DIGITAL RECORDS WORKSTREAM MEETING 3 29-09-2024
RICHARD BIM In Asset Management, as BIM4Housing, was set up on the basis of outputs. It’s not a talking shop, we have aims and we have a methodology which has evolved over the last four years with the help of all of the people involved. But it’s about collaboration and working, and I don’t know if any of you were in the interoperability meeting the other day. After the first 15 or so minutes I was making a coffee, and it went silent there was no sound and I was thinking what’s happened? And it stayed silent for the next hour, do you know why? Because they were all head down contributing to the Miro board. So, that’s it, that’s what it’s about, it’s about getting things done. (shares slides on screen).
We’re all volunteers, we’ve all got day jobs, so any wasted time is wasting everybody’s time and it’s more than proportional, so we look at achieve. We have working groups which we define as a group of people who are there to look at the bigger picture and direct and come up with issues for the workstreams. The workstreams are very much task and finish, they’re got something definable to do which they complete in 5 or 6 meetings. if there is more to be done, or maybe it divides up, obviously we do that. Our current working groups are safety critical assets, that started out as fire critical assets but we’ve expanded that out, PFI and decarbonisation. Our workstreams are digital records, interoperability and business case.
Now, there’s been a lot of talk in the meetings about crossover because obviously those top 3 certainly link very closely. So we’ve spent a lot of time in the last couple of meetings for each group in getting a definition, making assumptions as to what exactly are we trying to do. We’re looking at what, how, why, and normally we would do that looking at why first because if you don’t know why you want a piece of data then you’re going to get flooded with stuff that you don’t need. We’ve not been prescriptive on those necessary definitions and assumptions, they’ve been arrived at by the group, so what we’re going to do today is confirm and agree those definitions and assumptions so that we can then crack on and get moving.
ADAM SANDERS This goes back to where we kind of started in terms of we’re trying to define this digital record as a concept and if you ask different people you get different things. And I think when I work with clients they talk a lot about the golden thread and they’re sticking documents and scanned items and calling it a golden thread. But the key thing is unless we know what these documents are and what’s in them they are not useful. So it’s all very well having something in a digital format, but you’ve got to know what it is and where it’s come from. It might be that you’ve got historical stuff that’s scanned and you just know it’s a manual and you might look at it later.
But broadly we’ve got two types of record, we’ve got static information which could probably be specification of a product, it might be an installation certificate, something that probably doesn’t change. But we’ve then got this kind of ongoing service records which are relatively fluid which are much more dynamic. So I would have put static versus dynamic, (on the slide) it is static/passive, but these dynamic records which say yes this boiler has been serviced, yes this door has had a recent inspection. And that’s managing the asset and any routines around it. And then we have to think about the record itself. Is it just a document which we’re going to refer to in that somebody will read? Or is it machine-generated data which we need to use throughout other systems?
So, is that document, that record, is it a whole load of information that we can take out and put into other systems? If it was a door it might be it’s an FD30S, it might be what the last inspection date was, and the only real way to make that data useful is using APIs because there is no one system that’s going to make use of all of this data. So, we might have a BIM system, we might have an asset management system, we might have some FM systems or some kind of reactive maintenance management and scheduling software. You might have some compliance management software. The problem is that each of those systems tend to silo data and without APIs to make that software talk to each other we massively decrease the value of that data. And if we can’t keep the data record up-to-date it becomes relatively worthless,
If you’ve got a door survey and you’ve got it in all of these systems or one of these systems, if you’re then going to have another door survey that first one is irrelevant now. So your system becomes worthless if you can’t trust it to be up-to-date. So the only way you can get value in these digital records is to make sure that our systems talk to each other and that that data is up-to-date.
ANDREW PERRY You’re laying it out in a very good logical fashion, but I think if we take it right back to the start, if you like, the benchmark data, the actual data points about where we get the records from the design construction team. It’s how those are made available. Are we making an assumption that actually the base data is going to be available to us in a digital format? i.e. in CAD or a format of CAD that can be readily reviewed and revised in the digital format through the life of the facility. You’re spot on with the records of plans and serviceable assets, if we do have, say, fire doors, if we’ve got a dynamic record of fire door condition and damage, is that gap widening? It’s how we come back to the very, very start…
RICHARD So, your question is are we making the assumption that we’re going to be delivered with that digital…
ADAM SANDERS I think the key here is that if you’ve got a system that’s managing a floor…if you’ve got some kind of BIM system and you’ve got those floor plans you’re going to need an asset management system that deals with it. So if you’re thinking in terms of floor plans, but somebody that’s a facilities manager is thinking in terms of individual assets within the buildings, your BIM system has got to talk to your FM system.
