AI in architecture: Examples

One might argue that architecture is an art, which is true for the most part. However, designing a building takes more than a creative mind – it also involves a lot of calculations, and nothing calculates better and faster than a computer, especially one powered by AI.

Artificial intelligence has the potential to revolutionize architecture and construction in various areas, including the design process, efficient usage of space, and providing solutions to overcome safety and cost concerns.

Here are several use-cases of AI in architecture:

Generating design

Artificial intelligence technologies like Machine Learning have a great influence on modern building designs. Machine learning works much like the human mind by learning from existing data and making incremental improvements on given tasks with each attempt.

Several software companies have already developed effective machine learning frameworks tailored to architecture projects.

One of the most notable frameworks developed in recent times is the Generative Adversarial Network (GAN).

GAN is a form of unsupervised machine learning framework that uses two independent artificial frameworks[1] to develop usable designs. The first neural network, the generator, creates whatever design you want, such as a floor plan or a simple window design. Then, a second neural network, the discriminator, decides whether the generator’s output passes out as a good-enough floor plan or window design[2].

This back-and-forth process goes on until the discriminator is ‘satisfied’ with the generator’s output. Essentially, the idea here is to refine the design over time using an unsupervised model or a more collaborative method incorporating human input.

Creating sustainable designs and operations

Construction and building operations account for nearly 40% of the total global carbon emissions[3]. This has necessitated the need for structures that require less energy to build and operate. Artificial intelligence and machine learning algorithms can help architects streamline operations, thus reducing carbon emissions during construction.

Once construction is complete, architects can deploy AI-powered energy analytics platforms to adjust building temperatures automatically based on various factors, including weather, occupancy levels, and other system parameters.

The average commercial building uses approximately 22.5-kilowatt hours of energy per square foot[4]. Considering the rising costs of energy, coupled with increasing energy requirements in both residential and commercial buildings, architects now have to come up with sustainable designs that incorporate energy-saving systems.

The past few years have seen an increase in the use of networked devices like air quality monitors and smart vents that increase energy efficiency and provide real-time feedback on essential building systems like HVAC systems.

A good example is an IBM project at the Al-Bahar Towers in Abu Dhabi. The system features louvers that open and close as necessary to reduce solar gain and carbon emissions[5]. The result is a significant reduction in energy requirements without sacrificing the comfort of the building’s occupants.

Giving the clients a walk-through of the structure before it’s even built

Over the past decade, virtual reality has gained widespread popularity in the video game industry. The AI-powered technology seems to be creeping its way into the architecture space as more firms use it to give their clients a walk-through of the proposed structure before breaking ground.

This way, clients can have an accurate and realistic view of the structure and give feedback, thus allowing architects to make necessary changes beforehand. Considering the fact that most construction projects exceed their initial estimated budget by more than 16%[6], using AI in architecture is vital to managing added construction costs, which mainly arise from post-construction changes and improvements.

Monitoring large workloads

Drones have become an integral part of the construction process- they help monitor site progress. They’re also used to create an accurate visual layout of the site and create contour maps that often result in huge cost savings on surveys and catching errors before moving the machinery out of the site.

By measuring and taking note of the position of each stockpile on the site, AI-powered drones can help increase the efficiency of earth-moving operations on a construction site. They can also monitor and direct trucks and excavators throughout the site, thus optimizing the process even further.

Introducing parametric architecture

Parametric architecture utilizes algorithmic processes to create different geometric shapes that differ from the traditional lines and right angles that have long been the norm of architecture projects. The result is more creative projects, some of which the architect might have never imagined.

This design feature combines the features of CBD tools with geometric design to form a sort of architectural programming language. With parametric architecture, an architect can create several variations of an idea, then settle for the right one. The design feature also enables you to simulate interactions with the design to make any alterations before deciding on a specific design.

Reducing occupational hazards

According to the Bureau of Labor Statistics, roughly 150,000 accidents happen each year at construction sites[7]. Construction workers also get fatally injured five times more often than with any other profession[8].

Accidents in construction sites mainly result from falling, electrocution, faulty machinery, and being struck by falling objects. Construction workers safeguard themselves from some of these dangers by wearing protective gear. However, in some cases, the protective gear is either faulty or ineffective, thus prompting architects to seek other means of safeguarding the construction site.

Using artificial intelligence in architecture can reduce accident occurrence significantly. AI-powered machines and robots can significantly reduce the number of human workers needed for a construction project, thus reducing the possibility of accidents.

Architects can also use AI in architecture to inspect the safety of a construction site beforehand to stay informed of possible dangers. Additionally, AI can help architects and building managers visualize the connection between accidents and various working conditions; connections that would have otherwise gone unnoticed.

