Generative AI and the construction Industry - a conversation with OpenAI’s ChatGPT
Discover the changes and possibilities that generative AI is bringing to the construction industry!
Caity Taylor, a Senior Solution Engineer at Avvir, recently wrote a byline for Construction Pros about the automation of BIM and Reality Capture Analysis. You can find the original article on Construction Pros here.
According to the Construction Financial Management Association , the average pre-tax net profit for general contractors is between 1.4 and 2.4 percent and for subcontractors between 2.2 to 3.5 percent. With margins this low, there is little room for error, and contractors should absolutely consider utilizing technology and tools to mitigate project risk.
Unfortunately, today’s pace of construction moves so quickly that the most commonly available “tech resources” such as heavy Building Information Modeling and Reality Capture datasets are underutilized due to inaccessibility.
In order to increase the ROI for BIM and Reality Capture programs, the next big focus in construction technology should be on connecting data sets. This will let platforms collaborate and produce automated analysis with less human supervision, allowing teams to focus on other important tasks.
There are two major themes emerging that can immediately assist with connecting data with the tools available today. First, creating a connection between BIM and Reality through continuous updates and synchronization. Second, automating the translation of BIM to Reality will avoid unnecessary discrepancies in the first place and greatly reduce errors.
During the design and coordination phase of most major construction projects, there is a lot of effort put into spatially coordinating BIM elements prior to installation. If everything is installed precisely per the plans, in theory, there should be no issues on site. However, the reality is that adjustments happen in the field all the time - often for good and intentional reasons.
Today’s general contractors and subs need a method to seamlessly pivot from ‘reality capture data’ back to ‘design intent.’ AR and VR solutions can provide this context by overlaying the design models and reality capture data, but often there is still a heavy component of human QAQC to identify where the issues lie on site.
One of the most common complaints of a VDC manager is that models were coordinated or “clash free,” but installation errors on site have caused new issues that could not have been prevented through a typical coordination process. A reality capture system is needed that updates the model in near real time to proactively capture as-built conditions and identify downstream impacts. In this way, teams can continuously coordinate their models as conditions onsite inevitably change.
Throughout a building’s life cycle, commercial spaces will undergo major renovations every 5-10 years. Therefore, it is critical that each renovation is properly documented. Although it is a common “close-out requirement,” As-Built documentation is rarely very accurate. This results in downstream costs and delays for owners who are typically responsible for providing this documentation to consultants at the start of a project.
As owners become more advanced, and regulatory measures require more robust close-out documentation, the “As-Built” model is becoming a more common requirement. Creating this As-Built model can be a costly and time-consuming task, leading to errors and resulting in delayed final payments. Automating the placement of modeled geometry will drastically reduce the time required to produce quality As-Builts . The resulting model is also much less prone to error due to human oversight.
The reality is that until the translation of BIM to fabrication is automated, there will always be necessary site modifications and installation errors resulting in rework. One solution to this problem is to deploy robots to directly translate design to construction. Robotics in the construction industry is not an entirely new concept, with advancements in manufacturing, prefabrication, and logistics in the last few decades. However, one of the most disruptive advancements in the industry is deploying robotics on construction jobsites tasked with highly repetitive, dangerous, or tedious tasks.
Today’s construction robotics capabilities include laying out and even building walls, operating autonomous equipment, using imaging technology to gain valuable insights, and remotely operating equipment to minimize safety risk imposed on staff on the jobsite.
Construction robotics can even be used for surveying, surveillance, and inspection. When partnered with the right applications, these autonomous site capture technologies offer a wide range of benefits, from increased productivity, safety outcomes, and even overall project outcomes.
As the construction market evolves and grows alongside new technological innovations, robots could soon become commonplace on our job sites. This would result in general contractors and developers leaning into the technology and startup space more aggressively through partnerships and pilot programs across portfolios to test new solutions and evaluate outcomes.
Construction Robotics in the field offers considerable advancement in efficiency and safety protocols. Robots perform tedious, repetitive, and dangerous tasks, freeing up site personnel to focus on the bigger picture and keep projects on track. Additionally, utilizing robotics to monitor and record jobsite conditions continuously offers insight into project status, progress, and quality that has never before been available.
The adoption of construction technology across the industry will allow better data management and more opportunity to carry learnings from project to project. By taking steps to connect BIM and Reality Capture, initial investments will be greatly increased.
To learn more about Avvir and our reality analysis platform, check out our solution page.
5/10/21
Team Spotlight
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat.