Error Calculation of Flight Delay Prediction using Various Machine Learning Approaches

Whenever an airline gets late by more than 15–20 minutes, this is termed delayed by the Federal Aviation Administration. This delay results in severe damage for airlines that operate passenger aircraft. Hence, forecasting is critical throughout the decision-making procedure for all corporate airline operators. Airlines can get delayed due to many reasons such as adverse weather, a technical fault etc. This results in the delayed arrival of the airplane at the destination airport and leaves the passengers dissatisfied. Airline and meteorological facts, a forecast model of the on-arrivals airlines, is recommended. This paper tries to employ a variety of Machine Learning and Deep Learning methods including Decision Tree, Regression analysis, Probabilistic Ridge, Random Forest, Gradient Boosting Regression, Logistic Regression, and Long Short-Term Memory (LSTM) for error calculation. Proposed Model provides forecasting if a specified aircraft will arrive on time or not and enhance the precision of predictive models by employing hyper tuning methods.