论文部分内容阅读
为了合理预测潜在危险区域人员是否参与疏散的决策行为,以模糊逻辑与离散选择模型为理论基础,构建了能够反映影响人员疏散决策行为的因素及其影响程度的混合预测模型。该模型以二项Logit模型为主体结构,兼用模糊逻辑对主观、定性变量进行处理,在显著性分析的基础上,筛选对疏散决策行为影响较大的解释变量并确定其在效用函数中的系数。采用美国Andrew飓风的疏散行为调查数据,估计和测试了该混合模型,并将其与以往相关模型的预测结果进行了对比。分析结果表明:该混合模型对测试样本的预测准确度达到了85%,相对于以往的疏散行为预测模型,对于调查数据具有更高的拟合度和更好的预测效果。
In order to reasonably predict whether people in potentially dangerous areas are involved in evacuation decision-making, a hybrid prediction model that reflects the factors influencing evacuation decision-making behavior and the degree of influence is constructed on the basis of fuzzy logic and discrete choice model. This model uses the binomial Logit model as the main structure, and uses both fuzzy logic to process the subjective and qualitative variables. Based on the analysis of significance, the model selects the explanatory variables that have a great influence on evacuation decision-making behaviors and determines the coefficients in the utility function . Based on the survey data of evacuation behavior of hurricane Andrew in the United States, the hybrid model was estimated and tested, and compared with the forecasting results of previous related models. The analysis results show that the hybrid model achieves a prediction accuracy of 85% on the test sample, which has a higher fitting degree and better prediction effect than the previous model of evacuation behavior prediction.