技术领域Technical field
本申请涉及交通技术领域，尤其涉及一种集成出行策略的设计方法及系统。This application relates to the field of transportation technology, and in particular to a design method and system for integrated travel strategies.
背景技术Background technique
随着信息和通信技术（ICT）的发展，出行即服务（(Mobility as a Service ，MaaS）的出现为出行管理提供了一个新的策略。通过数字平台整合交通资源，实现集成化、一体化的交通出行，从而向出行者提供“门”到“门”的出行服务, 理想化的MaaS平台打通了火车、地铁、公交、出租车、共享汽车、共享单车等多种交通方式的壁垒，通过一个手机应用即可实现所有交通方式的查询、下单、支付并具有行程规划、预约出行等功能，可以为用户提供一段时间跨度内的出行服务。With the development of information and communication technology (ICT), the emergence of Mobility as a Service (MaaS) provides a new strategy for travel management. Integrate transportation resources through digital platforms to achieve integration and integration. Transportation, thereby providing travelers with "door" to "door" travel services. The ideal MaaS platform breaks through the barriers of trains, subways, buses, taxis, shared cars, shared bicycles and other transportation modes, through a The mobile application can enable inquiries, orders, and payments for all transportation modes, and has functions such as itinerary planning and travel reservations, and can provide users with travel services within a period of time.
然而，相关技术中已有的出行服务所针对的是如何提供单次的出行服务，在时间跨度上不能针对用户的需求进行精准辨识。例如，已有的出行服务虽然可以针对用户单次出行提供不同的出行方案，但是在一段时间跨度内，比如一周、一月内，已有的技术无法对这段时间内的用户需求进行辨识，无法为用户推荐较为合理的出行策略。However, existing travel services in related technologies are aimed at providing single travel services, and cannot accurately identify user needs in a time span. For example, although existing travel services can provide users with different travel plans for a single trip, within a span of time, such as a week or a month, existing technology cannot identify user needs during this period. It is impossible to recommend a more reasonable travel strategy for users.
因此，亟需提供一种技术方案，解决相关技术中存在的时间跨度上不能针对用户的需求进行精准辨识的问题，为用户提供MaaS式的出行服务。Therefore, there is an urgent need to provide a technical solution to solve the problem in related technologies that the time span cannot accurately identify user needs, and provide users with MaaS-style travel services.
发明内容Contents of the invention
本申请的目的在于提供一种技术方案，解决相关技术中存在的时间跨度上不能针对用户的需求进行精准辨识的问题。The purpose of this application is to provide a technical solution to solve the problem in related technologies that the time span cannot be accurately identified according to user needs.
基于以上问题，本申请提供一种集成出行策略的设计方法，方法包括以下步骤：Based on the above problems, this application provides an integrated travel strategy design method, which includes the following steps:
获取RP出行调查数据，RP出行调查数据包括出行起点、出行终点、出行时长以及出行方式；Obtain RP travel survey data. RP travel survey data includes travel origin, travel destination, travel duration and travel mode;
对RP出行调查数据进行中心点聚类，获得聚类结果，聚类结果包括不同类型用户对任一类型交通方式的搭配范围；Perform center point clustering on the RP travel survey data to obtain clustering results. The clustering results include the range of matching of any type of transportation mode by different types of users;
根据聚类结果，采用D-efficient设计，在各个类型用户对任一类型交通方式的搭配范围内选择一数值水平作为类型交通方式的额度水平，以生成选择情境，选择情境中包括至少两个集成出行策略，任一集成出行策略包括至少两种类型交通用具的使用额度。According to the clustering results, D-efficient design is used to select a numerical level as the quota level of each type of transportation mode within the range of each type of user's matching of any type of transportation mode to generate a selection situation, which includes at least two integrations. Travel strategy, any integrated travel strategy includes usage quotas for at least two types of transportation equipment.
进一步的，在进行中心点聚类时，采用K-medoids聚类算法对RP出行调查数据进行聚类。Furthermore, when performing center point clustering, the K-medoids clustering algorithm is used to cluster the RP travel survey data.
