Airlines must manage flight schedules adequately to maintain consumer satisfaction and profitability in operations. This study aims to create a prediction model for aeroplane delays by applying machine learning techniques. This research aims to evaluate the potential of multiple machine learning algorithms in predicting flight delays using a large dataset that includes precise aircraft specifications, airline operating data, and records of previous delays. Throughout-depth pretreatment techniques are used in the study to ensure data accuracy, and feature selection is carried out to pinpoint important delay variables. A variety of measures are used to assess the model, and the decision tree regression model’s evaluation metrics show overall high performance. The R-squared score of 0.933 indicates that the features of the model account for around 93.3% of the variance in the target variable, exhibiting an outstanding capacity to capture data variability. The low values comprising a Mean Squared Error of 102.71 are also observed.