AI and its place in the future of BIM

Craig Johnson, Business Development Manager - Steel - Structures Division at Trimble Solutions (UK) explores the topic of Artificial Intelligence and imagines its future place within the world of BIM and construction

Having experienced the benefits that digital ways of working can offer, the value it can deliver and the challenges it can help solve, the construction industry is perhaps more open to the adoption of new technology than ever before. The question now is what is next for the industry? How can it push this progression even further?

All technology, of any kind, is in a state of constant development, as we strive to increase its value and ask ourselves 'what's next?'. BIM is no different, with the industry looking for ways in which it can further improve the efficiency and productivity benefits the technology offers to our detailing, engineering, fabrication and construction workflows. Perhaps the most recent development in this case is the idea of parametric design, or data driven design as it is also known, with an increasing number of detailers and engineers already adopting this innovative way of working.

Through the use of parametric design tools and modelling software, parametric design enables detailers to input their required rules, parameters and design algorithm, with the relative output then generated by the computer. Pushing this technology further is the idea of computer-driven design. Here, users can input the basic required parameters and allow the computer software to automatically generate various different design iterations at speed. As well as saving valuable time and facilitating complex structural geometry, this process can also help to determine and identify the most optimum and efficient design solution.

In many ways, parametric and computer-driven design is a starting point for a new way of working, one where the software and technology is provided with more power but the user is still ultimately in control of the inputs and outputs. Now, with more people adopting parametric design within their BIM workflows, what is next?

While cloud-based software, such as Trimble Connect, is not necessarily new itself, it continues to be a great and effective way of enabling a connected workflow, facilitating collaboration and ensuring an open route of communication between project teams - three things that have proved especially valuable throughout the last twelve months. Essentially a huge data storage facility, a project's BIM model, along with all of its associated drawings, schedules and documentation, can be stored in the cloud, easily accessible for teams to review and individually work on.

However, what happens to this data once a project has been completed? Often, the majority of it will remain in the cloud, unused and unutilised by its owner. Yet, the rise of Machine Learning and Artificial Intelligence (AI) could change this.

In simple terms, AI is a form of Machine Learning, where existing information and data is used to develop its own intelligence system; to learn and think in a way similar to humans and provide its own solutions. Typically, the more data a machine is exposed to, the better it will become at detecting and internalising patterns in said data and, in turn, understanding and providing insights.

Within the context of the construction and BIM industry, AI has the potential to successfully harness and utilise the significant amount of past project data that is presently unused; helping to further improve and enhance our productivity and efficiency levels as a result.

While every building and structure is itself unique, detailing and modelling tasks can often be repetitive by way of nature and/or design. For example, various concrete panels, steel beams and columns and their various connections are often all found within a construction design project. It is these very data similarities where the potential for automation arises - something that parametric design has already given us a glimpse of - with a company able to utilise its experience and known good design choices from past projects to help automate, design and optimise the new.

Putting this into practice, consider the task of detailing a complex steel connection. Through AI and Machine Learning, it is possible that BIM software (in the future) may be able to detect patterns and similarities between a user's new model and their previously completed designs, automatically suggesting and recommending solutions based on past projects and the library of data. In this case, the optimum steel connection could feature fewer welds, fewer bolts or even less steel, saving money and materials, as well as being quicker and easier to fabricate and assemble.

It is evident that such automated technology could deliver very real time-savings to a project, both in terms of the initial detailing work and in improved accuracy. However, it could also contribute towards helping project teams achieve a more optimum and efficient design. Imagine if AI technology had the capacity to look at past designs and categorise what worked well and what didn't; taking this existing data and using it to not only design the new, but improve it.

Pushing this idea further still, what if collaborative platforms could then feed fabricator and construction information, such as costs and time, into this? The result could be a new age of BIM designs that are driven, not only by design, but fabrication and construction too. What was easy to fabricate? What was easy to install? What was most cost-effective? What was most successful?

Ultimately, however, both the success of AI in these complex environments and how much we, as an industry, are able to get out of such technology depends greatly on acceptance. It relies on a sense of trust - trust and confidence in the solutions that this automated and machine-learned software suggests. Only then can we reap the rewards of our technological advancements.