Leveraging State-of-the-Art Large Language Models for Accident Analysis in the Highway Construction Industry

Work zone hazards in the highway construction industry pose significant risks to worker safety,necessitating effective accident prevention measures. According to the Occupational Safety andHealth Administration (OSHA)'s severe injuries database, the highway construction industry ranksamong the top 10% of contributors to all cases. However, existing research in construction safetyprimarily focuses on work zone vehicle intrusion, building construction, and the mining industry,resulting in a notable gap in understanding accidents specific to highway construction, includingrehabilitation and new construction projects.To bridge this gap, this study proposes leveraging advanced large language models (LLMs), suchas GPT-3, to enhance data-driven analysis. While LLMs have demonstrated their effectiveness invarious scientific domains, their application in construction safety remains limited, with minimalexploration of their potential for analyzing textual accident data. Previous industry researchpredominantly relies on traditional machine learning approaches for natural language processing(NLP) tasks, indicating untapped potential to gain deeper insights and improve accident analysisin highway construction safety. This study aims to achieve efficient data processing, NLPunderstanding, and contextualized insights by utilizing publicly available databases such asOSHA's severe injuries database. The use of these databases underscores the novel and convenientapproach employed in this study, utilizing state-of-the-art LLMs to analyze extensive amounts oftextual accident data and facilitate identifying major causes of accidents. This analysis iscomplemented by traditional statistical approaches, clustering algorithms, and classificationtechniques. The research outcomes are expected to significantly contribute to improving accidentanalysis and prevention in the highway construction industry.