A common type of landing accident is hard landing. Hard landing, a phase where landing of flights is difficult, which can also make the passengers life to risk. Early detection of this phase results in safe landing. To predict whether it is hard landing or not, an idea has been introduced to obtain best prediction. Hybrid LSTM (long short-term memory) algorithm was used. The study presents a machine learning system that can be used in the cockpit to help the flight crew make decisions about go-arounds based on likelihood of a hard landing. To forecast challenging touchdowns, here the study offers a hybrid approach that uses attributes that model the temporal dependencies of aircraft data as inputs to a semantic network. A machine learning model for cockpit will read data from flight such as tyre elevation, speed, and other values to predict the type of landing. In this LSTM, the features such as Pilot (DH2TD), Actuator (AP2DH), and Physical (AP2TD) together make hybrid LSTM. If predicted hard landing, then it instructs to avoid landing or divert to other route. In the existing method there was a result of 85% sensitivity and 74% specificity, which improve with the hybrid LSTM to a sensitivity of 95% and specificity of 96%.