Podcast: Play in new window | Download | Embed
In this episode, I talk with Robert K. Otani, P.E., LEED AP, chief technology officer of Thornton-Tomasetti and founder of CORE studio, about the role of AI (artificial intelligence) and ML (machine learning) in engineering. We discuss the evolution of these fields through technology and examine impactful case studies of AI and ML in engineering design and construction.
***The video version of this episode can be viewed here.***
Engineering Quotes:
Here Are Some of the Questions I Ask Robert:
- How does your engineering and project management experience at Thornton Thomas contribute to shaping the firm’s technology strategy?
- What triggered your shift toward working on AI and ML in the last eight years?
- Are you training a computer to mimic human intelligence, which typically takes five to 10 years to develop?
- Can you share more case studies or projects where AI and ML have been successfully implemented?
- Instead of tedious tasks like counting cracks, how does automation free up designers to focus on creative repair details, opening new avenues and letting engineers excel at what they do best?
- What thoughts do you have on using AI and ML applications to make engineering projects more efficient and effective?
- What challenges are associated with integrating AI and ML into engineering projects?
- How do you see AI and ML becoming part of the daily workflows of engineers?
- Can you make tools, like the 15-parameter beam design, where a chat interface quickly generates and verifies SD-level designs?
- What advice would you offer to those entering or already in the profession who are keen on specializing in AI and ML in our field?
Here Are Some Key Points Discussed in This Episode About How AI and ML Are Revolutionizing Engineering Design and Construction:
- Building AI applications boils down to knowing your field inside out. The key is using AI tools to fix real, annoying problems. Those who’ve been in the trenches, dealing with repetitive tasks, are the ones who can spot what needs fixing. In the world of machine learning and AI, it’s about tackling those issues head-on.
- In 2015, a team member brought up machine learning, highlighting its role in big tech. Realizing its potential, Robert saw it as predicting solutions by fitting curves to smart data, like what engineers do. This sparked the idea of creating an app for the whole team, offering fast, accurate solutions and saving time in our daily work.
- The aim is to teach a computer to mimic human intelligence, usually a five- to 10-year process. By efficiently using real information and tapping into collective expertise for fast problem-solving, the goal is to speed up engineering, providing quick answers that can be later validated through traditional methods.
- In the AI landscape, a startup called T2D2.ai emerged from an R&D project in 2017 or 2018. Within six months, they trained a model using computer vision to detect concrete cracks and expanded its capabilities to assess various exterior damages. This tech significantly cuts down building inspection time compared to manual efforts, showcasing a shift in how work is done through AI. It’s a glimpse into the industry’s early stages of utilizing AI, with the potential for future innovations to create entirely new problem-solving approaches.
- The concept involves adopting frequent, cost-effective inspections to prevent major issues, with a focus on tracking changes over time and observing the evolution of existing damage — an often-overlooked task due to its manual nature. This digitized approach introduces fresh possibilities and adds value through technology, offering a more proactive and cost-effective strategy compared to traditional industry practices.
- Project management, encompassing technical, financial, and staffing aspects, is often overlooked in our field, with critical information scattered across various platforms. The challenge is unlocking and streamlining this data, especially in the outdated accounting and finance sector. As APIs slowly emerge, there’s a move toward starting with data analytics before gradually integrating the predictive capabilities of AI, a domain just beginning to unfold in the industry.
- In engineering, the main challenge is overcoming resistance to change with AI and ML. Ongoing presentations aim to educate teams and make engineers see AI as a powerful tool, almost like “Excel on steroids.” Achieving 95% accuracy using engineers’ spreadsheets highlights the potential of AI in simplifying tasks and transforming workflows.
- In engineering, we shifted from optimizing 2D designs due to tech limits to leaning too much on 3D software. Concerns about later changes led to envisioning AI and ML automating early design. The goal is for engineers to manually work on parts of the building, balancing AI for efficiency without excessive 3D modeling, ultimately delivering efficient, visually appealing designs to clients.
- Acknowledging the efficiency of FAA software and common designer reliance, the focus is on recognizing that clients in early project phases primarily need broad structural characteristics rather than intricate details. The aim for the next five years is to trim the 30% to 40% fee typically allocated for these early design stages, with a dual focus on delivering enhanced final designs and achieving cost savings on projects.
- Aspiring engineers stepping into AI and ML should prioritize training through weekend and online classes, mastering both tools and fundamental engineering principles. Success involves seamlessly combining tool proficiency with a strong understanding of principles.
More Details in This Episode…
About the Guest: Robert K. Otani, P.E., LEED AP
Robert K. Otani, P.E., LEED AP is chief technology officer at Thornton-Tomasetti, Inc., a 1,700-plus person multidisciplinary engineering and consulting firm, and founded the CORE studio, an applications development, advanced computational modeling, and R&D group at his firm. He has extensive structural design and project management experience involving commercial, infrastructure, institutional, cultural, and residential structures on projects totaling over $3 billion in construction. His professional and academic interests focus on informed architecture relating to optimized structural typologies, high-performance structures, and structural sustainability. He served as president of the Structural Engineers Association of New York in 2007 and has been an adjunct professor at Pratt Institute School of Architecture and Columbia Graduate School of Architecture, Planning, and Preservation.
About the Host: Nick Heim, P.E.
Nick Heim, P.E., is a civil engineer with six years of experience in the repair and restoration of existing structures, and host of the AEC Engineering and Technology Podcast (“AECTECH”), where he brings valuable insights and expertise to listeners worldwide.
Nick’s interests lie at the intersection between the built world and technology, and he can be found looking for the ever-changing answer to the question, “How can we do this better?” Nick can be found on LinkedIn, producing content about the use of technologies in his civil engineering career and small business.
Sources/References:
Thornton Tomasetti
CORE Studio
AI: Artificial Intelligence
ML: Machine Learning
AEC Industry
Google
Amazon
Facebook
ChatGPT
T2D2.ai
SAP Software
3D FEA Modeling
Nvidia
Connect with Robert K. Otani, P.E., LEED AP, on LinkedIn
We would love to hear any questions you might have or stories you can share on how AI and ML are revolutionizing engineering design and construction.
Please leave your comments, feedback, or questions in the section below.
To your success,
Nick Heim, P.E.
Host of the AEC Engineering and Technology Podcast
