Extending the Open BIM Standard to Serve Permitting and Regulatory Approvals
Many developers and building owners feel that regulatory approvals take too much time and effort. This poses a significant delay and risk for building projects. One way to speed up the building permit process is to automate it as much as possible.
Because 2D PDF files are limited in their capabilities, several governments are developing BIM-based building permits. Regions leading this development include Singapore, the Nordic countries, the UK, Japan, South Korea, and Dubai.
The trend is to use the open IFC standard for model submissions instead of native BIM models. However, the standard has not yet offered universal definitions for automating compliance checking. For example, IFC has had a class for walls (IfcWall) but no standard way to distinguish between specific types of interior walls, e.g., partition walls, fire-rated walls, or sound-insulating walls. This has forced companies or authorities to define custom property sets or user-defined classifications.
The RIR project
At this year’s WDBE summit, Tomi Henttinen (pictured) presented buildingSMART’s RIR (Regulatory Information Requirements) project, which will improve model machine-readability for regulatory purposes.
Under Tomi’s leadership, the international RIR team focused on four areas:
1. Building permit application data, such as building identification and characteristics
2. Life-safety, structure, and fire-safety-related information
3. Energy, environmental (e.g., carbon footprint), and re-purposing data
4. Accessibility of buildings
The process looked at data needed at four regulatory stages: concept approval, technical approval, construction approval, and approval to occupancy.
The results
The RIR project introduced nearly 300 new IFC properties to the IFC standard and almost 200 new Types. These include the definitions for the previously mentioned interior walls, for example.
The project also created a use case and a ruleset, the Regulatory Information Delivery Specification (IDS), which you can download.
Tomi concluded that the RIR project's results will improve communication between applicants and regulators. Design and model-checking apps can use the new definitions to pre-check designs before they are submitted to authorities. Overall, automation will make building approval more streamlined and faster.
EVENTS
AI in AEC 2025 has extended its call for presentations until November 1, 2024.
InfraBIMOpen 2025 tickets are now available for individuals and groups.
FROM AEC BUSINESS
Safeguarding AEC Data: Best Practices and Insights from Egnyte’s Kevin Soohoo
In this episode of the AEC Business Podcast, I had the pleasure of speaking with Kevin Soohoo , Senior Director of Global AEC Practice at Egnyte. We delved into the critical topic of data security in the AEC industry, particularly in the context of AI.
Construction and AI: What Contractors Need to Know from ABC’s New Report
The Associated Builders and Contractors (ABC) has just released its fourth annual construction technology report, which dives deep into AI’s evolving role in the construction industry.
“ABC contractor members and the overall contracting community want more information on AI and how it can help them improve safety, quality and profitability—and win more work,” said Matt Abeles, ABC vice president of construction technology and innovation. The newly released ABC AI Tech Report delivers on this need, highlighting AI-driven case studies, resources, and thought leadership from ABC’s Tech Alliance.
Simplismart Secures $7M to Make AI Adoption Easier for Enterprises
AI adoption is skyrocketing, with OpenAI projected to generate over $10 billion in revenue next year. However, many businesses still struggle to integrate AI into their operations. Simplismart, a company founded by former Oracle and Google engineers, aims to address this challenge by simplifying AI deployment. With a recent $7 million funding round led by Accel, Simplismart is set to accelerate its efforts.
Recommended by LinkedIn
AI in AEC NEWS
Boston Dynamics teams with TRI to bring AI smarts to Atlas humanoid robot
🦾 Boston Dynamics and the Toyota Research Institute (TRI) announced on Wednesday their plans to integrate AI-based robotic intelligence into the electric Atlas humanoid robot. This collaboration will build on TRI’s work with large behavior models (LBMs), which function similarly to the well-known large language models (LLMs) used in platforms like ChatGPT.
🤖 “In machine learning, up until quite recently there was a tradeoff, where it works, but you need millions of training cases,” TRI’s head Gill Pratt explained. “When you’re doing physical things, you don’t have time for that many, and the machine will break down before you get to 10,000. Now it seems that we need dozens. The reason for the dozens is that we need to have some diversity in the training cases. But in some cases, it’s less.”