进一步的，采用K-medoids聚类算法对RP出行调查数据进行聚类包括以下步骤：Further, using the K-medoids clustering algorithm to cluster the RP travel survey data includes the following steps:
对RP出行调查数据进行预处理以构建数据集，数据集中包括各个出行者的多项属性数据，属性数据作为样本点的坐标，出行者的属性数据包括该出行者在预设时间范围内发生的公共交通行程次数、小汽车行程距离和自行车行程时长；The RP travel survey data is preprocessed to construct a data set. The data set includes multiple attribute data of each traveler. The attribute data is used as the coordinate of the sample point. The attribute data of the traveler includes the events that occurred for the traveler within the preset time range. Number of public transport trips, distance traveled by car and duration of bicycle trips;
随机选择k个样本点作为中心点，计算剩余样本点到任一中心点的欧式距离；Randomly selectk sample points as the center point, and calculate the Euclidean distance from the remaining sample points to any center point;
根据计算出的欧氏距离，按照与就近中心的原则，将剩余的样本点分配到各个中心点所代表的类别中，实现初始聚类；According to the calculated Euclidean distance, and in accordance with the principle of the nearest center, the remaining sample points are assigned to the categories represented by each center point to achieve initial clustering;
重新计算聚类中心，遍历类别内的每一个样本点，计算样本点到该类别剩余样本点的欧氏距离，选取欧氏距离最小时所对应的组，将其作为新的中心点；Recalculate the cluster center, traverse each sample point in the category, calculate the Euclidean distance from the sample point to the remaining sample points in the category, select the group corresponding to the smallest Euclidean distance, and use it as the new center point;
根据新的中心点重新进行聚类划分，直至所有的中心点不再发生变化或达到最大迭代次数，以获得当前k种类别的聚类结果。Clustering is re-divided based on the new center points until all center points no longer change or the maximum number of iterations is reached to obtain the current k categories of clustering results.
进一步的，任一样本点到某一中心点的欧氏距离通过如下公式表示：Furthermore, the Euclidean distance from any sample point to a certain center point is expressed by the following formula:
； ;
式中：di为第i个样本点到该中心点的欧氏距离，xij为数据集第i行，第j列的属性数据，xmj为第m个中心点的第j列属性数据。In the formula:di is the Euclidean distance from the i-th sample point to the center point,xij is the attribute data in thei -th row andj- th column of the data set,xmj is thej -th column attribute data of them-th center point .
进一步的，采用K-medoids聚类算法对RP出行调查数据进行聚类时，聚类的类别个数通过以下公式确定：Furthermore, when using the K-medoids clustering algorithm to cluster RP travel survey data, the number of clustering categories is determined by the following formula:
； ;
式中，Ck为第k类别的数据，s为Ck中的数据，ok为Ck的质心，取SSE下降趋势转折点的K值为聚类的类别个数，聚类中每一类别作为同一类型的用户。In the formula,Ck is the data of thekth category,s is the data inCk ,ok is the centroid ofCk , andthe K value of the turning point ofthe SSE downward trend is the number of clustering categories. Each category in the cluster as users of the same type.
进一步的，方法还包括：Further, methods include:
设计至少一种类型交通方式的至少两种补贴方案，根据聚类结果和补贴方案，采用D-efficient设计，在各个类型用户对任一类型交通方式的搭配范围内选择数值水平作为类型交通方式的额度水平，以生成选择情境，选择情境中包括至少两个集成出行策略。Design at least two subsidy plans for at least one type of transportation mode. Based on the clustering results and subsidy plans, use D-efficient design and select numerical levels as the type of transportation mode within the range of each type of user's matching of any type of transportation mode. quota level to generate a choice scenario that includes at least two integrated travel strategies.
进一步的，集成出行策略至少包括以下类型交通用具使用额度的一种或多种：Further, the integrated travel strategy includes at least one or more of the following types of transportation usage quotas:
公共交通的可用次数、网约车总可用里程、网约车里程单价、租赁式交通工具的可用时长。The number of times public transportation is available, the total available mileage of online car-hailing, the unit price of online car-hailing mileage, and the available time of rental transportation.
进一步的，方法还包括：Further, methods include:
进行SP调查数据收集，获得SP调查数据，并对SP调查数据进行预处理，SP调查数据包括：出行者个人属性数据、出行行为属性数据和集成出行策略属性数据；Collect SP survey data, obtain SP survey data, and preprocess the SP survey data. The SP survey data includes: traveler personal attribute data, travel behavior attribute data, and integrated travel strategy attribute data;
构建Mixed Logit模型，并将经过预处理的SP调查数据输入Mixed Logit模型，使用极大模拟似然估计算法对Mixed Logit模型进行参数估计；Construct a Mixed Logit model, input the preprocessed SP survey data into the Mixed Logit model, and use the maximum simulation likelihood estimation algorithm to estimate the parameters of the Mixed Logit model;
对参数估计结果进行评估，并根据评估结果判断是否需要对集成出行策略进行优化。Evaluate the parameter estimation results and determine whether the integrated travel strategy needs to be optimized based on the evaluation results.
