Round table: Getting the right data at the right time

As the data used to create models is increasingly exported into VR and AR headsets, or micro devices, the challenge becomes ever more prescient.

It was in the context of this need that experts from across the sector met at the offices of law firm Penningtons Manches for the annual BIM+ round table to discuss data – what we need, what we’re getting, whether the industry can pull off seamless transfer of data that’s both accurate and timely? Can the treasure trove of data be stored, filed and accessed in useful format? If not why not, and what could we be doing better?

Denise Chevin: How important is data to BIM and why has it become the new buzzword in government?

David Jellings: BIM really is simple – it’s access to information you need when you need it. That information is based upon data. Data is just numbers but once structured it becomes information and means something; then when it’s applied it becomes knowledge and after experience it becomes wisdom.

We have to start creating data – if it’s not there or configured we can’t share it, and if we can’t share it we can’t use it, and if it’s changed or corruptible or inaccurate then BIM will make things worse, not better. Unless we have data veracity then BIM Level 3 will never work and BIM Level 2 can’t work.

Alex Lubbock: Data is critical to us. We are there to support the government to deliver projects better, therefore not having a line of sight in terms of operation delivery of programmes is an issue. Getting better line of sight is about having structured data available at the right time. My role is not to nail the best BIM Level 2 contract ever, it is looking at systems and process re-engineering based on quality data to inform insight to provide best value for money.

You have to understand what you are asking for. In BIM language that would be: What do I need to know to drive my organisation? What do my decision-makers need to know about the business to make leading changes, rather than looking at lagging indicators of performance? Digital gives us that opportunity.

The round table participants

(l-r) Marek Suchocki, EMEA sales development executive, Autodesk; Edonis Jesus, BIM leader at Lendlease and chair of BIM4Heritage; Alistair Kell, principal at BDP and head of information technology and process; David Jellings, managing director, BIMObject; Francis Ho (seated), construction lawyer, Penningtons Manches; Eddie Tuttle, associate director, policy, research and public affairs, CIOB; Jon Harris, director of BIM, Mace; Riz Calder (seated), consultant, Rider Levett Bucknall; Alex Lubbock, head of digital construction at IPA, the government’s Infrastructure and Projects Authority; Richard Bates, associate and BIM manager, Alinea Consulting 

The public sector for social infrastructure buys the same thing many times in a bespoke way – therefore takes no efficiency in what it procures and this does not support the supply chain to align behind what we really want. We have recognised this and are working on a more efficient way to engage a fragmented supply chain that is challenged on margins and own operation models.

If we can move to a more standardised approach then it gives us a huge opportunity to help the industry move forward. By moving to a manufacturing approach by 2019 in conjunction with the Construction Sector Deal and the Industrial Strategy, there are mechanisms to collaborate with industry. It’s not new but now we have the digital tools and data make a market shift.

DC: Where are we now in terms of getting the right data then?

Jon Harris: Are we getting the right information and is it consistent? It depends where you sit in the project lifecycle but I will say no. It’s a mixed bag but as for digitally integrated working between different disciplines, that is still a big issue.

Plus, for a Tier 1 contractor, the time given to evaluate the tender information is often not enough, sometimes the information we get is not structured and fit for purpose or isn’t at the level of definition (LOD) indicated.

We have to go through a process of evaluating it, and are there some tools evolving to support this process? Kind of – but the practice of “design and dump of unstructured data” is still there.

Alistair Kell: Capturing and analysing data isn’t happening in any uniform way across any organisation yet but everything is still changing and in an early stage. We are talking as if it’s fully developed, as if all the information is standard and as if appointments are clearly understood with deliverables, but it’s still far too immature.

That’s not a negative view – BIM 2 is a stepping stone to improving everything and we are still on that journey. Everybody understands the benefit of working this way but it’s still maturing and there is a long way to go – but we are beginning to see benefits. Let’s keep the faith and have evangelists.

Edonis Jesus: “Key things like data capture and data management need explaining to clients.”

Marek Suchocki: You have to understand where you are as an organisation and get some understanding on how you can participate in the process. Shooting for the moon isn’t going to happen at all.

Where can you get benefits if you’re a client? If you’re a client and currently have no relationship to your asset management policy, ask some basics so at the end of project you can populate your asset management more effectively. You don’t suddenly want to be doing COBie [Construction Operations Building Information Exchange] data validation when no one understands you and you get no value.

It’s about good practice, not best practice, and moving forward in incremental steps together, leveraging standards. It starts with clients knowing what they need from the supply chain and specifying up front.

The really important part is realising we are all clients: if you’re representing an infrastructure client they might have an Employers Information Requirement (EIR) but that would go to Mace to deliver it, but they are also a client to their supply chain, so they must specify their needs.

It’s a cascade and if you break that at any point that’s when we don’t have the data. But if you want perfection no one will deliver that. Ask for simple stuff up front and keep it simple through the process and we will get somewhere. 

DC: Is putting data requirements in a contract the answer?

MS: As usual the dysfunctional procurement routes that pervade the industry and see the wrong parties only brought in when it’ s too late was flagged as a major problem in terms of generating the right data in formats that would be useful further down the line. Perhaps writing data requirements in the contract could help.