ANDREW PERRY It has, and it should be embedded as part of the API, I think that’s one of the big challenges. I work for Amey complex facilities, we’re doing a lot of Lidar and Matterport scanning to create a digital format that can sit and talk to our CAFM system so that we can then start to dynamically track assets and how they’re used and how they’re deteriorating etc.
ISHKA HEART I’m from SNG a housing association in London and I work in the development department, responsible for helping specify both the documents that we need delivered on new schemes and also BIM models and COBie data. And so the approach that we’ve been taking is very much distinguishing between the two. We want certain documents, we need also certificates, drawings, those PDFs, but then we also ask for data points and now we maybe can get that in our BOM models and COBie data. Or we can get it in just like a spreadsheet and start pulling out key data points from the documents, so I think it’s a really good approach that you’ve mentioned that, static/passive or dynamic because we then push certain data points into our systems as well as maintaining them in the documents.
So I think the way we’re very much starting to think is documents and data which then in the future when they need to be maintained it’s what else are we specifying when we go and change an asset? We’re specifying certain documents and records and the PDF must be delivered, but then also we want additional data fields and where they sit in the system. So, I completely approve of this sort of approach, but distinguishing between the two is really important because it can get confusing.
ADAM SANDERS We’re often seeing, Adina who also runs this group and is not here today, a point that she made previously they put a lot of work into this handover information when they complete a development. And what’s really painful for them is that they know that information model won’t get maintained, especially when it goes into social housing. If it goes into a commercial environment sometimes it’s more likely to. So it’s quite nice to her that the social housing sectors are starting to pick up, but I think the challenge we’ve got is that the people day to day aren’t maintaining that information.
So with the housing management system, that would connect to your asset model and go oh yeah, that vent, that AOV, it’s specified here, it needs replacing. I can see the design requirements were X and it will get replaced with the right thing rather than the contractor going oh yeah, I’ll just stick one of those in, whatever is available off of the shelf.
I work for a company that produces inspection software and we do a whole load of stuff with building safety and it’s so frustrating to watch five different consultants go to the building, collect the same information and it each report, because they consider it a PDF, doesn’t corroborate each other. hey might all be slightly right, but if you gave the next person the context of the building they will start to corroborate each other and they’ll all think in the same way and the building will get managed as that single system much better.
So I think that the thing that we’re trying to do with this group, there’s three tranches to it. There’s business case, which I think we’ve just kind of touched on which is stop losing your data, stop wasting your data and make sure you can maximise the value out of it. We've got this tranche which is what is a digital record, which kind of needs to go into how do we use it and what do we need out of it. And then there is a third tranche which is interoperability and how you get that record to move better between systems because an API can obviously only really exchange fairly structured data.
And if we’ve got two different data structures from two different providers it’s very difficult to marry that together, and although everybody seems to think that AI can solve everything it’s not great at taking unstructured data and passing it between two systems that want structured data.
GEORGE Basically what we’re now doing is looking at, this is just a quick review that I’d like you to carry through of what you and Adina managed over the last two meetings. Essentially, what we’re doing is giving you the background as to how we’ve arrived at what you’ve just seen.
