Walk the Line: Ross Kilshaw from WSP

We caught up with Ross Kilshaw, Associate Design and Data Strategist at WSP to talk about the standardisation of data, customer choice and why working backwards can be game-changing for moving forward.

Ross Kilshaw

Associate Design and Data Strategist
WSP

📢 Listen below to also find out what the industry can learn from Lego! 👇

Could you tell us a bit about yourself and your role at WSP?

“Having moved from Urban Splash, I joined WSP as an Associate Design and Data Strategist within the digital innovation team.

This is a new team at WSP, we specialise in DfMA and offsite manufacture delivery. My role looks at the strategic approach to design and data in offsite manufacture, applying DfMA principles to improve design, project workflow and delivery.

“It’s no secret that data is one of the most valuable commodities in the world right now, but I think the construction industry is one of the last to really make use of it.”

One of the big problems I find is the standardisation of data. When it's left to employees to interpret how they should input data you don't get any standardisation, so it's a lot harder to collect and analyse. My work is around trying to use tools and techniques to automate data input, and from there get some real analysis back.

There are so many different types of data, one of the challenges can be sorting each type from one another. For example, if you're talking about procurement documents, I always think you've got three types: design drawings, specifications, and schedules. From there, you’ve got to understand, where does that data best fit? In this example, the dimensions clearly sit better on the drawings, and then the quantities sit better on the schedule. Then, you can start using current software, Revit or Inventor for example, to start automating the production schedule.

Gerald Desmond Bridge Replacement, Long Beach, USA

How did you end up in offsite construction?

Coming out of university I specialised in computational design, and through my Master’s I got more into self-taught software engineering. I'll be the first to admit I'm no expert, but the thing that excites me is understanding the basics of software engineering. You start to understand its potential.

From there, I went into AECOM and was in the specification team, which sits in the architecture design department. It was all about breaking a building down into separate systems and understanding how those systems come together. When you think about a building as systems, that's basically what an offsite manufacturer is: you've got the wall panels, the roof, the floor cassette. I started getting more and more interested into how offsite manufacturing can be optimised for the site, and for the employees' requirements.

I joined Urban Splash as the Business Intelligence and Improvement Manager at the factory in Alfreton. That was my first experience at an offsite manufacturer. We were a small design team, so I got into the nitty gritty of what happens within the factory.

“I got used to walking the floor to spot problems and then walking back into the design room to start solving them.”

Tell me about some of your successes in the last 12 months?

The success for me is my honed ability to critically analyse what has gone wrong in any given situation and then take those improvements forward into the future. At WSP, I’m working on some big projects focussing on permutations and customer choice, which have been interesting and key for the business. One thing I always start presentations with is the question of: “How long do you spend choosing something to watch on TV at night?” Most of us will sit for a while being unable to choose before giving up, and that’s the idea of customer anxiety. If choosing a TV show is stressful, then consider the house-buying process.

It is so easy to get caught up in producing lots of choice, but really, we should consider what the optimum number of choices is for our customers. We’ve been working on beginning with a customer’s requirements, then working backwards to develop a successful product.

What are some of the challenges around focussing on customer choice?

If you increase permutations, you reduce the standardisation. If you reduce standardisation, you're not getting the benefit of economies of scale, and you risk losing muscle memory because of the lack of repetition in movements across the factory floor. Within that same vein, the lack of repetition and mastering of a particular component can lead to quality issues.

I think you must start with production rates when trying to analyse your bottlenecks. The lines are only as fast as your slowest station. So, even if you've got one station that works really quickly because there is quite a lot of standardisation, the whole floor will be slowed down by another station if there is so much variation that it’s causing issues. If you want to increase permutations, you’d have to start looking at the whole sub-assembly ecosystem. You would need to manufacture sub-assemblies in separate factories and then have them all transported to an assembly factory, where you’d have a just-in-time methodology. The more manufacturers you include in this ecosystem, the quicker the efficiencies get because they're seeing the benefits of economies of scale and repetition. Whereas, if you try to do all of that under one roof, there is almost always too much going on.

I’m seeing a lot of clients coming into WSP and saying that they can't get architects to design offsite manufacturing to be profitable. And I think that is the key. Obviously, you get the benefits of sustainability, but at the end of the day, it has to be profitable. I think one of the problems is that we're trying to adapt the traditional design processes to suit offsite manufacturers when in actual fact, we need to go back to the drawing board and say, "Well, how should a design team be set up to optimise this project for offsite manufacture?"

I always want to know why people from production, or from the factories, aren’t being involved with the design team. If you're starting with the customer and working backwards, the factory is going to see that product a lot later in the design process than the concept architect will.

One of the things we're looking at now is collaborating with offsite manufacturers to create a kit of parts. If you already know that the manufacturers can build the kit of parts, you don't end up with abortive work or heavy re-designs because the manufacturing team gets the concept design, they analyse, and most importantly, they improve it. You can’t just choose to make small changes to one part during the end stages of the process, because it has a huge knock-on effect on the rest of the floor.

What do you think the opportunities are for technology in offsite at the minute?

I think the opportunity for technology is incredible. Although it’s important to have the ability to draw with pen and paper - because it’s a quick way to sketch ideas - a lot of architecture is based on rules and regulations, and rules can be coded and automated. To codify the rules, you need standardisation of data, you need the program to know the size of a door and what it requires. Whereas at the moment, each person will call a door a different thing, making it impossible to automate.

Uniclass have done a good job in categorising unique identifiers to systems and spaces within a building. There are spaces and locations, systems, and then components. When I'm designing a building, I’m subconsciously going through a specific list of questions in my head, all which relate to spaces, systems and components.

For example, within the team, we're talking about how many decisions need to be made to design a staircase – we've got to 150 decisions. Then, if you start breaking it down, considering what the employer actually wants to have control over, they might not care about what's going on behind that wall lining, so you’re down to maybe 80. They might just want something cheap, you’re down to 30 etc. You almost want a decision engine, that branches into the right questions, depending on the first few answers given. That is how I got into my first experience with software engineering, it's understanding, what do you need to be asking to solve that problem and how can you do it in a way that increases customer engagement.

Is there anything I haven't asked you about, that you think I should have? Or anything you wanted to speak about today?

One other thing which I think is really exciting is the idea of sub-assembly ecosystems. For example, the car industry is extraordinarily successful because the factories are successful. They could have a factory in Hull that is designing one small component before it is transported to another factory in Germany to be assembled. What we see a lot of in offsite manufacturing is everything being done under one roof.

“But, if one company has got a wall-making machine, what is stopping them from running it 24/7 and servicing other factories who don’t have one? ”

There are some very interesting conversations going on at the moment with a couple of offsite manufacturers to discuss how they could potentially profit further from their advancements in technology. They could effectively produce enough wall panels for themselves, but then go further to produce wall panels for multiple other companies working on multiple other projects. From there it is a question of using the systems available to manage that kind of dynamic scheduling, which is really exciting. I think it’s then good to consider that if you do have this sub-assembly ecosystem, what's stopping you having a flying factory on the site itself, to put everything together? With that concept, you use systems to begin reducing travelling distances, time and money.

---

And to end, who are you passing the mic to next?

“I’m passing the mic to Dale Sinclair here at WSP. Dale’s got some fascinating ideas about the ‘big picture’ of how we should design and make buildings and cities in the future. He was a co-author of the DfMA Overlay to the RIBA Plan of Work and has extensive experience within the construction industry. He’ll be an interesting guest!”

🎤