AK: We are talking as if BIM Level 2 is a thing in itself – it’s not. It’s many different things to many different people with many different tasks across whole life of building. But if I don’t know what you want to make of the data, I will work the way I work in a standard way and it won’t be immediately valuable. It’s that joined-up approach with the Asset Information Requirements and Employers Information Requirements.

But contractual aspects are holding us back: the industry doesn’t understand what it should be briefing for and how contracts should be shaped so that BIM is embedded not an appendix.

Francis Ho: There are different markets though. The major clients can take a world view of projects. The ultimate cost of the facility is in the lifecycle – operations and maintenance and they then have what outputs they need from construction from which they create what they need to get from BIM.

But most contracts aren’t put together that way. Two-stage tendering is perfect for contractors who get involved early in design, get involved in the BIM model and can bring their key suppliers into the BIM model. But how can you do that if you’re getting it for the first time with only six weeks to look at it? When you’re also looking at IRs, looking at contractual terms and pricing? There’s a huge amount to do.

Alistair Kell: “The generation that was born with technology is coming into our industry with a different mindset”

Contracts in an ideal world are put in place as soon as possible but that’s not how things happen. Contracts are done at a quite late stage when the design is already done. And how do we get mid-size developers to invest in this? It’s difficult. Do they really care about asset management? The public sector does, but not private.

DJ: Contracts have to support the process – standards have to address the problems. The way I look at the data conundrum is that we know what the end game should be, but there has to be a clear understanding in the industry of the benefit of doing that – it’s only the recognition of the importance and benefit of data that will drive the culture change.

DC: What are the best ways of delivering data for FM?

Edonis Jesus: Defining data needs a lot of work. We mostly use COBie data so that gives us good guidance in terms of data requirements but not focused on operations and maintenance.

BIM4Heritage is trying to define where is the data that will actually inform later the maintenance and repair, the health and safety – where is the key data for someone who is going to be maintaining this heritage building?

Riz Cader: When BIM was coming into play we looked and asked what information we need to run the building. You design and build it for five years but you operate it for 60 years. We quickly decided COBie wasn’t the right route for us and in 12 years I’ve never used it.

We went to CAFM [computer-aided facilities management] and used that data to fill in employer’s requirements so at completion the BIM file produced probably 80% of what was required. And so we always “start with the end in mind” by going to the client and asking what information they need to run the building.

Some clients have full CAFM system and some have an Access database, and so we tailor the information we require from the contractor and designer based on what the client wants. Asking for everything isn’t always of value for the client.

DC: Does better data mean better quality?

RC: With the issue of poor quality very much on construction’s agenda, the topic turned to the role digitisation should play – if only we could get the data right.

MS: If you can define what you need then you can check you got what you asked for. If I need a weatherproof membrane in a bathroom facilitated by tiles and I want it to be consistent across the estate – how you realise that doesn’t matter, as long as you have performance requirements you can check it.

David Jellings: “We know what the end game should be, but there has to be a clear understanding in the industry of the benefit of doing that.”

AL: If the data is accurate then increasingly that standard repeatable process will mean you get a better quality product. But we are talking about empirical analysis of stuff. You can talk about processes and improving quality of manufactured production but at some point there still has to be a response that is emotional.

Eddie Tuttle: If you’re buying a car you get a quality product at the end of the day, but is that the case buying a house? We have huge issues with the business model – we have a supply chain which is barely at BIM Level 1 and it’s really conceited to think we are further ahead than that. A client or government should be driving it further.

DC: Looking to “tomorrow’s data” – are we building the right foundations for the future?

FH: What do we do in the future? These assets are in use for decades so should we be thinking about capturing more data on the basis that in future we will have tools to identify it. Technology firms are capturing all the data they can. Are we limiting ourselves?

EJ: We need to think longer term. We have tech like AI capturing data, the smart building approach, even the smart city.

AL: The smartphone can tell me when the next bus passes by, if it’s on time, how long it would take me to walk or should I jump in a taxi. Google Maps understands traffic flow and it can predict traffic jams at different times of day – that’s all going to be in this space. That’s connected data and valuable. And buildings will move towards that inevitably.

Richard Bates: The idea of collecting more stuff for what we may or may not be able to do with it in future is important: if we collect it now we can figure out later what to do with it. But if we haven’t got that consistency and that data to start with how we know we can trust it, we don’t know it will ever be useful. Or maybe computers can figure it out in future?

EJ: These issues come down to poor education and the need to explain the bigger picture to people. For example, the aim of the government implementing digital technologies is to improve productivity in construction so the bottom line is increased by 1% then wages go up – it’s actually explaining the policy in context of industry and economy.

Most of my time is spent trying to explain to clients the bigger picture of the benefits of BIM implementation for them. Key things like data capture and data management need explaining to them. If you start collecting data today, in 10 years’ time you won’t have to pay for it again.

A survey you pay for today you will need again in five years’ time if you haven’t recorded that info somewhere – try to make them think in longer term. It might even be 100 years’ timeframe – capital expenditure is 50% and operational is 80 so how to use that data to create efficiencies.

Not just clients need this education – most of the industry needs to understand the implications. Then they can see it will help them exchange with others and improve their own performance and contribute to other people’s performance.

AK: The generation that understands and was born with technology is coming into our industry with a different mindset and provided we can keep them positive and not go into contractual niceties then actually there’s something potentially quite exciting on the horizon.

AL: We must stop focusing on a narrow pool of talent and start branding construction as the exciting industry it is – the talent is out there.