进一步的，构建效用函数，效用函数通过以下公式表示：Further, a utility function is constructed, which is expressed by the following formula:
； ;
式中，Vntj表示观测的效用，n表示出行者，t表示选择情境，j表示集成出行策略选项，Z表示出行者个人属性变量矩阵，Y表示出行行为属性矩阵，X表示与集成出行策略有关的属性矩阵，γ、φ和β表示待估计的参数，εntj表示误差项，服从独立相同的Gumbel分布；In the formula,Vntj represents the utility of observation,n represents the traveler,t represents the choice situation,j represents the integrated travel strategy option,Z represents the traveler’s personal attribute variable matrix,Y represents the travel behavior attribute matrix, andX represents the integrated travel strategy. The attribute matrix of ,γ ,φ andβ represent the parameters to be estimated,εntj represents the error term, and obeys the independent and identical Gumbel distribution;
根据效用函数，计算受访者对集成出行策略的选择概率，选择概率通过以下公式表示：According to the utility function, the respondent's selection probability for the integrated travel strategy is calculated. The selection probability is expressed by the following formula:
； ;
求解Mixed Logit模型的对数似然函数，对数似然函数通过如下公式表示：Solve the log-likelihood function of the Mixed Logit model. The log-likelihood function is expressed by the following formula:
； ;
其中，如果在选择情境t下，选项i被受访者n选择，则等于1，否则/>等于0；Among them, if optioni is selected by respondentn in selection situationt , then Equal to 1, otherwise/> equal to 0;
对对数似然函数求解，获得集成出行策略中不同类型交通用具的使用额度的参数估计值以及相应的P值，当P值大于预设阈值时，判断参数估计值在置信区间上显著，当P值小于或等于预设阈值时，判断需要对集成出行策略中与该P值对应的交通用具的使用额度进行优化。Solve the log-likelihood function to obtain the parameter estimates of the usage quotas of different types of transportation equipment in the integrated travel strategy and the correspondingP values. Whenthe P value is greater than the preset threshold, it is judged that the parameter estimates are significant in the confidence interval. When Whenthe P value is less than or equal to the preset threshold, it is determined that the usage amount of the transportation equipment corresponding to theP value in the integrated travel strategy needs to be optimized.
本申请还提供一种集成出行策略设计系统，系统采用如上所述的集成出行策略的设计方法。This application also provides an integrated travel strategy design system, which adopts the integrated travel strategy design method as described above.
综上，本申请提供一种集成出行策略的设计方法，首先根据RP出行调查数据进行聚类，再根据聚类结果采用D-efficient设计，针对不同类型用户生成集成出行策略，用户可以根据自身的偏好、需求、行为习惯等因素，决定使用哪一种类的出行策略，从而可以满足不同用户在一定时间跨度上的出行需求，降低出行成本，具有可以引导公共交通出行，降低碳排放等有益效果，能够促进可持续出行，助力交通领域的碳减排。In summary, this application provides a design method for integrated travel strategies. First, clustering is performed based on RP travel survey data, and then D-efficient design is used based on the clustering results to generate integrated travel strategies for different types of users. Users can according to their own needs Preferences, needs, behavioral habits and other factors determine which type of travel strategy to use, which can meet the travel needs of different users within a certain time span, reduce travel costs, and have beneficial effects such as guiding public transportation travel and reducing carbon emissions. It can promote sustainable travel and help reduce carbon emissions in the transportation sector.
附图说明Description of the drawings
图1为本申请一种实施例提供的集成出行策略设计方法流程图；Figure 1 is a flow chart of an integrated travel strategy design method provided by an embodiment of the present application;
图2为本申请一种实施例提供的选择情境示意图；Figure 2 is a schematic diagram of a selection scenario provided by an embodiment of the present application;
图3为本申请实施例提供的采用K-medoids聚类算法对RP出行调查数据进行中心点聚类的流程图；Figure 3 is a flow chart of center point clustering of RP travel survey data using the K-medoids clustering algorithm provided by the embodiment of the present application;
图4为本申请另一种实施例提供的选择情境示意图；Figure 4 is a schematic diagram of a selection scenario provided by another embodiment of the present application;
图5为本申请另一种实施例提供的的集成出行策略设计方法流程图。Figure 5 is a flow chart of an integrated travel strategy design method provided by another embodiment of the present application.
具体实施方式Detailed ways
以下将结合附图所示的具体实施方式对本申请进行详细描述，但这些实施方式并不限制本申请，本领域的普通技术人员根据这些实施方式所做出的结构、方法、或功能上的变换均包含在本申请的保护范围内。The present application will be described in detail below with reference to the specific embodiments shown in the drawings. However, these embodiments do not limit the present application. Those of ordinary skill in the art may make structural, method, or functional changes based on these embodiments. are included in the protection scope of this application.
如图1所示，本申请实施例提供一种集成出行策略的设计方法，包括以下步骤：As shown in Figure 1, this embodiment of the present application provides a design method for an integrated travel strategy, which includes the following steps:
S1、获取RP（Revealed preference，行为调查）出行调查数据，RP出行调查数据包括出行起点、出行终点、出行时长以及出行方式；S1. Obtain RP (Revealed preference, behavioral survey) travel survey data. RP travel survey data includes travel starting point, trip end point, travel duration and travel mode;
S2、对RP出行调查数据进行中心点聚类，获得聚类结果，聚类结果包括不同类型用户对任一类型交通方式的搭配范围；S2. Perform center point clustering on the RP travel survey data to obtain clustering results. The clustering results include the matching range of any type of transportation mode for different types of users;
S3、根据聚类结果，采用D-efficient设计，在各个类型用户对任一类型交通方式的搭配范围内选择一数值水平作为所述类型交通方式的额度水平，以生成选择情境，选择情境中包括至少两个集成出行策略，任一集成出行策略包括至少两种类型交通用具的使用额度。S3. Based on the clustering results, use D-efficient design to select a numerical level within the range of each type of user's matching of any type of transportation mode as the credit level of the type of transportation mode to generate a selection scenario, which includes There are at least two integrated travel strategies, and any integrated travel strategy includes usage quotas for at least two types of transportation equipment.