ADAM SANDERS I hope this is slightly stating the obvious here, but I think the key thing is that pretty much everything that we’re doing now is digital, we’re jut not extracting the value out of it. I think this is all fine, this is just about people having to work together and certainly I see it from the development or the construction, handover, asset facilities management, and then kind of the compliance and building safety are all working in a very siloed way. And then the key benefits out of it, we are efficient, although I think this requires some capital investment, not just money but time to make sure that this is set up for each organisation. But everything will become much more efficient and it becomes much more effective because everybody has got the information that they need to do the jobs. So your contractor is doing the jobs quicker, you’ve got better visibility of what you’re managing, compliance and building safety becomes much more straightforward and decarbonisation, I don’t know…
GEORGE Just on that decarbonisation front, I’d like to loop it back to what Andrew was saying earlier and that’s having a good reliable baseline of information. Because if you’ve got your baseline of information, which may have been collected for maybe snagging purposes or it may have been collected for compliance purposes, but once you’ve actually recorded what you’ve got in the building, what it is and where it is, then you’ve got the opportunity to understand more what its embodied carbon is and what it’s operational carbon is. Without knowing what’s in the buildings and how they’re being used then you can’t actually come up with anything reliable, it’s just guess work.
ADAM SANDERS I will keep talking, I’m very focused on housing so please feel free to drag me back into other sectors, but I think most of what I’m saying probably applies to the other sectors anyway.
RICHARD I want to raise something that Ishka has put in the chat. How do you define the difference between digital and electronic? And does it matter?
ISHKA HEART I suppose when you’re saying digital do you mean we’ve got records or document records on the cloud now? Or do you mean we’ve got all of our records in databases that are being managed? So again, back to my point, I suppose that the difference between having documents, which obviously contain data and information, but then also we have systems wit data fields, individual data fields being updated.
ADAM SANDERS Absolutely, I should have dealt with that first slide better. Sometimes you’ve got data, so it could be JSON data or XML or something like that, where you’ve got a record, or its sat in a database and it’s queried, it’s somehow structured. Sometimes you’ve got that, but you’ve extracted it from a document and it’s really helpful to go I’ve got this PDF which I’ve been given because that’s an output from some activity. Whether it’s an EICR or a boiler service or a door survey and I’ve got the data that sits alongside it. And sometimes I’ve got a document and it was just given to me as a PDF because it was a Word document that got generated as a PDF and I’ve extracted certain metadata.
And I think we’ve got to be really flexible with how we handle all of these digital records because sometimes you’re going to get that perfect balance which is the nice PDF document and all of the data that builds it and sometimes you’re not. You’re just going to have that PDF and you’re going to have to manually extract some information that says it was done by this company on this date and it’s this kind of a document and if you want to get into it you have to read it. We’d all love everything to be perfect structured data, but the systems that we use have got to be able to be flexible to handle both. And if you get two surveys done and somebody’s put make and model in one field and then somebody has got manufacturer and model number and serial number as separate fields, you’ve got to have nice ways to handle that unstructured data.
GEORGE The people that have been involved in the BIM4Housing journey, people like Jack White, right from the start will know that I’ve bored people to death with my definition of saying that we need information to be machine-readable. So, exactly what you’re saying there, Ishka. The only thing is though, the more that I’ve gone into it over the last few years the more I’ve realised that that particular aspiration is still something we should be pushing heavily for, but there’s still going to be a lot of stuff in documents. So, what we’ve been trying to do as our Active Plan business is make ti so that we can create a level of intelligence in the existing documents. And you might have a 70-page manual of which there are three different pages in it that are relevant to a particular purpose.
Now, if you can put tags in that and hold metadata against those documents then at least you’re turning it into something, it is to some extent machine-readable because it takes you to the information that you can then manually read. So, I think that’s the journey that Adam is just talking about, we need to recognise that we can’t have the ultimate goal immediately but I think we should be moving towards it.
TERENCE HORSMAN I work at ORCA Service Technologies, we’re a software development company and we primarily focus on field service software and asset management. And my background is in facilities management. I’ve been involved with ISO/TC 267, co-wrote the standard 41016 on technology in FM and helped with that working group. And throughout the discussion it’s been really interesting, one of the things that comes to mind is standardisation. So, machine-readable data is I think one thing, and getting it XML or whatever format, we agree is great. But then what about the filed mapping?
And I think having some form of standardisation would significantly help in turning that machine-readable data into something that’s consistent and interoperable which could then also help to allow to bridge that gap with, for example, static PDF documents (as Andrew mentioned in the chat, Co-pilot is really good at extracting texts from PDF). I think those types of services are going to become a lot more affordable and a lot more attainable and a lot of software vendors will incorporate those into the solutions. So if we have some form of standardised formatting which we can put that information, it might allow us to get more improved analytics over time.
