The future of BIM is not BIM

The move from paper, pen, transparent paper, and razer blade to Computer-Aided-Design (CAD) and later to Building Information Modeling (BIM) all happened for most offices in the last 20-30 years. What a change, it's still ongoing and not over. We just started. In this article, I explore ideas and concepts to move on. They are based on the following assumptions:

• Designing and building need to increase speed, get cheaper, and more sustainable. Traditional project management would say that's not possible. I believe that's also true, unless we change the underlying conditions and approach the task(s) differently. More sustainability is the most important, but it will not become feasible unless we manage speed and cost. Therefore we need to look at it together.
• During design and construction, we create value for the client twice - First, when creating good and fitting concepts and, secondly, when assembling the building. Everything else is necessary in the best case and, at worst just waste. Yes, with this statement, I say that most planners currently earn money with something that is not in the core interest of the client.
• Most people in the industry are overwhelmed by using the CAD and BIM tools. Think how difficult it (is) was to enforce a layering system in the office. And now, we are talking about a more dimensional matrix of elements, attributes, their values, and classification systems. 

I often hear the argument that before, we managed to build as well, so we don't need CAD/BIM/to change. This is a mental simplification. Some people did work very well - most did just average. But as a company owner, you need to find the best way for most people, not only the star employees! It's about consistent, repeatable performance. Not the one-time lucky shoot or the success bought with blood, conflicts, and burnouts. Therefore in this article, I'm exploring the properties of the necessary next-generation BIM or CAD to support value creation in the concept and in the detail design phase.

The (not) BIM for work preparation & construction

Late changes and waiting for decisions or missing information is a critical factor for losing money on construction sites. Therefore work preparation needs to be focused on:

Providing all the necessary information to work uninterrupted on-site (or in the factory for pre-assembly). The planning is serving the site!

Often I hear from planners that it's too expensive to plan every detail. It's precisely the opposite. It's too costly not to plan everything. Especially when looking at the percentage split between planning and construction. 20% of the budget is spent on planning and 80% on building. So, where do you want to save 1%?

For this detailed planning, the static view of providing a plan/picture of what the final result should look like is not enough anymore. We need to design the production/assembling and disassembly process.

The best way of doing this - is to do it. Therefore planning means building the building first digitally. Not only - as we do it now - showing the final result in 3D or providing a high-level 2D abstraction. Future tools should offer a virtual environment to plan the assembling holistically. And I'm not talking about a 4D simulation - an animated schedule here. We need a system to connect efficiently:

• Resources, e.g. formwork, steel, and concrete.
• Tasks, e.g., Delivering the scaffolding, putting it, laying the steel, bending the steel, pouring the concrete
• Different interchangeable methods to do the tasks. E.g., A concrete slab could be preconstructed or made on site. The necessary tasks would vary. And the optimum way would depend on the project and the overall availability of the company's resources. These resoureces would need to be dynamically taken into account. 

And all this is integrated with the next-level BIM system. So I imagine a system where you drag and drop the different resources and apply the tasks to them. Drag the steel on the scaffold - it will distribute according to the load requirements of the static model - the speed will be determined by the amount of worker and crane resources. It could be gamification of work - best done by the people who will do the job - or at least with their involvement.

Sketch hierarchical. 

The three-dimensional geometry would be just a byproduct of planning the work. The next level is leaving the project view and connecting the different projects.

To optimize the processes when the algorithm picks the optimal methods based on all the companies' planned activities at a particular time. E.g., the company has some scaffolding, but to cover peaks, it rents more. What's the optimal mix, taking into account some unforeseeable circumstances? A task there learning from the past project's data can help.

This approach is closer to the kind of parametric modeling known in the product design industry than the modeling we currently do to produce plans.

Quite a challenge and a complicated system - but imaginable, and it does not need to do everything from day one. Starting by combining resources and tasks would be the first step when connecting all the different projects across the company, and all the other involved companies to optimize them could come later.

Looking at planning in this way, let's answer the question of who should also do the planning for the site. Should an external planner do this planning? Is the external company even able to do this, or is the control over the preconstruction planning a core process for every construction company?

I believe the latter. It's the only way to differentiate from other companies - salaries and material prices are very comparable between the different companies - only their process intelligence differentiates them. Every company that does not control planning for the site depends on the quality of the planner's plans and gambles with success. The solution is to either build up a department internally or subcontract to a closely aligned partner that knows your needs very well. The catch is the common tender practice of giving a contract just a week before construction starts do not work anymore! The company needs to be involved much earlier. We already see a trend in wood construction, but that is also necessary for other trades.

The (not) BIM for conceptual design

I started the last paragraph with the sentence: 

"late design stages and waiting for decisions and information is the root cause of waste on-site ."

