| With the acceleration of urbanization,the urban transportation network,which is based on rail transit and supplemented by buses,shared travel modes,and park-and-ride,has become an important form of alleviating traffic congestion.The access problem is a major difficulty that restricts the efficiency of subway travel,and affects the willingness and experience of passengers.Exploring the decision-making process of travelers choosing the access mode will help the government and rail transit operators understand the factors that affect the choice of passengers’ access mode and their behavior characteristics,and provide theoretical support for operation management.First of all,this article investigates the process of travellers’ choice of connection methods in real life through RP research,and found that when faced with multiple access modes,nearly half of the travellers’ decision-making behaviors are not completely rational.Their decisionmaking process can be divided into two stages,first quickly eliminate inappropriate modes based on important attributes,and then comprehensively evaluate the remaining options to make choices.The attributes that individuals rely on in the elimination stage are different.In order to explore the two-stage decision-making process in depth,this article designs an SP survey that includes 20 connection scenarios and finds that travelers have different sensitivity to different access context variables.Keep other attributes unchanged,when the individual’s choice changes with the change of a single variable,the attribute is considered to be a non-compensatory attribute that he pays attention to in the sequential elimination phase.Both individual socio-economic attributes and connection situations will affect the noncompensatory attributes on which travelers make decisions.Therefore,the use of exclusion rules is heterogeneous.Then,this paper uses logistic regression model to capture the non-compensatory attributes of individuals in the sequential elimination phase,and constructs a two-stage access mode choice model that considers the heterogeneity of the elimination rules.Each transport mode has different transportation characteristics,suitable for different access scenarios.According to the descriptive statistical analysis of the SP survey results,when the weather is bad,the number of people access by walking and cycling will be greatly reduced,and they will use better sheltered motor vehicles to access.After the connection distance exceeds 2km,the number of walking and bicycling decreases significantly,and as the connection distance increases,the number of motor vehicle users gradually increases.When the reason for travel is more urgent,individuals will tend to use faster access modes.Based on these phenomena construct three exclusion rules.Finally,compare and analyze the effects of multinomial logit model,elimination by aspects model,two-stage model and the model proposed in this paper.From the perspective of prediction accuracy,the effect of the model proposed in this paper is always better than the multinomial logit model,which shows that when choosing a access mode,there are some travelers who will use heuristic decision method to quickly eliminate inappropriate modes.When the connection distance is 3-5km,the model proposed in this paper performs better than the elimination by aspects model and the two-stage model.This shows that the noncompensatory attributes that individuals pay attention to are not the same,that is,the exclusion rules used by individuals are heterogeneous when making heuristic decision-making.The model proposed in this paper can better capture the non-compensatory cues of individual attention. |