ADAM SANDERS We’ve talked about these data standards quite a few times, and obviously there is IFC and COBie and bits and pieces and HACT have done a lot around this as well. The problem is they come up with a…I hate people who go ‘the problem is’ and they never come up with a solution…but if you’ve got these rigid filed types, do they get updated? And then what happens when something needs to be extended? It’s a problem that we have all of the time because we’ve got multiple contractors…a client will have multiple contractors, they change them every few years and each contractor wants to do things slightly different. So we have a whole load of mobile tools that do that data collection.
And we’ve just got to the point where we say OK, look, we’ve got some standard field types and you can map your fields to those field types, and I’m kind of hoping that’s what the interoperability group come up with because we’ve got to be really flexible with stuff. I’m really scared of, not scared but, of the AI thing trying to create structured data out of documents because some of the mess that I’ve seen it create is absolutely astonishing. If you compare two documents and it doesn’t know the nuance of what its comparing, and it’s great if you’ve got somebody with a brain that’s looking at it and some AI goes, oh, I think it’s this, and they can check and go yeah, it is that, you’ve just saved me a whole load of retyping, that’s when it’s brilliant.
The problem is you give that to an organisation, especially a large one like a housing association, especially a large one like a housing association, and they’ll try and push that off to a £20,000 a year temp who, not having anything against them as you’ve got to start somewhere, will overstep the mark. And I think that’s where the software needs to be really clever.
GEORGE That’s exactly why we set up the interoperability group. So, just to provide some context, Adam just mentioned HACT (Housing Association Charitable Trust) and it’s a organisation that’s funded by the large housing association, housing providers. And they’ve come up with a series of standard taxonomies, standard structures etc. They’ve looked at some of the BIM standards, for example, which is the buildingSMART standards, to go through those. I was a member and have contributed to the original HACT standards for the asset information requirements for housing associations. And one of the challenges there is that when we were then going back to the subject matter experts, for example, something like a fire door, what information do you need to know about a fire door?
When we asked that question of subject matter experts they couldn’t explain that to us. And the reason was they didn’t have a context in which to do it. The IFC, for example, which is the BIM standard, is one way of looking at it, but if you’re a manufacturer you’re probably going to look at a completely different standard. Or if you’re a wholesaler or a merchant you’ll use another standard called ETIM. Now, most people in the BIM world have never heard of ETIM and yet it’s what the procurement people are going to use. So the reason that we’ve got the interoperability group is to try and rationalise a lot of that because what we don’t want to do is get drawn into that when we’re trying to define what a digital record is.
RICHARD Exactly, so let’s go right back to that…go back to that slide. What, are we agreed with that definition of a record and a digital record? Does anybody disagree with that definition? Ok, I’ll count that as nobody disagrees. How, for digital machine-readable data, APIs, PDF and metadata, 360 photos and assets, do we all understand what all of those things are? I think possibly we don’t.
GEORGE I’m saying machine-readable data in the same way that Ishka’s talking about and that is that it’s not something, you might have it as a PDF but then if a human has got to read it and interpret it, then I would say that isn’t machine-readable data. I think it needs to be in a form that can be, maybe some of the data that’s in it. For example, a fire door’s rating needs to be addressable in some way through that process, either by the section in the document being identified, or it needs a data tag against it. Now, as far as the PDF with metadata and classification is concerned, what we’re pushing for is using Uniclass which is wildly underused. So Uniclass is a way of doing that because what you can also do is use Uniclass to say what the content is of a document and therefore it’s not just about products and systems, but we’re applying Uniclass for content…
RICHARD Do we need to agree that at this meeting, George? Do we need to say do we agree with you on Uniclass? Does anybody disagree with what George just said about Uniclass?
JACK WHITE Yes. I don’t want to lose sight of we want things to work across the sectors but we need things to work internally within our organisations and if we focus on just trying to get it consistent across the sector, it’s enough of an effort to try and get culture change within the organisation and to take this onboard if everyone has to learn a new language, a new everything, it makes it doubly hard. And so I would suggest we take the frameworks around Uniclass etc and apply that way of thinking, but not necessarily apply it verbatim.
