Predictive model to determine patterns and trends that influence work accidents

Abstract

Safety at work is a concern for any organization that aspires to excellence, not only because it guarantees an accident-free workspace, but also because it increases the confidence, commitment, and motivation of employees. Using advanced tools and analytical techniques to explore large and small data sets can help identify these issues early enough to make timely decisions. Despite concerns about accidents, few analyzes have been conducted to date to identify specific trends or patterns, so this study focuses on a database containing information on accidents that occurred between 2015 and 2023 at a company. of the State. The main objective is to analyze the causes of accidents. Machine learning algorithms and data science techniques were used to identify patterns and trends in workplace accidents. The data is then classified in detail to better understand the changing behavior based on linear regression. After analyzing the forecasts, it was determined that they were very consistent with the actual results, which confirms the precision of the model used.