作为一种可选的实现方式，在本申请实施例中，针对不同类型的交通工具，可以设置不同的使用额度，集成出行策略至少包括以下类型交通工具使用额度的一种或多种：公共交通的可用次数、网约车总可用里程、网约车里程单价、租赁式交通工具的可用时长。As an optional implementation manner, in the embodiment of this application, different usage quotas can be set for different types of transportation vehicles. The integrated travel strategy at least includes one or more of the following types of transportation usage quotas: Public transportation The number of available times, the total available mileage of online car-hailing, the unit price of online car-hailing mileage, and the available time of rental transportation.
如图2所示，其示例性的示出了本申请实施例设计的选择情境。在本申请实施例中，设计的选择情境包括了三种集成出行策略，每种出行策略中包括不同类型交通工具的使用额度。例如，策略A被配置为用户购买策略A所需支付金额为230元。在购买策略A后，用户可以在策略A的有效期限内任意支配策略A所包含的各类交通工具的使用额度。即，在策略A的有效期限内，购买者可以按照策略内容，随意搭乘公共交通工具两次，可用网约车总里程110千米，可在1小时内随意使用自行车，在任一交通工具的使用额度内使用该种交通工具，均不会产生额外费用。关于策略的有效期限，可以根据实际需求进行设置，例如，可以将策略的有效期限设置为一周或一月，可以适用于不同类型人群的工作通勤、社交就医、休闲旅游等出行需求，使出行者可以在一个应用上享受多模式的出行服务，解决了频繁切换应用与支付方式纷杂的困扰，提升了出行的便捷性、幸福感。As shown in Figure 2, it exemplarily shows the selection scenario designed in the embodiment of the present application. In the embodiment of this application, the designed selection scenario includes three integrated travel strategies, and each travel strategy includes usage quotas for different types of transportation. For example, strategy A is configured such that the amount required for users to purchase strategy A is 230 yuan. After purchasing strategy A, the user can freely control the usage quota of various types of transportation included in strategy A within the validity period of strategy A. That is, within the validity period of strategy A, the buyer can take public transportation twice according to the content of the strategy, the total mileage of available online car-hailing is 110 kilometers, and can use bicycles at will within 1 hour. There will be no additional charges for using this type of transportation within the quota. The validity period of the policy can be set according to actual needs. For example, the validity period of the policy can be set to one week or one month, which can be applied to the travel needs of different types of people such as work commuting, social medical treatment, leisure travel, etc., so that travelers can You can enjoy multi-mode travel services on one application, which solves the problem of frequent switching of applications and complicated payment methods, and improves the convenience and happiness of travel.
根据以上说明，本申请提供一种集成出行策略的设计方法，首先根据RP出行调查数据进行聚类，再根据聚类结果采用D-efficient设计，针对不同类型用户生成集成出行策略，用户可以根据自身的偏好、需求、行为习惯等因素，决定购买哪一种类的出行策略，从而可以满足不同用户在一定时间跨度上的出行需求，降低出行成本，具有可以引导公共交通出行，降低碳排放等有益效果，能够促进可持续出行，助力交通领域的碳减排。Based on the above description, this application provides a design method for integrated travel strategies. First, clustering is performed based on RP travel survey data, and then D-efficient design is used based on the clustering results to generate integrated travel strategies for different types of users. Users can Preferences, needs, behavioral habits and other factors determine which type of travel strategy to purchase, thereby meeting the travel needs of different users within a certain time span, reducing travel costs, and having beneficial effects such as guiding public transportation travel and reducing carbon emissions. , can promote sustainable travel and help reduce carbon emissions in the transportation sector.
作为一种可选的实现方式，在步骤S1中，RP出行调查数据可以从目标城市的出行数据库获取。在本申请实施例中，从目标城市的出行数据库获取一周内的居民出行调查数据。主要包括出行起点、出行终点、出行方式以及出行时间等属性，样例数据如表1所示。As an optional implementation method, in step S1, RP travel survey data can be obtained from the travel database of the target city. In the embodiment of this application, the resident travel survey data within a week is obtained from the travel database of the target city. It mainly includes attributes such as travel starting point, travel end point, travel mode, and travel time. Sample data is shown in Table 1.