GEORGE Yeah, what I’m thinking, databases allow you to hold multiple data sets against an asset or a system or a space or whatever. So, what I’m suggesting is that Uniclass is a comprehensive one that is used by the industry.
JACK WHITE Yeah, I don’t think it is, I think for a start it kind of confuses form and function and when you look at what it’s got in terms of, it doesn’t say ‘door’ and then give attributes and give you a type of door. It will say ‘steel door, timber door’, not that you’re going to get a door that’s half steel and half-timber, but in that situation what would you do? So, I don’t really like the way it’s structured, in all honesty, and I’d prefer to take it just on a functional level on that form set by attributes.
GEORGE Have you got an alternative as a classification system?
JACK WHITE We’ve come up with our own classification system.
ADAM SANDERS I think that’s really interesting and, I think, that’s what I was talking about a second ago, correct me if I’m wrong. The problem is when everybody comes up with their own classification systems it’s difficult to move that data between systems. But if you can pick certain, I would call them fields because I don’t really understand Uniclass, but if you make certain fields and map them and yo go, well, these two are the same, these are different, but they should roughly map and have structured data that is flexible enough so that when you when you port it if you’ve got red, green, blue in your list but somebody has bot mauve and teal and purple, then fine, it extends it and you might get some duplication and you’ve just got a little bit of ongoing data cleansing. But as long as the data isn’t wrong I don’t think it’s a problem, I think it just makes reporting more difficult.
ISHKA HEART I just wanted to clarify, you’ve got obviously classification next to PDFs here, so are we talking now about classifying documents and applying metadata to documents and PDFs, or have we moved into the space of using Uniclass to classify asset types and products?
GEORGE No, what I’m saying is, I think it’s largely accepted in the industry that as far as design and construction is concerned that Uniclass is probably the main standard that everybody is working to. It's not perfect, I absolutely agree Jack, but there are 7 or 8 tables and the problem is that people don’t understand, in some cases, that it’s not just about products and systems. There are a whole range of different other tables that you can use to classify things and you don’t need to use Uniclass to define, for example, what properties you’ve got against things, that’s a different mechanism. So I think having something that is standardised, it doesn’t mean that you’re tied to that because, as you know, you can hold multiple classification references against the same item. It’s just having something that is consistent.
So, picking up on what you’re saying, Ishka, I think that what we’re doing we are using Uniclass to provide some context for documents in terms of what they contain. Because within Uniclass there’s information types, for example, and there’s also Uniclass now for risks. All fo these things, they overlap each other depending on what the purpose is of what people are looking for. But should we leave that to the interoperability group?
RICHARD Have we agree our definition that we need for this group? That’s for how. George, you brought up Uniclass.
GEORGE Yeah, maybe I went into a rabbit hole there.
RICHARD Right, we’ll ignore Uniclass for that part of the discussion, although it’s interesting it’s not as pertinent to us as it is to interoperability.
GEORGE But having a classification, I think that’s the point, we need to have some classification.
RICHARD OK, so we can agree on the how.
GEORGE Yeah, and the other thing, everybody now is taking lots of 360 photos and people are also pulling together asset information as well. What we’ve been doing is making it simple for people to use the 360 photos and tag the asset data and use them to create asset information from the 360 photos. And I see that as being one way of turning a static thing, which is the 360 photo, into something that is dynamic.
ADAM SANDERS I think this is a really key point, this needs to be available, it needs to be portable and needs to be available to all stakeholders. There was a post on Linkedin the other day which George pointed out and infuriated both of us, somebody that said oh, the golden thread is just a collection of documents. And it winds me up when you’ve got people that are supposed to be really quite knowledgeable in this sector and are responsible for building safety and safety cases saying effectively stupid things like that. If we want this data to be valuable, and if you look at the government guidance it has to be portable. The government guidance says something like the golden thread has to be available to the people who need to use it where they need to use it, when they need to use it.
So that means if you’ve got a contractor and they are onsite they need to be able to get to that information onsite wherever they are. And if you’ve got somebody in an office who is managing the CAD or managing BIM models, then obviously they need to be able to modify that stuff. So that portability that needs to be available to all stakeholders is really important because you might have somebody that actually needs to look at the granular data of something, you might have somebody that needs to look at PDF that built that data, or vice versa. So it’s not, going back to your point Ishka, if you’ve got a document, you’ve got metadata, but you might also have the data that built that document behind it.
