[Paper Review] GPT Models in Construction Industry: Opportunities, Limitations, and a Use Case Validation

Abdullahi B. Saka, Ridwan Taiwo+5 more|arXiv (Cornell University)|May 30, 2023

BIM and Construction Integration115 references12 citations

TL;DR

The paper reviews opportunities and challenges of GPT models in the construction industry using a critical review, expert discussions, and a materials-selection use-case prototype.

ABSTRACT

Large Language Models(LLMs) trained on large data sets came into prominence in 2018 after Google introduced BERT. Subsequently, different LLMs such as GPT models from OpenAI have been released. These models perform well on diverse tasks and have been gaining widespread applications in fields such as business and education. However, little is known about the opportunities and challenges of using LLMs in the construction industry. Thus, this study aims to assess GPT models in the construction industry. A critical review, expert discussion and case study validation are employed to achieve the study objectives. The findings revealed opportunities for GPT models throughout the project lifecycle. The challenges of leveraging GPT models are highlighted and a use case prototype is developed for materials selection and optimization. The findings of the study would be of benefit to researchers, practitioners and stakeholders, as it presents research vistas for LLMs in the construction industry.

Motivation & Objective

• Assess how GPT models can support the construction project lifecycle.
• Identify opportunities and challenges in applying LLMs to construction tasks.
• Provide expert-driven insights to guide researchers and practitioners.
• Demonstrate a use-case prototype for materials selection and optimization.

Proposed method

• Conduct a critical literature review on GPT/LLM use in construction.
• Engage expert discussions to validate insights and identify practical considerations.
• Develop a use-case prototype focused on materials selection and optimization.
• Synthesize findings to outline opportunities, limitations, and research directions.

Experimental results

Research questions

• RQ1 What opportunities do GPT models present across the construction project lifecycle?
• RQ2 What are the main limitations and challenges when adopting GPT models in construction contexts?
• RQ3 What practical guidance or best practices emerge for researchers and practitioners?
• RQ4 How can a use-case prototype for materials selection demonstrate the viability of GPT models in construction?
• RQ5 What future research vistas are suggested by expert discussion and case studies?

Key findings

• GPT models offer opportunities across the construction project lifecycle.
• Key challenges include reliability, data privacy, domain specificity, and integration with existing workflows.
• A use-case prototype for materials selection and optimization was developed to validate feasibility.
• Findings provide guidance for researchers, practitioners, and stakeholders on LLMs in construction.
• The study outlines research vistas and future directions for LLM applications in construction.
• The paper emphasizes benefits for both researchers and practitioners.