表1Table 1
如图3所示，在步骤S2中，可以采用K-medoids聚类算法对RP出行调查数据进行中心点聚类。具体的，可以包括以下步骤：As shown in Figure 3, in step S2, the K-medoids clustering algorithm can be used to perform center point clustering on the RP travel survey data. Specifically, it may include the following steps:
S21、对RP出行调查数据进行预处理以构建数据集，数据集中包括各个出行者的多项属性数据，属性数据作为样本点的坐标。出行者的属性数据包括该出行者在预设时间范围内发生的公共交通行程次数、小汽车行程距离和自行车行程时长；S21. Preprocess the RP travel survey data to construct a data set. The data set includes multiple attribute data of each traveler, and the attribute data is used as the coordinates of the sample points. The traveler's attribute data includes the number of public transportation trips, car trip distance and bicycle trip duration that the traveler took within the preset time range;
S22、随机选择k个样本点作为中心点，计算剩余样本点到任一中心点的欧式距离；S22. Randomly selectk sample points as center points, and calculate the Euclidean distance from the remaining sample points to any center point;
S23、根据计算出的欧氏距离，按照与就近中心的原则，将剩余的样本点分配到各个中心点所代表的类别中，实现初始聚类；S23. Based on the calculated Euclidean distance and the principle of the nearest center, allocate the remaining sample points to the categories represented by each center point to achieve initial clustering;
S24、重新计算聚类中心，遍历类别内的每一个样本点，计算样本点到该类别剩余样本点的欧氏距离，选取欧氏距离最小时所对应的组，将其作为新的中心点；S24. Recalculate the cluster center, traverse each sample point in the category, calculate the Euclidean distance from the sample point to the remaining sample points in the category, select the group corresponding to the smallest Euclidean distance, and use it as the new center point;
S25、根据新的中心点重新进行聚类划分，直至所有的中心点不再发生变化或达到最大迭代次数，以获得当前k种类别的聚类结果。S25. Re-divide the clusters based on the new center points until all center points no longer change or the maximum number of iterations is reached to obtain the current k categories of clustering results.
作为一种可选的实现方式，在步骤S22中，基于目标城市出行者的属性数据进行K-medoids聚类，随机选择k个样本点作为中心点，在本申请实施例中，k的取值范围设置为大于1且小于或等于10,计算剩余样本点到任一中心点的欧氏距离，欧式距离可以通过如下公式（1）计算获得：As an optional implementation method, in step S22, K-medoids clustering is performed based on the attribute data of travelers in the target city, and k sample points are randomly selected as center points. In the embodiment of this application, the value of k Set the range to greater than 1 and less than or equal to 10, and calculate the Euclidean distance from the remaining sample points to any center point. The Euclidean distance can be calculated by the following formula (1):
（1） (1)
式中：di为第i个样本点到medoids的欧氏距离；xij为数据集第i行，第j列的数，xmj为第m个中心点的第j列数据；数据集的行可以代表出行者，列可以代表提取的属性数据。In the formula:di is the Euclidean distance from the i-th sample point to medoids;xij is the number in thei -th row andj -th column of the data set;xmj is thej -th column data ofthe m -th center point; Rows can represent travelers, and columns can represent extracted attribute data.
作为一种可选的实现方式，采用K-medoids聚类算法对RP出行调查数据进行聚类时，聚类的类别个数K通过以下公式（2）确定：As an optional implementation method, when using the K-medoids clustering algorithm to cluster RP travel survey data, the number of clustering categories K is determined by the following formula (2):
（2） (2)
式中，Ck为第k类别的数据，s为Ck中的数据，ok为Ck的质心。In the formula,Ck is the data of thekth category,s is the data inCk , andok is the centroid ofCk .
取SSE下降趋势转折点的K值为聚类的类别个数，聚类中每一类别作为同一类型的用户。The K value of the turning point ofthe SSE downward trend is taken as the number of clustering categories, and each category in the cluster is regarded as the same type of user.
在步骤S25中，根据k的最优取值进行聚类，并得到最终的多模式组合聚类结果，如表2所示，其示例性的示出了可进一步应用的聚类结果。In step S25, clustering is performed based on the optimal value of k, and the final multi-mode combined clustering result is obtained, as shown in Table 2, which exemplarily shows clustering results that can be further applied.
表2Table 2
如表2所示，本申请实施例中对RP出行调查数据进行中心聚类，将出行者分为3类用户，分别为公共交通用户、汽车用户以及多模式用户。在本申请实施例中，RP出行调查数据为从目标城市的出行数据库获取的一周内居民出行调查数据。对RP出行调查数据进行聚类，将一周内搭乘公共交通次数的搭配范围在[10，14]内，且乘坐小汽车里程的搭配范围在[0km，45km]内，且使用自行车时长的搭配范围在[4h，7h]内的用户归类为公共交通用户。将一周内搭乘公共交通次数的搭配范围在[0，5]内，且乘坐小汽车里程的搭配范围在[90km，150km]内，且使用自行车时长的搭配范围在[0h，3h]内的用户归类为汽车用户。将一周内搭乘公共交通次数的搭配范围在[6，9]内，且乘坐小汽车里程的搭配范围在[50km，100km]内，且使用自行车时长的搭配范围在[2h，5h]内的用户归类为多模式用户。需要说明的是，公共交通次数取值为区间内的整数。As shown in Table 2, in the embodiment of this application, the RP travel survey data is centrally clustered, and travelers are divided into three types of users, namely public transportation users, car users, and multi-modal users. In the embodiment of this application, the RP travel survey data is the resident travel survey data within a week obtained from the travel database of the target city. The RP travel survey data is clustered, and the matching range of the number of public transportation rides in a week is within [10, 14], the matching range of the mileage of taking a car is within [0km, 45km], and the matching range of the duration of using a bicycle is Users within [4h, 7h] are classified as public transport users. Users whose matching range of the number of public transportation rides in a week is within [0, 5], and the matching range of car mileage is within [90km, 150km], and the matching range of bicycle riding time is within [0h, 3h] Classified as a car user. Users whose matching range of the number of public transportation rides in a week is within [6, 9], and the matching range of car mileage is within [50km, 100km], and the matching range of bicycle riding time is within [2h, 5h] Classified as a multimodal user. It should be noted that the number of public transportation times is an integer within the interval.
