartmentation, smoke control and detection. We want to try and make sure we have the right information about these products to protect against that particular scenario. We have a particular challenge (spread of smoke) solved by compartmentation. Of all the 2 or 300 bits of information on a fire door what are the six items relevant to that particular risk and have we got them?
There is a group of Tier 1 contractors who engaged with BIM4housing last Autumn. The major driver they got to collaborate was the liability they have through the Defective Premises Act. Part of that is understanding how they are going to get everything done by Gateway 2 so they don’t get delays. What they want to do is move Contractor design portion from workstage 5 to workstage 4 which is a massive change.
What we could then do is put in place a methodology where products could be tested against the requirements to make sure you’re only using products from a supplier that have got the data we need. When the designer is picking from a curated set of products, when they issue that the the supply chain, if there is a need to change, that technical submittal process needs to be really robust. When that is then approved we need to make sure we are buying what was actually agreed . Again, this requires digital information.
JIM CREAK responds by saying he is still uncertain how the systems that are done by specialist contractors and the components within that system are encompassed, the potential change of contractor over the lifetime of a building. Some life critical systems are really reliant on competent engineers and contractors to make that system effective and efficient in the lifetime expectancy.
JAREK says that the idea that the commercially sensitive information needs to be digital and then it can be shared in a secure way is a good idea but practically very challenging. Either the technologies are not there or there is a problem with integration between platforms, also people using the information may share it. He liked Nick’s comment about pull instead of push which simplifies a lot of things and is possibly the way forward for the industry. Dictionaries mixed up with classification methods is challenging because we have different dictionaries and different classification methods, but possibly this ‘pull’ will simplify things.
He’s not sure that George’s example, in which he pulled all this information together, is physically possible. He’s thinking about Google as a model of management of information: the information is everywhere, it’s not centralised. It’s copied multiple times and interpreted in a number of ways.
GOERGE replies that if we can add structure to the data then within data libraries its straightforward to be able to identify permissions for certain people in certain roles to see certain information and that can be done through the applications. He agrees with Nick that it has to be pull based rather than push based and we therefore have an ecosystem of data that is being managed and kept up to date from many different sources. But as a user you just want to know information about a particular asset or product, someone like Jim might want to drill into the detail of how it was manufactured and installed, whilst someone else might just want to know how much it cost - both of those could be commercially sensitive information.
JAREK asks how do you search for the relevant information, how do you filter for the project specific and relevant information? GEORGE replies that you create that through queries that are tied back to the applications that are looking for that information. He thinks probably a data workshop is necessary to explain this in more depth. JAREK says the BIM library is trying to be reinvented at the moment and the biggest problem is with management classification of the data to make it useful and shareable and to create an interface to make it useful. JIM and DEBBIE mention the problem of the knowledge competency gap.
JIM mentions a 360 camera which is able to identify the assets within an existing build environment. GEORGE is doing a lot with 360 with various different specialists. Image recognition is the holy grail but George has not come across anybody (yet) who can actually do it in a significantly complete way to make it useful. Build-ups don’t do it - what they do is compare elements to the model, if something isn’t already modelled they can’t do anything.
DEBBIE, picking up on Nick’s point regarding the definition of what is machine readable, suggests that perhaps before the next advisory group there should be a small group looking at what is meant by machine readable. JIM suggests that all proposals should be done in a machine readable format.