INTEROPERABILITY WORKSTREAM MEETING 25-06-2024
RICHARD We tend with these meetings not to have too many attending, we’re not looking for 30-40 people, we want everybody to get involved. The key word in this workstream is work, it’s not about sitting around having a quick chat for a couple of hours, we like people to get involved and actually get down to it.
PAUL OAKLEY I’m an architect by profession, involved in the CAD developments in the 80s and 90s. Then became chair of the Autodesk user group within the UK, helped with a lot of the development of the Autodesk space software in terms of AutoCAD architecture. Implemented the standards for UK, sat on BS1192 committees, PAS 1192 Part 2, sat on various ISO committees. And was head of BIM at BRE for about 5 or 6 years. Now, I’m back to running my own consultancy, working with various parties on asset management and certification training and international work as well.
RICHARD Can I suggest that we start off from square one: what do we mean by interoperability? Secondly, what are we looking to do in this group around interoperability?
PAUL OAKLEY This was a slide that we used talking about interoperability in about 2014. So basically, if everything is in one software solution we don’t have a problem. The whole thing with interoperability is you’ve got information from one person that you need to pass on to another person and you need to make sure that the information that they receive is in the formats that they want. Traditionally, the major issues were between all of the design team members, so architects couldn’t pass information to structural engineers, mechanical engineers, because they all use plug-ins on different variations of CAD software. And the perception was that BIM was going to resolve all of that, but we’d still got these sort of interoperability issues.
Now we’re thinking bigger picture, thinking more about how we deal with the thing as a whole. But as with the information back then, it was really the purposes why you need to exchange information. So, why do people want information? And what other people can provide the information to them to fulfil those purposes? That could be something as basic as an asset manager needs to do room booking service, so they need all of the spacial IDs, room capacities, equipment within rooms as a particular purposes. It could be cost, it could be all sorts of purposes. So, one of the tasks that we need to do in this workgroup is look at the purposes, particularly from an asset management point of view, why we want information.
The next bit is the software solutions that are used, and we're not really particularly talking about the design software. There's plenty of information already out there on that side of it, but really from the asset management point of view, what software is being used? This is where we really need your feedback on other things So, the software, obviously the vendor would be useful, but what sector is it used with? Other sort of useful information that we could collect that could help us deal with those exchanges in terms of the software that could be producing the information and the software that could be needing to receive it.
And also one of the things with interoperability is things like is it a one-way exchange? Or do we need to be able to roundtrip information? So, there is a need to go from one solution to another and then when it’s updated go back again. So, we need to look at some of those sorts of things. Data formats, so what formats do the software exchange information in? And then what data is required with them? Those are all of the key bits about interoperability to make sure that we can exchange from one to the other. One thing we need to do is to look at the purposes. There’s plenty of stuff on the CAD, BIM. This was 2014, a list of software that we looked at with Building Smart UK, there’s tons of software there. In noticed this image on LinkedIn, put up by Catenda, which has a whole load of software there. So there is plenty of stuff on the software that’s being used at the moment, but this is not particularly from an asset management point of view. So some of it deals with those, but I think it’s looking specifically from an asset management and facilities management point of view we need to look at.
Schemas and open data formats. Again, back in 2014 the slides were presented back then. So, there’s plenty of formats and steps at DXF, DWG which were used for 40-odd years, DXF I was using in the 80s for transiting information. Even down to basic CSV, JSON, XML, RDF which are open formats.But really it depends on the data schemas in terms of making sure that the data schemas are correct if you’re passing information in those formats. Which is where obviously things like IFC and COBie which have identified schemas, also make it easier to pass the information. But again, it’s looking at the formats that we can use on those to pass the information. So that’s some of them, but the particular software solutions and that will then help us work out whether we can carry out the transfers in existing formats, or we need to build specific data templates and structures to deal with those sorts of things.
RICHARD Paul, when we’re talking about interoperability, are we talking about interoperability of software? Or does it cover other things as well?
PAUL OAKLEY It’s how you exchange between people who want information and people that need it. Which is generally going to be via software applications most of the time.
NICK NISBET The Government Industry Operability Group looked at interoperability and we kind of moved on and said actually the key definition of interoperability for the industry is the ability of information to exist independent of the application used to create it, or the applications used to consume it. Because there are plenty of point-to-point solutions, APIs, direct links, which are offered by various software companies, often part of their lock-in strategy, and that’s not really interoperability.
PAUL OAKLEY I totally disagree. If you’ve do point-to-point between two software solutions to meet your needs, then that is interoperable.
NICK NISBET Well, it’s a point-to-point, it’s a point in time and it may solve a one off need, but you can’t guarantee it and you can’t rely on it. And we are talking asset management here where we’re talking about 60 years, not a matter of months. And so I think the point behind this shift was to say that in order to avoid those kinds of vendor lock-ins we need a vision of information as something that has an independent existence. And yes, some of the formats and some of the schemas that you’ve mentioned help with that, but the important thing is that information created one year should be able to be used the following year or 40 years later.
So, I think there’s huge industry or application vendor focus on the marketing of specific APIs to say therefore we are open, I don’t think that counts as open. And really we’re not serving the industry by saying there is software that will do this for you right now, because that’s not a guarantee that it will work in a month or a year or 2 years, whatever the update cycle is. So, I think we have to be really careful that we don’t go to people and say, yeah, if you’ve got something that works right now you have solved the interoperability question.
PAUL OAKLEY But that’s where I was talking about, you need to identify with any exchange whether it’s one-way tripping or you need to do round-tripping. So this is where understanding what the different software solutions is doing and is it a case of this is from design to asset management and then that’s it, it stays in asset management for the lifecycle. Or, it needs to be continuously round-tripping, so therefore the responses are different.
NICK NISBET The round-tripping question is always tricky because any application has cleverness built in which may not be transmitted. But yes, we can look at specific purposes and specific exchanges, but in the context of asset management the last thing you want to do is to say, well, we were able to get information from brand X to brand Y two years ago, therefore your information is safe. It isn’t.