在步骤S3中，根据聚类结果，采用D-efficient设计，针对不同类型用户生成集成出行策略。D-efficient设计是一种用于离散选择实验的设计算法。可以使用Ngene软件编写代码，采取D-efficient设计的方法进行离散选择实验情境设计，生成的情境数量可以根据实际需求进行设置，一般4-8个情境即可。In step S3, based on the clustering results, D-efficient design is used to generate integrated travel strategies for different types of users. D-efficient design is a design algorithm for discrete choice experiments. You can use Ngene software to write code and adopt the D-efficient design method to design discrete choice experimental scenarios. The number of generated scenarios can be set according to actual needs, generally 4-8 scenarios are enough.
步骤S3具体包括以下步骤：Step S3 specifically includes the following steps:
S31、在满足水平均衡、正交性和最小重叠性的要求下，根据聚类结果，在数值水平搭配范围内选择应用的数值水平，以生成初始的设计矩阵X。其中，设计矩阵X即表示本申请实施例设计的一种选择情景。S31. Under the condition of meeting the requirements of horizontal balance, orthogonality and minimum overlap, according to the clustering results, select the applied numerical level within the numerical level matching range to generate the initial design matrixX. Among them, the design matrix X represents a selection scenario for the design of the embodiment of the present application.
水平均衡要求属性水平在数据集中出现的机会均等，正交性要求属性水平的组合存在特定的相关模式，最小的水平重叠追求在同一组试验场景中尽可能减少属性水平的重复概率，以确保属性重要性估计的有效性。Horizontal balance requires that attribute levels have equal opportunities to appear in the data set. Orthogonality requires that the combination of attribute levels has a specific correlation pattern. Minimal horizontal overlap pursues the goal of reducing the repetition probability of attribute levels in the same set of test scenarios as much as possible to ensure that attributes Validity of importance estimates.
具体的，根据表2的聚类结果可以得出公共交通次数、小汽车千米数以及自行车小时数之间的数值水平搭配范围，在进行出行策略设计时可在搭配范围内选取应用的数值水平。例如选取公共交通次数取值范围内的整数，小汽车千米数取值范围内选取预设千米数的整数倍，自行车小时数取值范围内的预设时间的倍数。为了便于计算，本申请实施例中预设千米数设置为5或10，预设时间设置为0.5。Specifically, according to the clustering results in Table 2, the range of numerical level matching between the number of public transportation, the number of kilometers of cars and the number of bicycle hours can be obtained. When designing travel strategies, the applied numerical level can be selected within the matching range. . For example, select an integer within the value range of the number of public transportation times, select an integer multiple of the preset number of kilometers within the value range of the number of kilometers for cars, and select a multiple of the preset time within the value range of the number of bicycle hours. In order to facilitate calculation, in the embodiment of this application, the preset number of kilometers is set to 5 or 10, and the preset time is set to 0.5.
作为一种可选的实现方式，本申请实施例中还可以在集成出行策略中加入出租车折扣（10%，20%）。如图4所示，其示出了本申请实施例提供的一种选择情境下的出行策略选项。每一选择情境可以包括两个定制出行策略、一个随用随付出行策略以及不订阅选项。As an optional implementation method, in the embodiment of this application, taxi discounts (10%, 20%) can also be added to the integrated travel strategy. As shown in Figure 4, it shows a travel strategy option in a selection scenario provided by the embodiment of the present application. Each choice scenario can include two customized travel strategies, a pay-as-you-go travel strategy, and a no-subscription option.
