Leading Building Design Firm Enhances Modeling and Simulation with Generative AI

Challenges

The company faced inefficiencies in validating 3D design models, a key step in ensuring the accuracy and structural integrity of its commercial steel buildings. These buildings support major construction projects nationwide, where precision and compliance are paramount. With a production capacity of 7,000 buildings annually, any delays in the modeling and validation process directly impacted project timelines, customer commitments, and overall revenue generation.

The creation and validation of each 3D design model could require a maximum of 200 hours per project. Modelers needed to manually search through thousands of documents, including annotated contracts, contract amendments, and cross referenced these against the parts database to verify design accuracy. This labor-intensive, repetitive process created bottlenecks and slowed project timelines.

The company needed a faster, more reliable method to create and validate 3D design models. The challenge was to eliminate inefficiencies, automate document analysis, and accelerate model verification – all while ensuring the highest levels of precision and compliance in its modeling and validation workflows.

About the Company

• Leading US-based building design firm
• Designs and assembles approximately 7,000 buildings each year
• 400 modelers and 2,600 builders nation wide

Project Highlights

• Processed 2,000+ complex PDFs detailing various building components, including multi-page tables, annotated sections, horizontally split columns, free-text fields, and contract amendments.
• Leveraged Vision Language Models to reason over the extracted images and provide precise answers to
  in-scope questions based on the content.
• In practice, the firm reduced validation time by 4 hours per building and improved building plan quality across approximately 7,000 projects annually – resulting in $9M in annual economic value.

Solution

Recognizing the need for a more efficient and scalable approach, the company deployed C3 Generative AI for Modeling and Simulation to expedite model creation and validation, accelerate project timelines, and increase capacity. Given that the company produces 7,000 buildings annually, with millions of dollars in revenue dependent on the timely and accurate completion of these structures, ensuring fast and precise 3D design model creation and validation was a business-critical priority.

C3 Generative AI for Modeling and Simulation transformed the model creation and validation process by automating information retrieval and comparisons of customer contracts and the information specified in the 3D design models – enabling modelers to spend their time resolving discrepancies between source documents and the models instead of identifying them. The solution eliminated manual bottlenecks by instantly cross-referencing customer contracts, contract amendments, and the parts database, ensuring that each 3D design model adheres to project-specific requirements. Additionally, the AI-powered system ingested and interpreted handwritten notes on customer contracts and amendments, capturing last-minute adjustments and annotations that were previously difficult to incorporate into the validation workflow.

By flagging inconsistencies before construction begins, the AI-driven system prevented costly errors, reduced the time needed to create and validate 3D design models, and increased operational capacity.

C3 Generative AI for Modeling and Simulation

Key Capabilities

C3 Generative AI for Modeling and Simulation processed contracts and contract amendment PDFs using a custom parsing approach to extract critical insights. This included transforming contract data into images for enhanced analysis.

Ensured that the most recent contract amendments were utilized, resolving
any conflicting information from earlier versions to maintain data accuracy and consistency.

Designed and implemented distributed computing logic to handle computations efficiently, enabling the application to scale seamlessly across the customer’s organization.

Ingested 3D design models and cross referenced the information within the models against what was specified in customer contracts – and flagged any discrepancies.

All structured data sources were seamlessly ingested into the C3 AI data model. Relevant insights, such as 3D building design model data, are extracted to address nearly 100 predefined questions efficiently.