摘要
船型综合评价是一个多目标决策问题。为克服采用单一方法确定权重存在的片面性,基于主观赋权的层次分析法(AHP)、客观赋权的熵权法(EWM)和智能赋权的BP-神经网络法3类赋权法得到的权重,采用博弈论法来确定指标的组合权重,将各评价指标组合权重与逼近理想解排序法(TOPSIS)结合,建立博弈论–TOPSIS评价模型。并以一组消防船为例进行了案例分析。评价结果表明,博弈论–TOPSIS法与改进复合权重TOPSIS法的评价结果趋势一致,且博弈论–TOPSIS法的方案区分度更大,说明采用博弈论法确定的指标组合权重更加合理,能够为多方案船型评价工作提供更有效的评价工具。
The comprehensive evaluation of ship forms is a multi-objective decision-making issue.In order to overcome the one-sidedness of determining the weight by using a single method,we used game theory to determine the combined weights of pre-given indicators based on the weights obtained by three kinds of weighting methods,including analytic hierarchy process(AHP)of subjective weighting,entropy weight method(EWM)of objective weighting and BP-neural network method of intelligent weighting,and combined the weights of each evaluation index with technique for order preference by similarity to an ideal solution(TOPSIS)method to establish a game theory-TOPSIS evaluation model.A case study was conducted with a group of fireboats as an example.The evaluation results show that the game theory-TOPSIS method is consistent with the evaluation result of the improved composite weight TOPSIS method,and the game theory-TOPSIS method has certain advantages in distinguishing different evaluation methods,which shows the index weight determined by the game theory is reasonable and can provide an effective evaluation tool for multi-scheme ship evaluation.
作者
张明霞
韩丹
赵桐鸣
ZHANG Mingxia;HAN Dan;ZHAO Tongming(School of Naval Architecture and Ocean Engineering,Dalian University of Technology,Dalian 116024,China)
出处
《应用科技》
CAS
2020年第5期13-19,共7页
Applied Science and Technology
关键词
博弈论
逼近理想解排序法
层次分析法
BP-神经网络
熵权法
改进复合权重
消防船
船型综合评价
game theory
technique for order preference by similarity to an ideal solution(TOPSIS)
analytic hierarchy process(AHP)
BP-neural network
entropy weight method(EWM)
improved composite weight
fireboat
ship type comprehensive evaluation
