The era of Artificial Intelligence (AI), which previously progressed hesitantly, is now rapidly approaching with full force in the coming years. Even the construction industry, which initially faced difficulties in embracing AI, is now starting to integrate it.
AI has been in existence since the 1950s, but it's only in recent years, thanks to the emergence of voice assistants, autonomous cars like Tesla, and advanced language models such as ChatGPT, that it has gained wider acceptance among the general population.
At Miviso, we are always on the lookout for innovative tools that can enhance our work. Therefore, we're thrilled to announce that Miviso is also embracing AI. After conducting sufficient testing, we're convinced that AI will bring significant benefits to our work.
But what exactly is AI, and how can it help us in the construction industry? If AI has been around for so long, why hasn't it been more widely used until now? In our upcoming blog post, we will address all of these questions and provide insights into our own use of AI and its potential impact.
What is Artificial Intelligence?
Artificial Intelligence is a field within computer science that focuses on developing algorithms and techniques to enable computers to perform human-like intelligent tasks. In order to make these tasks usable for us, there are a few essential requirements.
Firstly, a massive amount of data needs to be available. Using this data, AI can identify and "learn" certain patterns based on our requests (e.g. ChatGPT: when we ask a question). The unique aspect of this learning process is that each time we use it and provide new data, AI can learn more. With this accumulated knowledge, AI can make predictions or decisions that may be difficult or even impossible for the human brain to recognize due to their complexity.
Another fascinating aspect of AI is its ability to assist us in monotonous tasks that require a high level of attention but are not particularly challenging. This is precisely why this technology is so attractive to us.
How can AI be useful in the construction industry?
In the construction industry, Artificial Intelligence can lead to increased efficiency, particularly in the planning stage. However, its benefits extend beyond just efficiency, as AI can also play a crucial role in ensuring safety at construction sites. Real-time monitoring using AI technology can detect potentially hazardous situations and trigger warning signals, similar to the autopilot feature in cars. Additionally, AI can provide valuable assistance in building maintenance and operation.
But why hasn't AI been used in construction earlier?
On one hand, the lack of understanding regarding the capabilities of artificial intelligence has contributed to a lack of trust in its potential, a perception that still persists today. On the other hand, the construction industry itself is complex and requires a high level of knowledge and experience to make experience-based decisions. Therefore, AI had to be taught how to handle these specific requirements.
In existing buildings, there are significant challenges to overcome, such as incomplete or missing data, especially with older projects that were completed before the introduction of digital systems. Additionally, the absence of uniform standards for data collection and analysis makes the interpretation of data more challenging. AI must be adaptable to various data formats and standards to interpret them correctly and accurately.
Despite these obstacles, the use of artificial intelligence in the construction industry will continue to grow in importance. We should view AI systems not as competitors to human competence and decision-making but rather as tools and support for us.
Miviso goes AI
What prompted Miviso to embrace AI?
At our company, we leverage artificial intelligence to enhance the accuracy and precision of our 3D building models. By harnessing the power of AI, we are able to streamline our workflow, reducing both our work time and customer wait times.
As a dynamic and forward-thinking company, we understand the importance of embracing innovation and seeking opportunities to improve efficiency. Given our enthusiasm for working with cutting-edge technologies and our drive to unlock maximum potential, it was only natural for us to wholeheartedly commit to artificial intelligence.
How does Miviso use artificial intelligence?
To harness the power of AI, we integrate plugins tailored to different CAD software providers.In our case, we work with plugins for Revit, which we consider a significant enhancement to our work after a successful initial testing phase. In the first step, it was important for us to determine whether AI tools would be suitable for our activities. Today, that question can clearly be answered with a yes.
In the next and current step, we want to find out which plugin is most suitable and meaningful for us.
At the core of AI in building modeling lies the fundamental principle of point cloud preselection, which relies on high-quality data provided by surveyors. Even at the point cloud stage, AI can recognize various building elements such as walls, windows, doors, stairs, beams, pipes, and more, laying the groundwork for subsequent 3D modeling processes.
Once the elements are selected, they are either sorted by color or organized into an appropriate layer structure, depending on the plugin being utilized. This efficient sorting process saves us valuable time and enables us to work with improved cleanliness and precision.
Furthermore, the layer structure offers additional benefits. It allows us to easily hide elements or point clouds that are not currently being modeled. This level of organization greatly enhances our workflow and significantly simplifies the building model modeling process.
Some plugins even offer the capability to place elements directly. By simply clicking on the corresponding point cloud, and inputting the necessary parameters, the element is automatically created and accurately positioned with minimal effort. This feature enhances our modeling efficiency and streamlines the entire workflow.
Nevertheless, it remains vital to possess proficiency in CAD software and effectively handle library elements when utilizing AI. While artificial intelligence plays a valuable role, the expertise and experience of our skilled employees remain essential. AI does not replace them; rather, it complements their work. "It is crucial to evaluate whether these features are truly necessary and if the cost-benefit ratio is suitable," explains Philip, Co-Founder of Miviso. "When it comes to simple architecture, AI plugins are considered 'nice to have,' as a skilled building modeler can still work efficiently without AI. However, AI tools prove invaluable for MEP projects, especially in dealing with complex technical building equipment. I already foresee tremendous potential in that area."
Lessons learned
In the pursuit of embracing innovations and tools, it is vital to foster collaboration and seek input from the entire team. Their opinions hold significant value and should be duly considered. With this in mind, we conducted a series of inquiries with our BIM modelers, and we have summarized the most important and interesting aspects:
To what extent do you see AI as an enrichment to your work? Sandu: "With the constant evolution of technology, staying up to date can be quite challenging. Artificial intelligence has significantly enhanced our work in various ways, enabling us to operate with greater efficiency and precision. It provides valuable assistance through information, automations, analyses, and recommendations, all of which enhance our workflows. That's pretty cool!"
How precise is artificial intelligence in its work? Can it be trusted? Philip: "A prerequisite for efficiently using AI in building modeling is a clean and high-quality point cloud. Once that requirement is met, you can rely on AI's capabilities. However, it's important to note that skilled building modelers cannot be substituted by AI entirely. An expert is always necessary to validate, interpret, and fine-tune the AI-generated results. The potential of AI is impressive, but it should not be blindly trusted."
What is the summary of the AI tests? Which plugin is best suited? Philip: "Our testing phase is still ongoing, and it requires ample time and diverse projects to determine the most suitable plugin for our needs. Factors such as usability, features, exchangeability, and product support are all taken into consideration. We are actively collaborating with AI plugin developers, providing them with valuable feedback. This mutually beneficial partnership exemplifies how collaboration can generate added value and drive progress. However, the final product selection is yet to be determined."
AI vs. BIM – What exactly is the difference?
Both artificial intelligence and building information modeling play significant roles in enhancing human work processes.
AI leverages data and pattern recognition and finds applications across diverse fields such as medicine, science, finance, transportation, and automation.
On the other hand, BIM is a digital approach that encompasses the creation, management, and representation of building information using a 3D model throughout its lifecycle.
Although AI and BIM are distinct technologies, their integration presents an exciting prospect for us. We believe that harnessing the combined power of AI and BIM holds immense potential and offers promising opportunities for our company's future.
Are you interested in the topic of 3D modeling (BIM) and want to learn more about it or do you have specific questions? We would be happy to have a conversation about it.
Michael Danklmaier
Miviso Co-Founder
Tel.: +43 664 4563309