And they’re different things to different people because some people just need to know that document exists, some people need to read the document in the context which it was originally created and other people just need the data within it because they need it for other systems.
GEORGE I don’t think we need to go into this in detail. I think the point is what we’re trying to do, obviously the challenge that we’ve for with all of these meetings is that people will come along to one and then they can’t make the next one and they join the following one. But we need you to understand that we’re not just dreaming this up, this has actually come from the collective group. For example, I’m just going to show here, this is what Adina done. Adina is information manager for Durkan Construction, Durkan are delivering various different developments. She’s coming at it very much from a BIM perspective and looking at things as an information manager, whereas Adam was just going through there looking at it from a practical point of view from an asset management perspective.
So, what Adina has been looking at here is how does this fit in with the BIM standards and how we’re actually producing interoperable data. And then looking at that from a point of view of who is doing things, how are they being managed, so she’s thinking of things in this sort of way. So, it’s a result of that that we then came to determine, because remember this is a workstream, we’re trying to get this done so that people like jack and you have got something to basically build upon. So, the next bit is why do we need digital records, and that’s what we’d like to focus on for the rest of this session. So, maybe that’s something we can come back to Andrew Perry on that. Are you all finding this useful?
ADAM SANDERS I’m finding it quite interesting because I spend my life saying this stuff and I’ve got Jack, Ishka and Terence kind of also saying the same stuff. So, it’s really nice to hear that there is progression and I feel like that progression has happened in the last 6-12 months.
GEORGE Yeah, people are using it. I’ve been at two conferences this week, both of which have been revolving around safety cases. It does concern me a little bit that people have focused on the safety case report because it’s the deliverable that everybody has got to be providing at the moment. They’re not thinking about how the information needs to be created and used by different parties to create that safety case report. But it’s great that people are bothered about pulling together the information for that safety case and therefore the digital record is a fundamental part of the safety case. Because if you don’t have up-to-date and reliable digital records then your safety case is going to be invalid.
ADAM SANDERS This is also my other bugbear about safety cases and everybody just focusing on the report, rather than the management system that should produce the report. Get your management system right, you’ve got a safety case and then hopefully it reports your snapshot in time. I know nobody is quite there in automating that bit, but this all comes back to the digital records being in the right format, or the right structure and in the right systems and those systems talking to each other. The question is how do we achieve that? Maybe one of the things that we should do is get a few people to talk through what they’re doing and how they’re making it work.
GEORGE Yeah, so could we…Christopher Abrams, you’re from the prison service, aren’t you?
ADAM SANDERS Should we pick on Jack? If you don’t mind, because it would be great to hear what you guys are doing and how you’re making it work, where you’ve come from, maybe even some of the challenges. And what are your pain points, what are the hurdles that you’re hoping to overcome in the next 6-12 months?
JACK WHITE I’ve set up a small team doing this, the golden thread for existing buildings for a housing association about 3 1/2 years ago. And we’ve done a lot of work almost exclusively internally because we just found that we’d understand what we were doing, we’d understand the problems, we’d understand how to deliver it and we’d make sure we tailored it for ourselves. But there’s been a lot of, we’ve had 7 layers of hierarchy 8 layers or hierarchy, 7 layers of hierarchy 8 layers of hierarchy, it’s been coming up with systems going out onsite, testing data collection, testing how we do different things and really trying to find a way that made it work for us. Now we’ve just got to a stage where it works for us and we’re operationalising it, but it’s been a long time coming. I would probably just say it’s being really clear in terms of what we’re looking for, talk about information and setting out. When I’m thinking about it I’m thinking I want data for people to be able to understand our portfolio, our buildings, and our assets within them at a high level and then on a project level I want documents, so someone might want to understand what a specific fitting is on something.
Now, we could fill a whole database with all that, but it just makes it harder to find. So, it’s trying to have that flexibility of designing something that you don’t know how it would be used. So don’t think we can just design it and then implement it, it will be design it, implement it, realise our mistakes, improve it and kind of go like that. There’s been too many lessons from mistakes to cover in less than an hour, so just in a very brief way I’d probably say it’s trying to get that really going down into detail to understand it, to then be able to come back and look at the bigger picture and understand, a bit like what we’re covering here, exactly what it is that you’re trying to achieve.