Preventing cost overruns

Most construction projects go over budget. In fact, most lenders often include a contingency budget for cost overruns. When designing a building, architects often have to consider the cost of materials, labor, and machinery in their final budget. But, considering the fact that construction projects often have deviations, even the best estimates often end up falling short.

That’s where AI in architecture comes in. Artificial intelligence can do more precise estimates whilst taking more factors into account. AI can consider things like the size of your construction project, the project manager’s performance, contract type, and many others. By compiling this with data from online sources, AI can accurately calculate the labor and material cost for every section of the project.

Encouraging offsite construction

Offsite factories are revolutionizing the construction process. Gone are the days when all matters pertaining to a construction project were handled onsite. Now, architects and building contractors are embracing the idea of making various building components offsite, then assembling them onsite.

For instance, architects and construction managers can have the walls and windows of a building constructed in an offsite factory, then assembled at the construction site. Most of these projects use autonomous machines and robots, which are not only efficient but also cost-effective.

Besides making the construction project run faster, embracing AI in architecture makes the work easier, safer, and more cost-effective.

Addressing labor shortages

Recent labor shortages, coupled with the desire to boost productivity, have prompted many players in the construction industry to invest in AI and data science. According to a 2017 McKinsey report, construction companies can increase productivity by as much as 50% through real-time data analytics[9].

Many construction firms are already using artificial intelligence and machine learning to plan for the distribution of labor and machinery across projects. By constantly evaluating progress and the location of workers and equipment in real-time, project managers can tell which job sites need more workers or equipment to complete the project on schedule.

Improving building maintenance

Besides construction, AI also plays a major role in building maintenance. After construction, building managers can install AI systems to detect any faulty parts and have them repaired before they cause significant damage.

AI in arhitecture also enables building managers and architects to monitor energy usage patterns and other utilities in the building, thus helping the client save energy and reduce building maintenance costs. A good example is the Nest Thermostat[10]. This AI-powered thermostat can help keep the building at optimal temperatures using minimal energy.

There are other building AI systems that can also help you detect leaks, manage building lighting, and detect malfunctions before they develop into a bigger problem.

Making construction sites more productive

Several companies are already offering self-driving construction machinery that performs repetitive tasks faster and more accurately than their human counterparts. Construction activities like pouring concrete, bricklaying, and demolition can be performed by autonomous or semi-autonomous machinery with the help of a human programmer who stipulates exact specifications.

Using autonomous machinery in construction reduces the need for human labor and the overall time required to complete the project. Construction managers can also use AI to track productivity and improve construction processes for better productivity.

Final thoughts on artificial intelligence in architecture

As more companies leverage AI in architecture, we’re bound to see a major shift from traditional building design and construction processes to more efficient and cost-effective methods.

Despite the fears of massive job losses in the construction industry, artificial intelligence is unlikely to replace the human workforce completely but rather alter construction processes, reduce expansive errors, and make building operations more efficient.

Do you want to know more about artificial intelligence? Take a look at our AI consulting services.

References

[1]Ibm. com. Neural Networks. URL: https://www.ibm.com/cloud/learn/neural-networks. Accessed September 18, 2022
[2] Google.com. Machine Learning. URL: https://developers.google.com/machine-learning/gan/gan_structure. Accessed September 18, 2022
[3] Architecture2030.org. Why the Building Sector. URL: https://architecture2030.org/why-the-building-sector/. Accessed September 18, 2022
[4] Iotacommunications.com. Benchmarking Commercial Building Energy Use Per Square Foot. URL:  https://bit.ly/3xIwENe. Accessed September 18, 2022
[5] Archdaily.com. Al-bahar Towers Responsive Façade Aedas. URL: https://www.archdaily.com/270592/al-bahar-towers-responsive-facade-aedas. Accessed September 18, 2022
[6] Propellraero. 10 Construction Project Cost-Overrun Statistics You Need to Hear. URL:  https://www.propelleraero.com/blog/10-construction-project-cost-overrun-statistics-you-need-to-hear/. Accessed September 18, 2022
[7] Bls.gov. URL: https://www.bls.gov. Accessed September 18, 2022
[8] Osha.gov. Falls are the Leading Cause of Death in Construction. URL: https://bit.ly/3BXeaez. Accessed September 18, 2022
[9] Mckinsey.com. Reinventing Construction.  URL: https://mck.co/3r0eL98. Accessed September 18, 2022
[10] Store.google.com. Next Learning Thermostat 3rd Generation. URL: https://store.google.com/us/product/nest_learning_thermostat_3rd_gen. Accessed September 18, 2022