S32、根据设计矩阵X和关于参数估计的一些先验概率分布信息来确定渐进方差-协方差（AVC）矩阵，来预测当前设计情境X下各个出行方式水平搭配设计所得数据产生的参数估计误差。S32 . Determine the asymptotic variance-covariance (AVC) matrix based on the designmatrix
作为一种可选的实现方式，设计的效率可以基于进行衡量；以最小为目标，采用Modified Federov算法对问题进行求解，迭代寻找近似最优可行设计。/>计算方法如下所示：As an optional implementation, the efficiency of the design can be based on to measure; to Minimum is the goal, and the Modified Federov algorithm is used to solve the problem and iteratively find the approximately optimal feasible design. /> The calculation method is as follows:
（3） (3)
式中：P是待估计参数的数量，代表单个受访者的渐进方差-协方差（AVC）矩阵，AVC矩阵是一个P*P矩阵, 其是设计矩阵X和先验概率分布信息/>的函数，/>可以从相似研究或预调查中获取，为了使/>与问题的大小无关，其被归一化为1/P次幂。where:P is the number of parameters to be estimated, Represents the asymptotic variance-covariance (AVC) matrix of a single respondent. The AVC matrix is a P*P matrix, which is the design matrixX and the prior probability distribution information/> function,/> Can be obtained from similar studies or pre-surveys, in order to make/> Regardless of the size of the problem, it is normalized to the power 1/P .
根据以上说明，本申请实施例提供的集成出行策略设计方法可以针对不同类型用户生成集成出行策略，用户可以根据自身的偏好、需求、行为习惯等因素，决定购买哪一种类的出行策略，从而可以满足不同用户在一定时间跨度上的出行需求，降低出行成本，具有可以引导公共交通出行，降低碳排放等有益效果，能够促进可持续出行，助力交通领域的碳减排。According to the above description, the integrated travel strategy design method provided by the embodiments of the present application can generate integrated travel strategies for different types of users. Users can decide which type of travel strategy to purchase based on their own preferences, needs, behavioral habits and other factors, so that they can It can meet the travel needs of different users within a certain time span and reduce travel costs. It can guide public transportation and reduce carbon emissions. It can promote sustainable travel and help reduce carbon emissions in the transportation field.
如图5所示，作为一种可选的实现方式，本申请实施例提供的集成出行策略设计方法还包括：As shown in Figure 5, as an optional implementation method, the integrated travel strategy design method provided by the embodiment of this application also includes:
S4、进行SP调查数据收集，获得SP调查数据，并对SP调查数据进行预处理，SP调查数据包括：出行者个人属性数据、出行行为属性数据和集成出行策略属性数据；S4. Collect SP survey data, obtain SP survey data, and preprocess the SP survey data. The SP survey data includes: traveler personal attribute data, travel behavior attribute data, and integrated travel strategy attribute data;
S5、构建Mixed Logit模型，并将经过预处理的SP调查数据输入Mixed Logit模型，使用极大模拟似然估计算法对Mixed Logit模型进行参数估计；S5. Construct a Mixed Logit model, input the preprocessed SP survey data into the Mixed Logit model, and use the maximum simulation likelihood estimation algorithm to estimate the parameters of the Mixed Logit model;
S6、对参数估计结果进行评估，并根据评估结果判断是否需要对集成出行策略进行优化。S6. Evaluate the parameter estimation results, and determine whether the integrated travel strategy needs to be optimized based on the evaluation results.
作为一种可选的实现方式，步骤S4中，出行者个人属性数据包括：性别、年龄、学历、职业、家庭结构、家庭小汽车拥有量、家庭月收入、日常和旅游中的主要出行方式。As an optional implementation method, in step S4, the traveler's personal attribute data includes: gender, age, education, occupation, family structure, family car ownership, family monthly income, and main modes of travel in daily life and tourism.
本申请实施例中，按照以下年龄范围对出行者的年龄属性数据进行划分，年龄范围包括：18岁以下，18-30岁，31-50岁，51-65岁，65岁以上。出行者的学历分为4类，包括：高中及以下，大学专科/职业技术学院，大学本科，硕士及以上。出行者的职业至少分为6类，包括：公务员/企事业单位人员，服务人员，个体经营者，学生，无业（包括家庭主妇、退休人员等），其它。家庭月总收入分为5个等级：5000元以下，5000-9999元，10000-29999元，30000-49999元，50000元及以上。家庭小汽车拥有量分为4个等级：0辆，1辆，2辆，3辆及以上。In the embodiment of this application, the age attribute data of travelers is divided according to the following age ranges, which include: under 18 years old, 18-30 years old, 31-50 years old, 51-65 years old, and over 65 years old. The academic qualifications of travelers are divided into 4 categories, including: high school and below, college/vocational technical college, bachelor's degree, and master's degree and above. The occupations of travelers are divided into at least 6 categories, including: civil servants/enterprise and institution personnel, service personnel, self-employed, students, unemployed (including housewives, retirees, etc.), and others. Total monthly household income is divided into five levels: less than 5,000 yuan, 5,000-9,999 yuan, 10,000-29,999 yuan, 30,000-49,999 yuan, and 50,000 yuan and above. Family car ownership is divided into four levels: 0 cars, 1 car, 2 cars, 3 cars and above.