PHIL THOMPSON For people’s information I was involved in the working group that out the guidance together for the code of practice for interoperability. But just the comments that Nick made then, they resonate with me. I don’t think you should look at this from a perspective of the vendors, you should look at it around the data. The vendors need to fit around the data, if you build a solution around what the vendors currently offer and what they’ll potentially offer in the future, you’re doomed to failure in the long term.
PAUL OAKLEY Initially the whole thing was to look at the purposes, so what do we need information for? And what purposes, from the asset point, do we need information for? Then, to understand what existing vendors and software and meeting those purposes. So, the purposes are the key things that we need to identify what we need those things for. Now, we could have a situation where actually they’re all using open schemas within those software solutions and there is no interoperability issue.But I don’t believe that 8until we’ve actually done the work, or at least started looking at this, we need to find out what is it that we’re actually dealing with here.
PHIL THOMPSON So what is the remit, the purpose of this group? What are we looking for the outcomes? Because if the code of practice for interoperability already exists, are we in danger of duplicating some of that work which is already done.
PAUL OAKLEY No, because we’re looking at a detail specifically. All that provides is a high level framework.
PHIL THOMPSON Agree, so is that our place to start?
PAUL OAKLEY No, because that defines a general framework which any good practice on interoperability should be looking at. So, it’s there, but that works already been done and completed. What we’re trying to do is move on to look specifically from an asset management point of view in terms of what are the purposes that people need information relating to asset management. I’m saying asset management, I’m using it in the broader…to manage and look after buildings, whatever you do with them, what are the purposes you need information for.
GEORGE WE have to come up with stuff that’s practical, otherwise we’re going to not be taken onboard. That GIG work, Nick was done probably four years ago.
NICK NISBET The definitions were published a couple of years ago and are online. Yeah, it started 4 1/2 years ago, maybe 6, and the deliverables are now out there.
GEORGE So, how many software applications in asset management have implemented them?
NICK NISBET The code for interoperability? Or that definition?
GEORGE The schema, presumably. What did you produce?
NICK NISBET The aim of the GIG was to promote client as client bodies, all the different departments and industry working together. And they agreed on this definition as being their requirement. The implementation of that has involved obviously looking at COBie and looking at IFC and looking at GML for geospatial information.
GEORGE For me, I’m working in the real world with people that are actually trying to do these things. They’re existing buildings and they are using existing software with data sets that have been put together over this last many years. So, I think we have to be practical and pragmatic about how we can help improve the level of interchange. And I would agree with Paul, interoperability between two software applications, you are actually…and API is interoperability.
It might not be the ideal way of doing it, just as Esperanto was supposed to be the international language. But people have stuck to their own implementations. So, I think that we need to be pragmatic, from this perspective I want to be able to encourage people to work in a way that they’re not just collecting things as PDF documents because that’s where we’re at, at the moment. People are not adequately exchanging the underlying data.
NICK NISBET I tried to outline the risk involved in those point-to-point solutions. One of the things we were offering is that irrespective of whether you use the file technology, or the database technology associated with COBie and IFC, it offers the industry a way of thinking about their data which is more complete than others. So, the independence of it is that you’re using a set of ideas that have been worked out over 30 years. So even if you don’t use any particular file representation, the work has gone into, particularly the IFC, but COBie 2 and COBie 3, offers the industry a framework for mapping its concepts and so on. So, that’s the other side of it- that the ability of information to exist independently really depends on recognising what the shared concepts are.
PAUL OAKLEY we haven’t gone into the details of some of the sort of things that we’ll be doing and the solutions that we have to support this. So, as Nick does know but most people don’t. I was also on the data dictionary ISO standard group. And part of the work that we did at BRE was we developed the Templater solution which is used with Active Plan, and Active Plan have already given that to BIM4Housing. So that provides a data dictionary which allows us to hold data definitions of things. So you can define what your things are, then you can define groups of information associated with those things and then you can define properties within those groups.
So when you look at particular purposes you can then define what things are associated with those purposes and then the groupings of properties and how you then define those. So, those can then be used through open formats, such as COBie, IFC, or any other things, for carrying that exchange in open formats.
RICHARD Can we just take a step back, please. Are we all clear what we’re doing in this workstream?
PHIL THOMPSON No.
RICHARD Are we clear what we’re doing? What we want to do? If this workstream goes on for 4 or 5 meetings and then we’re finished, what is the output? Where would we have got to? What are we looking to do?
ANDREI GRIGORAS Not quite from my end, Richard, sorry. I’m from Enfield and I work on asset management, I’m the asset data analyst. For me, it’s not quite clear the outcome. By interoperability we aim to align APIs through systems, that’s what I got from the discussion.
RICHARD Paul, why don’t we get a sentence together that tells everybody what we’re doing. Is that a good idea for a start point? Because I think if we aren’t agreed on that then we’re going to be chasing down rabbit holes for the next 5 weeks.
KARIM FARGHALY I can give you the sentence if you want. I came with the idea that when Paul started with the why, which is functional and non-functional requirements, we shouldn’t come and solve the interoperability aspect that Phil and Nick mentioned which is in the academic board called synthetic interoperability. I thought we were here to solve the semantic interoperability which is what information is required, how we can keep this information, how we can maintain it for 60 years, how it can be structured and common between all of the software, despite what software you are using. So we are telling people please provide this data points in this structure in that format, for our asset to manage for the 60 years.
We can’t go and talk about…I know we could talk about COBie and how good it is, but to be honest in the industry it’s not still used. It doesn’t have all of the data points that are required for asset management and I understand that it’s left like this for a purpose, but we could help asset managers in this point.
RICHARD Karim, give me your sentence again.
KARIM FARGHALY So, semantic interoperability, to make sure that the dried? 24mins 34secs data sets required for asset management maintainable and can be transferred from one platform to another.
RICHARD Does anyone disagree with that?
PAUL OAKLEY No, I don’t.
RICHARD So, are we adopting Karim’s sentence? Are we unanimous?
