RESEARCH ON MULTI-ATTRIBUTE DECISION MAKING AND VERIFICATION METHODS FOR SYSTEMS CONFRONTATION

Multi-attribute decision making (MADM) serves as a crucial method for optimizing schemes during the conceptual design phase of aircraft. However, the lack of effective quantitative evaluation methods for the accuracy of decision results poses a challenge. To address this, a method for verifying and optimizing multi-attribute decision results of aircraft conceptual design schemes based on System-of-systems (SoS) confrontation simulation is proposed. Initially, the general process of MADM is delineated, forming the SAWC modeling framework. Utilizing this framework, a MADM model for aircraft tactical-level conceptual schemes is constructed. Subsequently, a method for establishing a systematic verification environment for aircraft is designed, utilizing system modeling language, component modeling, and task completion rate as key components. Finally, a model verification method leveraging the Pearson correlation coefficient as the indicator is developed. Additionally, a model optimization algorithm based on particle swarm optimization with mapping function and weight factor as optimization targets is explored. The proposed method is validated through a ground attack mission scenario involving a ground attack UAV. Systematic verification demonstrates an improvement in the accuracy of the MADM model from 0.6976 to 0.9154. These results highlight the effectiveness of the method in verifying decision results within the aircraft domain, as well as explaining and optimizing the decision model. Ultimately, this approach can contribute to the systematic demonstration stage of aircraft conceptual design.