Abstract
Safety in the construction sector is a critical concern due to the high frequency of accidents and their impact on worker health, project timelines, and productivity. To the best of our knowledge, no prior study in the Kingdom of Saudi Arabia (KSA) has applied quantitative time series forecasting to construction accident data. This study analyzes over a decade of monthly accident records (January 2011-September 2022) from the General Organization for Social Insurance (GOSI) using three univariate forecasting models: Seasonal AutoRegressive Integrated Moving Average (SARIMA), Holt-Winters exponential smoothing, and Simple Exponential Smoothing (SES). The analysis identifies recurring seasonal patterns, long-term trends, and quantifies the impact of the COVID-19 lockdown on accident rates. SARIMA (1,1,1) (1,1,1,12) achieved the best performance, with a Mean Absolute Error of 74.75 and Root Mean Squared Error of 103.77, effectively capturing both seasonal cycles and trend behavior. By integrating historical pattern analysis with predictive modeling, the study provides a data-driven basis for proactive safety planning and accident prevention in the Saudi construction industry.