PHIL THOMPSON From my perspective, I'm talking about data sets. What about the data specifically relating to an item?
KARIM FARGHALY IN my definition, it’s a data set still.
PHIL THOMPSON OK, so how does that work in a transactional model?
KARIM FARGHALY A very good question. Every data set should be associated with an element of an item because we are because we are object-based orientations. So, the item is activity or the item is an element, I don’t care for now, I need to identify this data points and attach it to the right item or object. And then the way it’s structured, we can talk about this in session 5, but after we identify them. This is how I’m coming to this meeting.
PHIL THOMPSON I see where you’re coming from with that. Where my experience is is around systems and procurement and transactional data. So, in the real world, when we move items around, not just in a digital environment in a model or a twin, but when you actually procure products from the merchant, what piece of interoperable information do we need? What does the contractor need to send to the supply chain? How does the supply chain confirm that that’s what they’re going to send?How do they know that that’s what they’ve picked? How do they know that’s what they’ve loaded in the logistical model? How do we know that’s received at the site? How do we know that’s installed in the building? All of those things form part of the golden thread and they’re all transactional elements, but that information is not ready yet until it’s ready to be used in the model, but it’s part of this process.
RICHARD Who can bridge the gap between Karim and Phil?
PAUL OAKLEY The sorts of things that Phil was talking about, there was research projects such as tag & track that were done by BRE and various others which were then looking at that whole flow of information from initial concept to actually delivery onsite and the whole tracking system and logistics and all of that. I think that is outside of the scope of what we’re trying to deal with. We’re focusing particularly from the asset management point of view and their purposes. Some of that may then come in later on, as Karim said that’s 4 or 5 meetings down the line at least.
PHIL THOMPSON But what’s an asset, then?
PAUL OAKLEY A thing.
PHIL THOMPSON An asset in my mind is something that you’ve procured and then installed. It’s not an asset until it’s passed to the client.
PAUL OAKLEY A person is an asset, a building is an asset, a site is an asset. Anything of work is an asset.
PHIL THOMPSON OK, so is a drop-in anchor? 28mins 3secs that we get from the supply chain an asset?
PAUL OAKLEY It could be.
PHIL THOMPSON Ok, so then the scope of this work needs to cover all of those elements, right from the actual design all the way through to the transactional process to the building completion.
PAUL OAKLEY But this comes down to the purposes, what purposes do asset managers or people looking after buildings want the information for?
ANA MATIC Can we start smaller and then grow? Rather than start very big. Would it be possible to focus on maintainable assets? Because we all know what maintainable assets are.
PAUL OAKLEY Ana, at the moment assets and that are I think to start off with just the basic purposes, what do we need information for? Not the assets.
ANA MATIC No, but I’m trying to say can we focus on resolving this issue only for maintainable and fire safety critical assets.
GEORGE Or purposes.
ANA MATIC Or purposes, yeah.
RICHARD I think what Paul is saying is yes, when we get that far, but we haven’t got that far yet.
PAUL OAKLEY Yes, to start off with let’s just focus on the purpose that we need information for.
RICHARD Let’s talk about the information first.
ANA MATIC OK, agreed.
PAUL OAKLEY But then we filter down to other things.
KARIM FARGHALY Can I reply to Phil very quickly. This is a quick exercise that we are doing now at UCL as a research project. We started top-down approach, we took some sustainability certificates, so EU taxonomy ??? 30mins 9secs, and we looked at the requirements and we turned these requirements to information requirements and then to data points, And then we put the data points to associate with something. So sometimes, Phil, it’s associate to an activity, or it could be associate to a document, so lifecycle assessment or anything.
Or it could be associate to an asset which means this data point needs to be in the material passport or a building passport or so on. So, we can do exactly similar, but from the perspective of asset management. So, I will echo Paul, that we need to start with the why, why do we need this information? What are the requirements? And then let’s turn this to information. And during this we can associate it to an activity or to a phase or to an asset or to whatever.
RICHARD I must just stress with these workstreams, this could be phase 1, if you like. Some of the things that we’ve been talking about already could be phase 6. We take it in stages, each workstream completes a stage. it then may follow on to another, or another 5. At the moment we haven’t determined that.
GEORGE Alan from EARS, you’re providing onsite installation and inspection and things like that. So, the information that you need, and also the information that you produce, is critical, but it’s in lots of different systems, Now, one of the things that I’ve learned by working with people that are actually doing a lot of this stuff is that, in many cases, the asset information that’s being collected isn’t about the asset, it’s about the thing that’s been done to the asset. So therefore the relationship between the activity and the asset that is being related, that is often completely broken. so those are the sorts of things that I think the approach that Karim and Paul are taking enable us to get to. It comes back to this why is that information being collected? Is it being collected because you need it for compliance purposes?
Which means really you just need to know whether it’s been done or not. If you want it from a proper assurance perspective, you want to know what the tasks were that were completed. At another level you want to know what the competency of the person was that actually carried out the work, So, it is very complex, and therefore if we can start with the purposes, why do we want the information, we stand a chance of making some immediate progress. And I do think we have to have some immediate progress because, in my experience, this has taken far too long.
NICK NISBET The idea of purposes is very powerful. In my experience it comes back to we want the information because we want to avoid the uncertainty. Uncertainty may be risks or it may be opportunities, it may be a positive thing that if we have the information we can do something better. And so the model that we proposed as part of the golden thread was you start with the risks that you’re trying to address and from that you can identify which physical, spatial activities the risk is associated. And that starts to tell you what assets you need to track, and it may not be the individual fixing bar or whatever it was.
It may be that it’s just a whole system or a whole level of a building. So, some of the risks and opportunities that we want to manage are the long term risks and uncertainties and opportunities, and hence my definition of interoperability. But the purposes don’t exist unless there is some kind of choice/uncertainty/risk/opportunity attached to them. The short term risks can include being required to have documentary evidence of stuff for compliance purposes. But what we want to do is, above that, define what is the information that is implied by all of that evidence which is supporting what we’re trying to do. So, we start with the risk register and say which is our top risks and how are we going to address them.
