Target threat assessment is very important for air combat mission planning.Due to the uncertainty and fuzziness of target information in traditional assessment model,the model and algorithm for target threat assessment based on improved Generalized Regression Neural Network (GRNN) are proposed,with multi-aircraft air-combat formation as the starting point.This optimization algorithm can quickly find the optimal scatter coefficient through traversing values within the scatter coefficient interval,thus enabling the model to reach the optimal simulation output.Considering that most of the current air-combats are formation operation,the overall threat level of targets to our formation is selected as the assessment criteria,which improves the reliability of assessment results.Finally,an example is given to verify the effectiveness and correctness of this optimization model.