GEORGE That’s an interesting point Jack, actually because you and Ishka and lots of other people, you’ve been on this journey now for several years and there’s things that you’ve learned along the way that would be very valuable to the rest of the team, if you’ve got some sort of summary of what to do and what you came across. What we’re doing with the business case group is we’re getting the teams to come up with examples of things where things go wrong that could have been avoided with better information. But we’re getting them to actually come up with some tangible examples.
A good one that one of the hospitals has come up with was they had some external works done and they dug through a mains water and took the mains water out completely to the hospital. And that’s because their record information was about the building, not the site, so therefore they didn’t know where anything was. And they were very, very close to taking out the electricity as well. And the irony of the electricity one is that it was the contractor that almost did the damage that had actually laid it in the first place.
ADAM SANDERS I have a good story about that, we had an old gas main and we were putting something into the house under the foundations and in the pit there was this old disused gas main. So I pulled out this electric reciprocating saw to cut the gas pipe out. What I didn’t realise was that the electricity main was just underneath it and it blew the blade clean in half.
GEORGE Ishka, have you got any experiences that you’d like to share?
ISHKA HEART On the development side we’re pushing to specify BIM models, COBie data and trying to provide exceptional records of what was built. But the difficulty is then passing it over to the rest of the business and how that’s managed in the FM system going forwards, that’s sort of been the fact that we work in silos, from development to asset management, it’s one of our biggest challenges. It’s interesting, Jack, hearing you speak from the asset perspective, a question to you: are you specifying to the development teams exactly what it is that you need from new builds? Or are you just focused on existing assets?
JACK WHITE We’ve specified a little bit, but I’ve left it fairly light touch because i think it’s actually will work out cheaper for us to do what we need to do to our buildings than to out it in information requirements. We’ve got our asset hierarchy for how we set everything out, to have to go and teach whatever organisation is delivering something that that’s how we want it, it’s much easier and quicker for us to just do it ourselves afterwards. So we’ve set out some requirements, but it’s very light touch.
GEORGE That’s interesting, Jack, what if they, let’s say, for example, you’ve got an electrical contractor or a mechanical contractor and they don’t provide you the information that you need to satisfy Regulation 38?
JACK WHITE That happens almost every time. As Ishka is saying there from the other side, it’s so disconnected, but I think in a lot of cases we don’t have the information. We ask for certain information but it’s not, the project manager and development, it’s not really their goal. They want to get something signed off, make their targets and they don’t really want things to get in the way.
GEORGE So who from within your organisation, because now it’s a legal duty…
JACK WHITE There are speed limits on the roads, George, but people speed every single day. Legal duty, it’s a bit like for safety case, it’s a requirement to produce a safety case so people focus on producing a safety case, not producing safe buildings or safely managing the buildings. That’s not the…I think we look across the sector and in this sphere a lot of it is producing safety cases, rather than that safety management system to make sure you’re managing building safety and then produce that. And I think we say it’s a legal requirement, it’s a legal this, legal that, people don’t really understand legislation, they don’t bother reading it.
RICHARD OK, we have to get back to the point of the meeting which Ishka has thoughtfully posted in the chat. What is the expected output of this workstream and in what timeline?
ADAM SANDERS I think broadly it’s what we’ve discussed today. I’m not sure there is a huge amount more resolution that we can add to this.
RICHARD I think in the pre-meet on Monday you had a very good line you put in.
ADAM SANDERS Whilst Richard finds my amazing line, I think it’s about trying to articulate what we’ve said. We talked about active and passive documents, it’s coming up with some terminology like that where we go we’ve got business case that says this is why it’s really important, it’s going to make you more efficient, it’s going to make you more effective. We’ve got a bit of a definition of what are these digital records, how are they structured, what counts, how can we define them and what do they do. And then we’ve got a third tranche which is saying how do we get all of that shit to talk to each other.