And, as Ana said, a lot of people are trying to think about fire safety management at the moment, for all sorts of good reasons. And from that we can work downwards. And the other case that we heard was commercial transactions where the risk is non-performant products, I guess, which is what you’re trying to manage out and so on. So I think there is a process that we can say, well, what are the top 10 risks and opportunities that asset managers are trying to get a handle on? And that tells you what information you need to be transacting with people.
GEORGE OK you may not know this, but what we’ve done in BIM In Asset Management, we’ve learnt a lot from doing the BIM4Housing sessions. And what we’ve done is broken it out into smaller chunks that people can deal with, and what we’re doing here is one of those smaller chunks. But we’ve also got, we had a business case session last week which was very strong. So, the reason we had the business case one was for the exact reason that Nick’s just said there, actually identify what it is that’s going to be the driver. So, the business case group, that’s what they are concentrating on.On Friday, we’ve got the digital records group.
The digital records group, we’re trying to keep away from the interoperability bit because the digital records group is then looking at what type of record is likely to be produced and how are we managing them. And then we’ve also last week, we had a fire critical assets group. The purpose of that is to try and define what do we actually mean by fire critical assets. And what information should we be collecting about them. So, we’re trying to break this down into smaller chunks that we can address and I think if we could come back to Paul on what we’re maybe doing on purposes. I think we’ve got a definition, we can obviously evolve it if you’re not happy with it, but let’s start with the one that Karim has given us and come back to Paul.
PHIL THOMPSON I’m happy to go with the flow and I appreciate this is highly complex, or else if would have been solved a long time ago. But you made a point there about record keeping, which is great and I see that’s another stream. But how can you keep records and things unless you’ve got interoperable information in a business system? So, it’s kind of like one is intrinsically linked to the other. We can’t ignore the stream, we can’t separate these too much.
RICHARD We’re aware they’re going to overlap. It's just starting off we’re trying to keep them as separate as possible for simplicity. There will be joint meetings as we move further along.
GEORGE And they’re all recorded, so it’s absolutely open and transparent. So, if you’re interested in seeing what they’ve done, what I tend to do is watch them at 1 1/2 times speed which comes over quite well.
PAUL OAKLEY We hope to get a Miro board set up to start being able to populate these, but we need to get a way of collecting the information. So, we start looking at purposes, and Nick’s mentioned risks as one area, commercial opportunities. Again, there’s going to be a whole seres of different collections of purposes under different things. So to start off with it’s to basically mind-dump specific purposes why people need information in terms of what they’re trying to achieve. And then we go through that and start categorising under particular things. So, things like a risk, one set of purposes and meeting legal requirements or other things, with things like the fire is one of one of the key areas that they've been looked at the moment. If we’re looking at purposes, what sort of information should we be collecting against the purposes?
GEORGE Contractual commitment?
PHIL THOMPSON Can we not do this on a Miro board or something like that? And then everyone can map their ideas into it simultaneously.
PAUL OAKLEY we will do eventually, but unfortunately we couldn’t get it set up in time for the meeting. I think if you just put things in the chat for now and then we’ll try and fire it on to the Miro board for the next meeting. And then hopefully we can then build on it from there.
GEORGE We can set them up on the Miro board and give people access before the meeting so that people can contribute before hand. Jarek and Das from the BIM4housing design group have set up the Building safety Act Miro board, we’re getting quite a lot of good input to that. We’re going to align this to that. So, we’re putting stuff into the chat?
PAUL OAKLEY Yes, if you can think of, particularly those that are basically maintaining buildings, what are the purposes that you require information for. Also it would be useful, do we need to know things like are those purposes sector specific? If people have got general ideas about other things we need to gather against those purposes or gropings of them, that would be useful for input that way as well.
KARIM FARGHALY Paul, there is a Miro board in the chat, do you use this? And it has an interoperability part.
PAUL OAKLEY The trouble is we haven’t populate anything yet.
JAREK WITYK I just sent that link, at the moment it’s a blank space for this purpose because we literally set this up 5 minutes before the meeting. But feel free to dump in anything there, make comments. which will then be used and picked up.
GEORGE Why don’t we do that? Can we put stuff straight onto it, Jarek?
JAREK WITYK We can do. I can share screen and start typing whatever your’e saying if you want.
ANDREW WARING I would probably reflect some of the things that Nick said right at the start that if we focus on discrete interoperability of exchange of information between multiple platforms, multiple requirements, it can grow arms and legs quite quickly. Whereas if we focus on a centralised proposal for storing information then we can start to solve issues like ingesting the information into that kind of platform, finding information on the platform, finding information on the platform, whether through identification GS1 coding etc.
What that platform actually looks like, because I’m aware that GQL as a standard has just been ratified by ISO, so as a graph relationship database it’s quite a good starting point for this kind of posting of an IFC-based schema as a live solution. Because one of the biggest issues that I’m finding with, even if we do produce a nice IFC fill of rich data, is it’s still difficult to keep live as the building progresses, so having a live relationship database based on GQL would be a good proposal for that and then we could just start to solve how that is worked with, from an ingesting egress from a set up and management point of view.
GEORGE I think the point is, though, that’s the how, so that’s probably 3 or 4 meetings away. What we’re asking now is the why. So, the purposes, why do people need the information? So, what are the…maybe Stuart from ASFP’s point of view, why would you need information?
STUART SOUTHALL From the ASFP, obviously we’re looking at passive fire protection side of things specifically. And what we’re needing is how the architect’s design can be recorded, transferred to the contractor design, how does that link? You’ve got BIM software, Revit etc, how does that link to the contractor software that they’re using to design these specific systems, like a fire wall, damper systems, intumescent paint and all of that. And then how does that then link to installing it, recording that information of how it’s installed? That’s the bit we’re struggling with, how that BIM software links to them installing it with their own software and then down the process, right through into investigating the work that’s been done in handover stage. Does the design at the start match what’s actually been built.
How is that data recorded? Is it easy to look at that information for someone coming to the project and going, yeah, I can see that what that architect designed, intumescent paint on structured steel, is what’s actually been installed. And then you can follow that cycle through into maintenance contractors coming in. How do they interact with the fire systems? How are they able to see warnings to say don’t touch this wall? Are they able to look at that information and go, well, if we do this to that, what do we then have to do after to make that systems fit for purpose and fire safe? It even comes down to the people living in the apartments, what information do they need access to?
Do they need to see what the building is constructed of? Do they need to see what cladding is on there if they have concerns? All of these kinds of information, we have it available, but as people a lot cleverer than you or me in terms of information systems, is how does it transfer? And then how can you look back, at the point where people are living in that building, and can see a specific part of that building and that everything matches up.
GEORGE I have two observations. One is that what you’ve started to jump into there, and this is very common (particularly with engineers), is you come to the how and you start looking at the how and what the solution might be, before we get into the why> But then towards the end of that you started talking about the why. I think if we can start with the why, then we can produce the how much more effectively. So, the example that you gave there of tenants is a really good one.
Do the tenants really need to know what the design construction and the specification of the cladding is? Probably not. It may be that they need to have access to that if something happens, but they probably wouldn’t want to know that. They’d probably just want to know that it’s safe or not and that it’s been inspected. So, both of them, it comes down to purpose about why is that information needed because that will define the how.
RICHARD But George, you can’t have a why without a what. Why is what information needed? You’ve got to identify the what first, surely.
JAREK WITYK No, I disagree, you need to identify why first and then what.
RICHARD Why is that information needed…what information?
JAREK WITYK OK, they do link, but first is why, not what.
RICHARD Give me a why question that doesn’t include a what.
JAREK WITYK I want, as an organisation, to build a bridge. So, what is the bridge, but what I want to do is the purpose. I want to build a bridge first.
RICHARD We’re talking about information, aren’t we?
JAREK WITYK That’s related to that, then there will be information as a result. The information on its own…what’s the purpose of information? You have to…
RICHARD Why do I need this information?
GEORGE Let me give an example. Let’s say that Alan’s contractors need to be able to go and inspect some fire doors. The reason they need the information, they why, is they need to know where the fire doors are and they need to know what specification those fire doors should perform to so that when they actually carry out the assessment they know that they’re checking the right asset, and they know that the records that they’re taking about that are actually going to be recorded going forward. So, that’s the why they need that information.
RICHARD But that information needs to be identified to be there.
GEORGE Certainly.
RICHARD And that’s a what.
GEORGE But you don’t know what that is until you know that…the why is I’m being asked to do a fire door inspection. That tells me that I need information about the fire doors.
ALAN WHEAL I get exactly what Stuart has just explained, from end to end. If we’re talking about what assets are the most important to deal with, we’re talking about the golden thread, isn’t that about assets that are fire related assets. If they are fire related assets, surely the insurers should be helping us with this document. Surely the insurers should be telling us what they want because they’re insuring he buildings, they’re insuring the tenants, the occupiers, whether its commercial, residential. The problem is they don’t know what they want because engineering is such a broad…you talk about fire doors, George.
When they said fire doors need inspecting annually by an inspector that’s done a 3 day course, guess what we get as an industry? Thousands of fire doors being replaced that don’t need replacing. The problem is that we’ve got an immature industry, within the middle of all of this, called the fire industry that hasn’t grown up with BIM and modelling and Revit and all of the kind of good IT systems that are needed. Most of the operators are too small to even have an IT system. So, for me this is a much bigger problem and somewhere we’ve got to find a solution quickly which is why I’m on that because it’s a massive issue, Because the operators that we’re going to want to jump on this type of system won’t even know what a system is.
PAUL OAKLEY I’m trying to read through what people are putting in…put up some notes to start off with in terms of things we’re starting to capture.I think people getting stuff into the chat. Or, if you think I’ve misrepresented things that you’ve said just shout out, because I think the what, why and that are all important.They follow on from the initial set of things. All of these will develop to push the whole purposes piece.
GEORGE For example, if we look at what Alan was just saying there about the golden thread. By the way, the golden thread isn’t just about fire safety, it’s actually about structural safety, but it’s also about other risks as well. Obviously you’ve got things like mould, ventilation, there’s a whole series of different things.
PAUL OAKLEY The likes of legionella. Obviously a lot of us have come from the housing side where the focus has been on high-rise housing, but if we’re looking at this now this is a total asset management group. So that’s why I think that some of the purposes may be sector specific. So, it might be a case on certain things, some things will just be generic, but other things will relate particularly to specific sectors…it could even be infrastructure, never mind buildings.
GEORGE Yeah, we’re going to try and stick to buildings, I think, it’s wide enough. But, for example, PFI. One of the biggest risks for PFI FMs asset managers is data sheets because that’s where they get huge penalties. And therefore the coordination between the handover information, which is in the form of lots of drawings and schedules and things like that, and then the constant updates that are taking place. All of the different variations that are being recorded, many of which are actually being undertaken by the client without even telling the FM contractor that they’re doing it. And then they end up inheriting responsibility for it because the records haven’t been adequately maintained, and then they get penalised for not maintaining the records. So, that in itself is a key part…would that be a purpose, Paul?
PAUL OAKLEY Yes.
GEORGE Most of the inspection applications, and there is a growing number of them which is great that we’re capturing all of this information digitally. But unless they are actually referencing back to the same data set and that can be coordinated in some ways, then we end up with hundreds of golden threads, and there is no relationship between them. So, in many cases the purpose of the information, why are you actually producing that record of which fire dampers have been inspected? You’re doing it because you’re trying to make sure that the building is less risky from a fire safety point of view.
Yet if you’re collecting that information and it’s in a silo, it doesn’t relate to any thing else, then why did you do it? What’s the point? If it’s simply to get a compliance tick then that’s one thing, but if you’re actually trying to do it so you can make the tenants or the patients in the hospital safer, then you need to step back and say, well, actually why are we doing this?We need to be doing maybe a bit more. Fire risk assessments are something that I’m quite amazed of the lack of specification on them. You can have a fire risk assessment, people are bidding that they’ll do a fire risk assessment for £150, that’s probably an hour. So, what do you actually risk assess in an hour n an office block?
