[关键词]
[摘要]
当城市轨道交通系统出现异常,乘客容易产生盲目、恐慌、从众等心理,并做出非完全理性的决策,与现有MNL模型中的信息完备假设与完全理性假设不一致。为刻画非常态下乘客的不完全理性并考虑乘客个体差异,解决经典MNL模型在非常态下的不适用性,现针对轨道交通非常态情景,通过累积前景理论建模。首先,综合考虑时间、费用、舒适度与便捷性四个对乘客出行行为有直接影响的因素,构建了基于累积前景理论的轨道交通非常态乘客出行方式选择模型,用以刻画乘客的不完全理性。之后,开展问卷调查用以标定模型参数,结合调查结果构建服从泊松分布的差异化参考点,描述模型的参考点依赖现象,并将乘客个体差异引入模型。最终,算例研究结果表明,引入差异化参考点的泊松分布检验值满足大于等于0.05的检验标准,解释了乘客出现不同决策结果的本质,使综合前景呈现随参考点波动的势态。针对非常态情景,该模型体现了乘客面对非常态时的有限理性与个体差异性,效果优于完全理性假设,划分率计算结果相比MNL模型更贴近实际值,总体偏差减小了4.9%,微观计算结果准确率较MNL模型提升了25.4%,可以为轨道交通非常态下的交通需求预测与应急响应提供理论支撑。但受到损失规避心理影响,个别方式的划分率会被放缩,微观下的准确率也有进一步提升空间。
[Key word]
[Abstract]
When the traffic system is abnormal, passengers are prone to have blindness, panic, conformity and other psychology, and make incomplete rational decisions.This is inconsistent with the complete rational assumption and complete information assumption in the existing passenger travel behavior model, such as MNL. Therefore, the incomplete rationality of passengers under abnormal conditions is described through the cumulative prospect theory,and consider individual differences among passengers to solve the inability of classical MNL model.Firstly, comprehensively considering the four factors of time, cost, comfort and convenience, a model of rail transit non normal passenger travel mode selection based on cumulative prospect theory is constructed.It used to characterize the incomplete rationality of passengers.Afterwards,a questionnaire survey is conducted to calibrate the model parameters for rail transit abnormal conditions. Based on the survey results, a differentiated reference point following a Poisson distribution is constructed to describe the reference point dependency phenomenon of the model.Individual differences among passengers are introduced into the model to explain the essence of different decision-making outcomes among passengers.Finally, the results of the case study indicate that the Poisson distribution test value with the introduction of differentiated reference points meets the test criterion of greater than or equal to 0.05. It explain the essence of passengers' different decision-making results, and presenting a trend of comprehensive prospects fluctuating with the reference points.This model focuses on abnormal situations and reflects the incomplete rationality and individual differences of passengers when facing abnormal situations.The overall accuracy is higher than MNL model, and the average absolute error is reduced by 4.9% .The accuracy of microscopic calculation results has been improved by 25.4% compared to the MNL model. It can provide theoretical support for traffic demand prediction under abnormal conditions of rail transit. However, under the influence of the principle of loss aversion, the division rate of certain methods will be reduced. The accuracy at the micro level needs further improvement.
[中图分类号]
[基金项目]
国家重点研发计划(2019YFF0301403);中央高校基本科研业务费专项资金 (2019JBM041)