AI governance gap: 95% of firms haven’t implemented frameworks
In a survey of 600 CEOs, CIOs, and CTOs from large companies across the US, UK, and Germany, only 5% of executives reported that their organization had implemented any AI governance framework. However, 82% stated that implementing AI governance solutions is a somewhat or extremely pressing priority, with 85% planning to implement such solutions by summer 2025.
How AI can augment the design visualization process
AI introduces novel design possibilities, from generating tailored images based on textual descriptions to manipulating, enhancing, and even reanimating visuals.
📷 Image Generation: Create unique visuals by generating images from text prompts, allowing for personalized design assets.
🛠️ Image Manipulation: Utilize tools like Photoshop to refine AI-generated images, keeping desired elements and modifying others. This iterative process enhances creativity and precision.
🎨 Image Enhancement: Begin with a digital sketch, let AI produce a high-quality output, and refine the results. This back-and-forth mirrors traditional rendering software, speeding up completion time while expanding creative horizons.
🎥 Reanimation: Transform static images into animated clips through descriptive text prompts, turning complex animations into quick, cost-effective creations.
🖼️✨ AI in design is an evolving journey, opening the door to boundless innovation and exploration. This post, crafted in September 2024, offers a snapshot of the ever-changing intersection of AI and design.
AI is hard (to do well)
Generative AI (GenAI) is both promising and challenging. Keir Regan-Alexander, founder of Arka Works, offers strategies for its effective implementation in professional settings.
🔍 The Reality: AI isn’t a silver bullet nor a scam. Its success depends on the nuanced integration into existing workflows, requiring human oversight at critical steps.
📉 Challenges: Businesses face inflated expectations versus reality. Structured data, ethical frameworks, and skill training are essential for meaningful AI integration.
📈 Progress: OpenAI’s classification system aims for advanced AI capabilities. The journey to maturity involves gradual adaptations, focusing on core technologies like diffusion and language models. 🎯 Strategy: An impactful AI adoption process emphasizes team-led experimentation and incremental changes, maintaining control and sound judgment.
🤖 Future Outlook: AI isn’t a magic solution but a tool to enhance productivity and enjoyment. We’re early in this journey, so stay curious and experimental.
Keir’s insights stem from his experience at Arka Works, where he blends architectural practice with Generative Design and AI innovations.
Construction tech start-up sends AI-powered “micro-factories” to the US to build affordable housing
🏗️ British construction tech start-up AUAR is expanding to the U.S. with two pop-up robotic micro-factories, aiming to revolutionize house building. Founded in 2019, AUAR licenses its tech to builders, enabling them to construct low-energy homes at standard prices and address the housing crisis in collaboration with Rival Holdings.
🤖 Each micro-factory, developed with ABB, can produce up to 180 homes annually, automating design and construction with AI-powered software. Packed in shipping containers, they offer quick deployment and require minimal specialized training. AUAR’s “hardware-as-a-service” model ensures accessibility with low upfront costs.
🏡 Rival Holdings’ first project will focus on a low-energy Accessory Dwelling Unit (ADU), using AUAR’s sustainable tech to reduce costs and improve quality.
Austin will use AI to evaluate residential construction plans
🚀 Starting next summer, Austin will use AI to help review residential building plans. In 2022, getting a construction permit took an average of 345 days. To streamline this, Austin signed a 3-year, $3.5M contract with Australia’s Archistar, with an option to extend for two more years at $6M.
🚦 The AI aims to automate a significant portion of the initial review process, starting with single-family homes, but humans will still oversee to ensure complexity is addressed. The city piloted this AI earlier, finding it accurate 75% of the time.
🖥️ While AI can’t replace human oversight, it can enhance efficiency in routine evaluations.
🔍 Professor Bill O’Brien from UT Austin highlights the value of AI in repetitive tasks, freeing skilled workers for complex cases. Local builder Brita Wallace hopes it will standardize rule enforcement, while Trevor Odelberg advises careful human oversight remains crucial. If successful, Austin plans to expand AI use beyond single-family home applications.
AEC BUSINESS PARTNERS
Here are our fantastic supporting partners:
Helping contractors with high-risk contracts | Posts and articles about the process.
2moGreat advice! Thanks for this, Aarni
AI & Data Governance for AEC Firms / Construction Technology / Innovation / Start Up GTM / Advisory Board / Adjunct Professor / Philanthropy
2moI enjoyed our conversation, thank you for having me on your show Aarni Heiskanen