STUART SOUTHALL It comes down to the competency of those people doing it as well,, which is what the industry's working rapidly towards is getting people that are doing those reports, assessments that are educated to the right levels and competent in what they're doing so the information's clear and correct, guidance that we're putting out on the basis of that as well. So it's a challenge at the moment, George, and it's been a long time coming, but there’s work going into solving that problem that you just set out there.
GEORGE Yeah, I think the other thing is that probably the bigger problem isn’t the people that are carrying out the work, it’s the people that are procuring the work. So it’s important that if procurement is being done though a procurement department where maybe the head of fire or something like that is actually simply specifying what they need. And then a procurement person then is looking at bids that come in and they pick the lowest one, for example. That’s a competence thing as well. So we’ve got tο be careful that we’re not trying to boil the ocean here, that’s why we’ve got these different groups to try and look at the different elements. The business case group, for example, would be looking at what level of specification, why are we actually doing it.
PAUL OAKLEY Alan, you said compliance, could you just expand on that please?
ALAN WHEAL I think if I’m looking after any type of building as a responsible person or the FM responsible person, I want to make sure that I’m fully compliant in my job. And I don’t think those people have the ability to know whether they’re being told the right thing by their supply chain. So it’s yeah, we’re gonna give you a certificate, they’re happy they’ve got a certificate, but was it really compliant? You’ve got to get to the point where, well, I don’t know who audits this. i guess it needs an audit process.
PAUL OAKLEY You've got compliance with legal requirements, contractual requirements, but what you're talking about is the tracking to make sure the information you have is compliant through the process.
ALAN WHEAL Yeah, and if you look at what you say the NIC and Corgi have been successful in doing what they’ve done. There’s a whole world of authorised person that’s sign in certificates and those certificates get audited by the NIC IC. So, it’s a ? 1hr 4mins 41secs system, but it’s also got an audit function.
PAUL OAKLEY Would say you also talk about compliance verification and validation?
ALAN WHEAL Yes. And if it’s my job to be the person that’s in charge of signing that I’ve got a compliant building, we’re going to end up with having no one that wants to take them jobs, unless we can get this right.
NICK NISBET There is a spectrum of compliance, starting with information compliance. Probably then you might list contractual compliance and at the extreme regulatory compliance. But they all require people to commit to making the requirements machine operable. That can be done. And then the D-COM project which was an Innovate UK project done by Cardiff University and several others, showed how you could create public records of which buildings have passed which regulations. And the opposite, private records of which regulations that a particular facility has passed and so on.
So, the infrastructure is there, people commit to machine-operable requirements and, of course, machine-operable information. And that can include handling things like supporting evidence, which is not data, it’s data about that data. But I think the tools are on place, the challenge is whether people are prepared to commit to the letter of their contracts, or the letters of their information requirements, or even the letters of the regulations. Most people exploit the ambiguities and leave them there in order to create bargaining positions.
PAUL OAKLEY Karim, you’re right, there is a process of getting the stickies down the different high level things and then it’s the way that we deal with things like grouping, tying things, together, cross-referencing and linking things. Because I think that’s the next stage after we dump the list of things down and try and rationalise some of the things. Because I think we then start to go through into these.
ALAN WHEAL One of the things that we haven’t talked about is this whole carbon commitment and ESG. That is big on the agenda at the minute, they’re having to do almost a golden thread on equipment for tracking carbon, right? So, can we learn anything from what’s going on it that world?
GEORGE Sure, I’ve got some exposure to that so i can bring that.
PAUL EYDEN I’m the BIM manager at the University of Warwick, so I deal in students rather than tenants. But there’s a lot of similarities, I’m sure. I just wondered with a lot of the whys that we’re putting on the board whether they in fact overlap quite a lot with the digital records workstream. And I was just reading again that purpose sentence that we put down at the start and whether we need to be looking at the why we need a unified framework. Does that need to be something that this group is considering? The pros and cons. Are we recommending open BIM? Are we recommending that people should stick with native design file formats? There’s a whole conversation around that and I just wondered whether that would be one of the outputs from this group.
PAUL OAKLEY My interpretation is it should be evidence based, so basically if we say here’s the purposes, here’s the software solution and they why. If the software solutions that are used all use a standard exchange format, then there is no requirement for an open format. The odds of that happening is very, very slim, so I don’t think we’ll be going that way. But until we’ve actually gone through the process and identified what is used. It was the same in the design side of things, we go back to what happened in Teamwork in the 2000s. The first Teamwork, everybody was forced to use a Bentley solution, so we thought everything would be interoperable because we’re all using Bentley solutions. But actually we found even that wasn’t interoperable between their own file formats. Then when we went in the following year into anybody could use any of their own software, then we found the limitations of open file formats. So I think that some of the work here in terms of let’s not prejudge what we’re going to come up with or what the solutions are going to be further down the line.
PAUL EYDEN Yeah, there’s also that issue of the longevity of this. 19650 talks about triggers, what are the triggers that are going to effect a change on that asset data for the life of that building, because we know software changes over the years. Is that something that we need to be aware of and make a record of? That’s just another angle on that.
PAUL OAKLEY Yeah, but I think this is where having understood…basically we’re talking about object models, understanding here is a thing, here’s sets of information against that thing and here are the properties. And ideally down to the units of measurement, enumerated possible values and those sorts of things. If you define that as a data schema, then actually the software is just going to take onboard that schema. But it’s whether that is necessary or not, I think that this process should help to dig that out.
ALAN WHEAL Is there a thing of who will own this data? Is ti the building owner that needs to own their data?