And I guess ultimately this is about awareness and if we’ve got a little bit of a working group and a few people that can contribute to that then we can help other organisations do it and find there’s some commercial interests here. I try and keep quite quiet about our commercial interests, but we want to be able to provide software and services that are relevant to other software providers, other housing providers, other FM providers. And I guess the group is about getting those people in the room and then they’ll be another stage to this which hopefully puts a bit more detail into it.
GEORGE Yeah, in answer to the point I think the first task of this was to actually determine what a digital record is and what it isn’t, and I think we’ve done that. The next step is to look at why we’re doing it, so some of the contributions already have come up. What would be very helpful is if you could give some practical reasons why you need a digital record, as some examples. I’ve thrown that one in there as the need for digital records in the safety case, but I’m sure that there’s others that we could do as well. And then finally, how do we then keep them up-to-date. So, I think we’ll perhaps run another session on the why, if people are interested in learning from each other as to how they’re using digital records and then why that is needed. And we may then follow that up with some feedback from the interoperability group looking at the detailed how as to what that’s been done.
RICHARD I think there’s something we did miss in this meeting. We were looking to confirm what a digital record was, but also George you brought something up about it being a valid digital record.
GEORGE Oh, definitely. In essence I guess, I think we’re saying the same thing in terms of…I think that’s covered in what is it.
RICHARD OK, Adam’s line was what does the digital record need to do.
GEORGE Oh yes.
RICHARD I think we know what we’re doing for the next meeting. So as George said, if anybody can put some examples in of reasons why it should be a digital record can you please do that.
PHIL THOMPSON Is all of this content that you’ve just been discussing on a Miro board?
ADAM SANDERS Its not really structured enough, I think we could get to that and I think that’s probably what we need to do.
PHIL THOMPSON It’s just that we had a call for the interoperability group in the week, I thought it was fantastic, and I’ve made some updates to the Miro board. I just think that if we aligned all of this stuff together it’s much more powerful.
ADAM SANDERS Maybe we should all be working on the same board. Let’s do that.
PHIL THOMPSON Hallelujah, put it in the same board in its own little box and then we can bring all of these connections together, that would be great.
RICHARD Are you volunteering to pull that Miro board together, Phil?
PHIL THOMPSON Well, we’ve already got the link, we sent it in the other group, it’s well populated. So I think just use the same link and just put a little border in the bit to do with this workshop.
RICHARD There was some issue with access on it, Jiss is in the process of sorting it out. When’s the next meeting?
ADAM SANDERS I think we need to get this board up, we need to have a couple of chats. 6 weeks, and we’ll make sure we’ve got some real goals to find and some outputs that we need to achieve.
CHAT
Andrew Perry (External)11:19 am
Is the value of the data - up to date and assured - actually realised?
Ishka Heart (Unverified)11:20 am
have you defined the difference between digital and electronic? but also, does it really matter? I feel like people sometimes use it interchangeably.
Andrew Perry (External)11:22 am
If PAS 2080 is to be successful the history of how the facility has / is operating has to be available to provide the data to support / enable the "options" case
Andrew Perry (External)11:22 am
Co-Pilot is good for extracting text from a pdf
for Govt FM contracts they are driving suppliers to collate / manage asset data in the FMS 02 format to allow x-ref to CIBSE M / SFG20 / NRM3 /Uniclass 2015 within an API / CDE
Ishka Heart (Unverified)11:39 am
Jack - is your bespoke classification for your asset types, documents or both?
Jack White (External)11:42 am
for our asset hierarchy, types of assets, the attributes associated with the assets, the set of types of each attribute. that for assets. then for documents and for organisations and individuals to look at competence
Ishka Heart (Unverified)11:42 am
People also think that if we have a CDE we have a Golden thread, and that the CDE holds all our data- when really, it holds our documents, it is not our FM system
Ishka Heart (Unverified)11:55 am
What is the expected output of this workstream and in what timeline?
Andrew Perry (External)12:01 pm
Digital records allow asset performance / deterioration to be tracked and remedial works to be managed in a proactive way to MAINTAIN compliance
The digital record needs to enable the monitoring of key parameters to be measured / monitored / tracked - eg - the gap in a fire door - as it increases over time it will show - which doors are affected- the rate of change and allow correction to be actioned before the gap increases and goes beyond the maximum limit and FAILS - similar for emergency lighting