Abstract: The design and improvement of the susceptibility scheme in survivability is a process of constant tradeoff of benefits and costs. A method of bomber susceptibility tradeoff based on support vector machine (SVM) is studied to optimize the benefit-cost of bomber susceptibility tradeoff. The SVM model is established to simplify the susceptibility calculation and the model parameters are optimized by cross validation and grid search. The function relationship between the bomber susceptibility parameter and the increased cost is fitted, and the sensitivity theory is introduced into the susceptibility tradeoff. The value range of the susceptibility parameter is determined from the perspective of optimal cost-effectiveness. Finally, the feasibility of the method is verified by simulation. The results show that the tradeoff optimization method can optimize the range of susceptibility parameters 〖JP2〗under the cost constraint. The accuracy is guaranteed and the computational efficiency is improved.

Key words: bomber, susceptibility, tradeoff optimization, support vector machine (SVM), sensitivity