GEORGE I think that’s probably more for the digital records team to consider, and the business case team. Which obviously anybody here is welcome to be part of. I think what we’re looking at here is looking at the interoperability element of that so that we can feed into that. But, it’s a very good point, and actually I’ve just had a message through from an FM that I asked the question this morning, what would make your life easier? And he said that typically the problem that he sees all of the time is that clients don’t know what assets they’ve got in their buildings and they then try and pass the responsibility for the accuracy of that to the FM company. Who then buy in CAFM systems, but don’t actually engage too closely in it, so you end up with a set of not exactly validated information.
And then the next time that’s done, it’s completely different. The client might have several buildings, all being managed by maybe different FM contractors for different purposes, and they’re all set up differently, so there is no standardisation across them. I think that’s an important part of interoperability, if we don’t have that on here could somebody put it up? Standardisation really is important as far as things are concerned because you’ve then got a chance of managing it.
ALAN WHEAL But that’s not strictly true, in the FM market we know that they did go out and do lots of conditioner surveys and they did get their assets in order. But they don’t want to give them to the client because if they lose the contract they want the pain to be with the next contractor. So, my view has always been, like a fire alarm big system panel software, if the clients have paid for that system they should own the management data on that software. But over the years we’ve seen many, many fire alarm companies have closed protocol software systems and they’d be unreal pain for the client.
So, is the purpose here the end client, the operator of that building, the asset owner of the building gets their own data to be able to operate their building properly? Because historically, the responsible person or an FM manager would be sitting there going, well, I know where to get that data, it’s over there with that contractor, it’s over there with that contractor. If I need it, if I’m asked for it, I’ll go and get it. They don’t have it to hand, iO&M manuals are destroyed the day after PC, in my opinion.
GARY McCORMICK I’m from Sovereign Network Group and I’ve been working with various HAs? 1hr 15mins 44secs over the past few years on exactly this problem and George and Alan have hit the nail I’ve been trying to hit on the head, in their last two points. And that is a purpose, I think, Paul. And that is BIM, OK, we’ve got BIM, we’ve got a myriad of things that we need to do for BIM. We’ve got a big brown fence that all of this BIM data gets thrown over and we’ve referred to then today as asset management. Now, asset management have got a fixed agenda of certain things like compliance, planned maintenance, like various things which could be financial etc. But there’s a thing further to this, and Richard and George, this is why I want to talk to you on Friday afternoon, and that is basically that we’ve got people, and Alan and George, to your exact points, we’ve got people in our organisations.
Your first point, George, we don’t know what’s in the building. Now, if we look at building safety and we look at the golden thread, failed. So we’ve got all of this conundrum to solve, but the people solving that conundrum on a day to day basis are people like contact centres. We’ve got a stickie on there, repairs and maintenance, first time fixes. The repairs contractor goes out to do a job in Flat 14 of Elm Court, he may know what the job is, he doesn’t know what the component is that he’s dealing with, he doesn’t know exactly what it is that he’s walking into. Oh, we can’t do the first time fix, we’re going to charge you again for that piece of work. The contact centre have people ringing up saying I’ve got a problem with my smoke alarm and they say what flat are you in? Oh, right, you’re in that flat. What floor is that on? Oh, you’re on that floor.
They’ve got no spatial awareness, they don’t know what components are in the building, they look like complete muppets in front of the residents, the resident’s satisfaction goes through the floor. So, all of this data, irrespective of definitions, whatever it is we want to get out of this group, and this is probably a group agnostic point, we need to solve these issues for those people. It is interoperability because there’s different systems serving these people on the front line day in and day out, but they’re not being served. So, that is a purpose, that there are people trying to do their day job properly and they’re not being able to because they can’t, if the data is available, they can’t get that data. Because we don’t know what’s in our systems, and even if we did we can’t get it into the systems that are used to do their job.
PAUL OAKLEY So you're saying what's in the building or where are things in the building?
GARY McCORMICK So, somebody rings up and says I’ve got a problem with my flat. And the operative that takes that call does not know what’s in that flat at all. They may not know what floor because housing systems don’t hold spatial information. They say it’s got 14 floors, but it doesn’t say flat 47 is on floor 8. So, COBie is good for that, is COBie the right way to go? I don’t know, but we need this kind of granular information to serve the people who are doing their day job to the people…we’re on this call for residents, that’s our primary purpose.
GEORGE Just to reinforce what Gary’s saying, we’re doing this day in and day out because the example that Gary’s just given is a perfect one. What flat are you in? I’m in flat 54. What’s the UPRN for that in the asset database? Obviously the user, the person on the help desk, then connects to that, and actually they find that the information that they’ve got in the housing database has got a different flat. Because when the building was delivered, all of the areas and the flats were called one thing and then it’s never been transferred over into the new naming convention, not updated. That’s a classic example.
The second one is if you’ve got something that is maybe part of a system, where does that asset sit in the rest of the system? What’s it connected to? So, if I’m carrying out a piece of work, what do I need…if I disconnect it, what other areas are going to be effected by that? In many cases, the asset information doesn’t give you that at all. You’re lucky if you’ve got the individual asset.
ALAN WHEAL I’ve got a funny story about a housing association who were doing decent homes upgrades to kitchen and bathrooms. Someone that didn’t even own the property got a new kitchen and bathroom because the asset management was that poor on this housing association. So, it’s pretty basic stuff that we’re talking about.
JAREK WITYK Just to refer to what you said, George, with records. I think we’ve got a massive problem in industry, (unless I am the isolated case, tell me), that the recorded information, as-builts whatever you want to call it, they’re never accurate, and they’re never checked by anyone. They accept it as they are at face value, but they’re never accurate. They’re being produced and rushed to achieve practical completion, fulfil contractual obligations. They’re just tick boxes: yes, I’ve got the drawings, yes, I’ve got the O&M, tick, tick, tick, and that’s it. No one checks it and it’s never accurate.
And we’re supposed to have a mechanism in place to check the information, and apparently the client is responsible for accepting it. But the acceptance in turn is not accepted by the client, it’s being pushed downstream, and then that on its own has massive contractual implications. And the people downstream are suffering the most in the end. But then the end result is that the records are just not accurate, they never were. And what do we do about that?
