The invention relates to an interference decision-making method based on a directional mutation search artificial bee colony algorithm, and the method comprises the steps: firstly carrying out the one-to-one evaluation of a jammer and a radar through the five aspects of time, space, power, frequency and interference pattern, and constructing an interference benefit matrix Bij through the benefit values obtained through the evaluation; secondly, performing threat level evaluation pj on the radar according to the working state of the radar, the radar system and the actual combat mission of the radar through a front reconnaissance plane; constructing a radar interference decision model and a target function by using the interference benefit matrix, the radar threat level and the constraint condition; and finally, performing solving through an artificial bee colony algorithm based on directional mutation search to obtain an optimal distribution scheme and an interference income value; and then issuing tasks to all jammers according to the optimal scheme. The method is suitable for interference resource task allocation under the condition that the multiple jammers cooperatively interfere with the networking radar, the number of iterations is reduced, meanwhile, the average profit value is larger, the optimization searching probability is improved, and the practicability is very high.