[Paper Review] Generative AI in the Construction Industry: A State-of-the-art Analysis

Ridwan Taiwo, Idris Temitope Bello+5 more|arXiv (Cornell University)|Feb 15, 2024

BIM and Construction Integration17 citations

TL;DR

A state-of-the-art analysis of generative AI in construction, proposing a framework for custom AI solutions and demonstrating a case study with contract document querying.

ABSTRACT

The construction industry is a vital sector of the global economy, but it faces many productivity challenges in various processes, such as design, planning, procurement, inspection, and maintenance. Generative artificial intelligence (AI), which can create novel and realistic data or content, such as text, image, video, or code, based on some input or prior knowledge, offers innovative and disruptive solutions to address these challenges. However, there is a gap in the literature on the current state, opportunities, and challenges of generative AI in the construction industry. This study aims to fill this gap by providing a state-of-the-art analysis of generative AI in construction, with three objectives: (1) to review and categorize the existing and emerging generative AI opportunities and challenges in the construction industry; (2) to propose a framework for construction firms to build customized generative AI solutions using their own data, comprising steps such as data collection, dataset curation, training custom large language model (LLM), model evaluation, and deployment; and (3) to demonstrate the framework via a case study of developing a generative model for querying contract documents. The results show that retrieval augmented generation (RAG) improves the baseline LLM by 5.2, 9.4, and 4.8% in terms of quality, relevance, and reproducibility. This study provides academics and construction professionals with a comprehensive analysis and practical framework to guide the adoption of generative AI techniques to enhance productivity, quality, safety, and sustainability across the construction industry.

Motivation & Objective

• Assess current opportunities and challenges of generative AI in construction
• Propose a framework for building customized generative AI solutions using firm data
• Demonstrate the framework through a case study on querying contract documents

Proposed method

• Review and categorize existing and emerging generative AI opportunities and challenges in construction
• Propose a step-by-step framework for data collection, dataset curation, training a custom LLM, model evaluation, and deployment
• Use a case study to illustrate developing a generative model for querying contract documents
• Evaluate the impact of retrieval augmented generation (RAG) on baseline LLM performance

Experimental results

Research questions

• RQ1 What are the current opportunities and challenges of generative AI in the construction industry?
• RQ2 How can construction firms build customized generative AI solutions using their own data?
• RQ3 How effective is retrieval augmented generation (RAG) in enhancing generative models for construction tasks?
• RQ4 Can a practical framework be demonstrated end-to-end with a contract document querying use case?

Key findings

• RAG improves baseline LLM quality, relevance, and reproducibility by 5.2%, 9.4%, and 4.8%, respectively.
• A practical framework is proposed for data collection, dataset curation, custom LLM training, evaluation, and deployment.
• The framework is demonstrated via a case study on developing a generative model to query contract documents.
• The study discusses opportunities to enhance productivity, quality, safety, and sustainability in construction through generative AI.