GEORGE I had a situation exactly like that a few months ago. We were going through doing a review of handover information and when we opened up the fire stopping drawings we saw that all the tabs were overlapping each other. So you couldn’t actually read…that fire stopping report was a waste of time. So, we raised that with the employer’s agent who was the one that was supposed to be signing these things off, and they said, oh, we didn’t actually look at them, we just said that we’d received them.
JAREK WITYK If we can solve this, the interoperability we come up with, never mind how beautiful scheme that will be, that will never work, because of that simple one reason.
PHIL THOMPSON Really interesting anecdotes and examples coming out here. And whilst they’re making everybody chuckle on the call, there is a serious side to this. My view is that if you reverse engineer these problems and go back along each stage and identify where it went wrong, you’ll identify the ambiguity of information. And especially when these examples talked about products within assets within the buildings. I could give you examples all day long of where contractors just change things throughout the design selection procurement stage.
And they even don’t check what’s being sent by the distributor, and the distributor has no audit trail about what they’ve put on the vehicle. And then invoices are paid by the contractor without actually any recognition of what was actually on the goods received note because they just want to pay them below a certain value, It’s the wild west.
KARIM FARGHALY I’ll talk about what we should do…(shares screen). This is what we have done and I think ti works well with what you’re trying to do. The right hand side where this is the requirements and I believe this is the business case. The left hand side, which is the information container, which I do believe it is the digital records. And this is the part in the middle that we need to achieve in a way. In our exercise we identified the data point applies for what and we give it the classification for the ambiguity perspective. The asset type is project building sone element, product material, the impact the data point type, is it just a binary? Yes or no question. Or is it quantitative, qualitative and so on.
And the requirement type which is something we are trying to do now with the why: why this requirement? Is it for safety? Is it for sustainability? Is it for carbon? Is it for what. So, we did this, we defined this, I know Nick will not wrong, but we structured our data point in this respective, and it worked. It worked to identify people.
RICHARD Is that helpful to you, Paul?
PAUL OAKLEY Yes, it is, it’s similar to the amount of grounds we had in the standards.
RICHARD Karim and Paul, why don’t you hook up and have a chat around a few of these things.
GEORGE And we’ll distribute the Miro board and people can carry on contributing to that over the summer.
PAUL OAKLEY We’ve started to get some things down, but keep firing the purposes that you can think of, we’ll speak to colleagues and organisations that are involved with the management of buildings and really try and find all of the different areas that they’re particularly looking at. And what are the key problems that they need to be, that they want information for. That’s going to help define out purposes and then we can start looking at the other bits. You’ve noticed I’ve tried to start to give some colours to things, that’s sort of an initial set of grouping, and then we’ll start talking about whether those groups are correct and other things to build up from this afterwards.
CHAT
[11:14 am] Nick Nisbet (External)
The ability of information to exist independent of the applications used to create or consume it.
[11:17 am] Joe Stott (External)
Personally I dont think there is a onesize description of Interoperability - maybe we should start by mapping out some scenarios / pain-points that we already know exist. I suspect the recommendations for each would be very different but may hopefully allow us to find common themes etc.
[11:27 am] Nick Nisbet (External)
Item, Space, and Task
[11:40 am] Adrian Mokus (External)
i was expecting people joining this meeting already know we have these separate workshops
[11:40 am] Nick Nisbet (External)
(Risk/Opportunity)->Associations/Triggers->Groups of Items/Spaces/Tasks
[11:42 am] Jarek Wityk (External)
access to the MIRO board via link:
https://miro.com/app/board/uXjVNZlu1W8=/?share_link_id=838109356194
Password: Bu1ld1ngSafety
The access will allow visitors to comment
[11:42 am] Nick Nisbet (External)
GIIG Glossary https://www.ukbimframework.org/wp-content/uploads/2023/06/GIIG-Glossary_V3.pdf
[11:45 am] Alan Wheal (External)
Insurance requirements
[11:48 am] Nick Nisbet (External)
Uniclass RK: Risk classification. https://uniclass.thenbs.com/taxon/rk
Anyone interested in IFC in Knowledge Graphs (round tripping) and GQL. I'm doing it.
[11:50 am] Adrian Mokus (External)
Sustainability
[11:50 am] Andrei Grigoras (Unverified)
Manage at a low level i.e. managing lifecycle, repairs of elements, is it communal repairs or individual residency repairs, maintanance, how tenants could access data points, what data points would they want/need, this is what comes to mind currently. Could add for compliance purposes etc
[11:54 am] Phil Thompson (External)
Why - Remove ambiguity, maximise integrity, create an audit trail, improve process efficiency
[11:57 am] Eyden, Paul (External)
Why - Collect information you need to make decisions on.
[11:57 am] Phil Thompson (External)
Open the board access to all
[11:58 am] George Stevenson (External)
contractual
knowing where things are
what systems they are part of
[11:58 am] Joe Stott (External)
First barrier in regards to building asset information (as things currently stand) is basic registration / transfer of the data we (design team) have put in great efforts in creating from IFC/COBie into the CAFM system(s). Sounds simple but im sure it never is.
[12:16 pm] Eyden, Paul (External)
In my experience, the CAFM provider should provide their own data import templates.
[12:20 pm] Joe Stott (External)
Should being the key bit.. We have transferred lots of projects into CAFM however what is never clear to me is how successful / complete this process is. (i strongly suspect lots of data is still being lost)
[11:58 am] Farghaly, Karim (External)
asset maintainability
asset resilience
[12:00 pm] Adrian Mokus (External)
asset levels of service
[12:02 pm] Alan Wheal (External)
simplifying compliance for the fm or responsible person
[12:03 pm] Hayhoe, Caroline (External)
Why - To keep Buildings and people safe
[12:06 pm] Farghaly, Karim (External)
Compliance could have subcategories based on complaint on what ?
[12:16 pm] Adrian Mokus (External)
standardisation is the key
[12:23 pm] Alan Wheal (External)
Provide trusted live data
[12:25 pm] Joe Stott (External)
So is data provides, transferred and verified