采用多目标决策对从国外引进的稻种资源进行定量评价.从173份国外稻种资源中筛选出3份表现最优的稻种资源为IR78875- 190-B-1 -3、IR78994 -B - 18 -B -B -B和IR82198 -24 -2 -2.另对173份稻种资源的农艺性状进行相关分析和通径分析,结果表明,各农艺性状与产量的相关性与多目标决策中的熵和熵权的结果是一致的.最后选择多目标决策评价法筛选的最优和最差的(各10份)稻种资源进行聚类分析,将其聚为两大类,结果证明,多目标决策评价模型在大量稻种资源的评价是有效的.该方法避免了由单一性状指标来判断稻种资源优劣所造成的偏差,可为水稻育种选择优良亲本提供理论依据.%The Multiple Objective Decision Making ( MODM) was used to quantitatively analyze foreign rice germplasms. Three best performing rice germplasms were screened out from 173 foreign rice germplasms, which were IR78875-190-B-1-3, IR78994-B-18-BBB and IR82198-24-2-2. The main agronomic characters of 173 rice germplasms were studied by correlation analysis and path analysis. The results showed that, the correlation between agronomic characters and grain yield was consistent with that between the entropy and entropy weight of main agronomic characters. Finally, the best ten and the worst ten rice germplasm resources selected by the MODM would be clustered into two groups. The results proved that MODM was effective in evaluating a large number of rice germplasms. The MODM method can avoid the bias caused by single trait decision methods, and provide more reliable decision for parents selection in breeding programs.
展开▼