Imagine a problem occurs on-site. People stop working and start discussing. After some time, a call to the back office/planner. More people get involved, and more money is wasted. It's not uncommon that several days pass by between finding a problem and the solution. The site needs to juggle resources and tasks. And most of the time, it only costs the construction company - not the involved planners. Therefore there is little incentive for planets to do a better job!

Even the most flexible production system will get overwhelmed when the overall goal is not stable. Therefore the avoidance of this already starts in the conceptual phase! And the best method to do this is communication with the client. Yes, listening, asking questions, and running experiments with the client to better understand the client's wishes. Once we know them, we must translate them into requirements and solutions. And this is an iterative process. Most clients need to see something to get started and understand their needs. The tools we use should support us in doing so. Especially as we are seldom dealing with only one client but with a complex system of different stakeholders.

We need a convenient way of capturing the requirements to manage this process. Some database with visualization capabilities helps to document them and automatically check the new solutions against the formulated requirements. As discrepancies will occur and when we either need to adapt the solution or refine the requirement. It's an exploration and not a waterfall approach.

We need a solution that enables fast decision-making. Therefore with minimal input, we need to provide as much information as possible about the consequences of the decision. 

I imagine a system where you directly design the user's space, not like now, where you draw walls enclosing the space. A delicate but distinct differentiation. 

It's about minimizing the amount of interaction with the computer, like clicks. I imagine a VR collaborative environment with some automated avatars that simulate the behavior of the users and can give direct feedback if something is not working. Here again, we could work with predefined tasks - going to the reception, using the bathroom, running down the stairs in case of fire.

I think the key is directly working in and at the space. The elements between are just a consequence of the use. We need the spaces and their connections (windows and doors). And as we prove in our work at abstract, everything else can be automatically generated. 

It won't be enough to have this eye-level immersed view of the project. We will need a bird's eye view as well.

• Zooming out and seeing the overall organization of the building, switching between different themed views. E.g., showing where the different uses are and how they are connected. Showing different vibration requirements in a laboratory building and so on.
• Looking at the performance of the building over one year, ten years in the context of climate change and maybe in the context I different surrounding buildings. 
• Looking at different costs, construction, operations, and disassembling costs. Comfort, environmental impact, and so one.

All this is already necessary for an early design stage to make the right decisions at a time when the cost of change is minimal. Unfortunately, the reality is that these decisions can only be taken late in the design process because that's when the information becomes available.

So the lengthy planning cycles with making decisions based on guesswork or "experience," or better-called tradition, need to be much faster and replaced by provable simulations in the conceptual stage. 

On a side note, one thing that changed is that we don't trust authority figures anymore. Before going to the doctor, we google and read about different diagnoses. And the same is true for clients. People won't believe it because the architect said this is a good solution.

What needs to change to enable change

First, we need to leave behind the additive working method of planners. Instead, in analogy to "preconstruction," we need to "predesign" and work with ready-made components we adapt to the project - no more of the common postponing of decisions to later stages. So instead of delaying the decision, e.g., which floor material to use, we need to work with assumptions and document them. E.g., There is a very high probability that there will be floor material. Therefore we can already assume it early on and change it later. This means we confirm the assumption appropriately and document this again. 

Second, we must leave behind the idea that planners are better than construction people and that they must tell them what to do. It's not them against us but us. It's - and - we find the best solution together. This change influences the contractual side as well. Design assist, full-service contracts, and IPD contracts will become more common. Therefore, we need to change how we procure design and construction services. Once the concept stabilizes, the construction companies need to be on board to optimize design and constructability. Connected to this is that companies need to align their business model closer to the customer value. This probably means a higher degree of specialization and communication competency. The fountainhead is dead.

Third, we must leave the idea behind that one software tool can solve all these different needs. There will be many various tools, that need an open API to exchange information seamlessly. To program a system that fits all needs is expensive and too complicated. Moreover, the risk for the companies is high. Better to work with smaller interchangeable parts instead of one monolithic chunk of software. 

What are the first steps?

A vision is excellent but worthless with some concrete steps we can take right away. The first one you can take is to think about your business model and how it aligns with the customers' needs - and how you can serve the customer better.

We are working on our second product, the abstrctBUILD marketplace. This is a convenient way of capturing existing design and building systems and automatically applying them to new designs. Therefore having a better foundation for decisions right away, just based on preliminary design models, you can use these ready-made templates - a way of predesigning. To build up this marketplace with "predesigns," we work together with YOU.

There will be three different ways you can use them.

• To optimize your internal processes.
• To build up another revenue stream by selling the design and building system.
• To build up another channel for your building services.

And of course, you can combine these different ways. Get in touch with us, and we will find the best way to get you started.