SP调查数据中出行行为属性数据包括：近一周的出行数据（包括出行方式、出行距离等和日常中的主要出行方式以及各方式的使用频率，主要包含以下9类出行方式的数据：私家车、公交、地铁、出租车/网约车、拼车、共享汽车、自行车、步行以及其它。The travel behavior attribute data in the SP survey data includes: travel data in the past week (including travel mode, travel distance, etc., as well as the main daily travel modes and the frequency of use of each mode). It mainly includes data on the following nine types of travel modes: private cars, Bus, subway, taxi/ride-hailing, ride-sharing, car-sharing, bicycle, walking and others.
SP调查数据中集成出行策略属性数据：包括可享用的公共交通次数和网约车里程、网约车里程单价、出租车折扣、共享单车小时数以及出行策略价格。The SP survey data integrates travel strategy attribute data: including the number of public transportation available and ride-hailing mileage, online ride-hailing mileage unit price, taxi discounts, shared bicycle hours, and travel strategy prices.
步骤S4中，还需要对收集的数据进行预处理，方法包括：剔除填写调查问卷时间少于一定值的数据，剔除家庭成员个数与家庭结构不相符合的数据，剔除选择情境方差较小的数据。通过这种方式，可以初步剔除SP调查数据中的异常数据，避免对评估结果产生干扰。In step S4, the collected data also need to be preprocessed. The methods include: eliminating data whose time to fill in the questionnaire is less than a certain value, eliminating data whose number of family members does not match the family structure, and eliminating data whose selection situation variance is small. data. In this way, abnormal data in the SP survey data can be initially eliminated to avoid interference with the evaluation results.
步骤S5中，对Mixed Logit模型进行参数估计包括以下步骤：In step S5, parameter estimation of the Mixed Logit model includes the following steps:
构建效用函数，效用函数通过以下公式(4)表示：Construct a utility function, which is expressed by the following formula (4):
(4) (4)
式中，Vntj表示观测的效用，n表示出行者，t表示选择情境，j表示集成出行策略选项，Z表示出行者个人属性变量矩阵，Y表示出行行为属性矩阵，X表示与集成出行策略有关的属性矩阵，γ、φ和β表示待估计的参数，εntj表示误差项，服从独立相同的Gumbel分布。In the formula,Vntj represents the utility of observation,n represents the traveler,t represents the choice situation,j represents the integrated travel strategy option,Z represents the traveler’s personal attribute variable matrix,Y represents the travel behavior attribute matrix, andX represents the integrated travel strategy. The attribute matrix of ,γ ,φ andβ represent the parameters to be estimated,εntj represents the error term, and obeys the independent and identical Gumbel distribution.
其中，选择情境是指采取D-efficient设计的方法进行离散选择实验情境设计生成的选择情境，选择情境的数量可以根据实际需求进行设置，一般4-8个情境即可。集成出行策略选项是指出行者n在图2中的出行策略A、出行策略B、随用随付和不订阅中的选择结果。Among them, the selection situation refers to the selection situation generated by discrete selection experimental situation design using the D-efficient design method. The number of selection situations can be set according to actual needs, generally 4-8 situations are enough. The integrated travel strategy option refers to the choice result of travelern among travel strategy A, travel strategy B, pay-as-you-go and no subscription in Figure 2.
作为一种可选的实现方式，可以将部分设为随机参数，以扑捉出行者对于集成出行策略的偏好异质性，/>的分布形式可以通过如下公式(5)表示：As an optional implementation, some Set as a random parameter to capture the heterogeneity of travelers’ preferences for integrated travel strategies,/> The distribution form of can be expressed by the following formula (5):
(5) (5)
式中，表示与属性k（某一个集成出行策略属性）相关的参数/>的样本平均值，/>是平均值为0、标准差为1的正态分布随机项，/>是/>的分布标准差。In the formula, Indicates parameters related to attributek (an integrated travel strategy attribute)/> The sample mean of ,/> is a normally distributed random item with a mean of 0 and a standard deviation of 1,/> Yes/> The standard deviation of the distribution.
根据公式(4)的效用函数，计算受访者对集成出行策略的选择概率，选择概率通过以下公式(6)表示：According to the utility function of formula (4), the respondent's selection probability of the integrated travel strategy is calculated. The selection probability is expressed by the following formula (6):
(6) (6)
将SP调查数据转换为可用Nlogit软件进行计算分析的格式，包括将分类变量进行有效编码。为更好地研究SP调查得到的伪面板数据与受访者的异质性，基于极大模拟似然估计方法进行Mixed Logit模型参数估计，使用Halton抽样方法对概率密度函数进行1000次抽样并将模拟概率均值作为积分的近似解。Convert SP survey data into a format that can be used for computational analysis with Nlogit software, including efficient coding of categorical variables. In order to better study the pseudo-panel data obtained from the SP survey and the heterogeneity of the respondents, the parameters of the Mixed Logit model were estimated based on the maximum simulation likelihood estimation method, and the Halton sampling method was used to sample the probability density function 1000 times and Simulate probability means as approximate solutions to integrals.
求解Mixed Logit模型的对数似然函数，对数似然函数通过如下公式(7)表示：Solve the log-likelihood function of the Mixed Logit model. The log-likelihood function is expressed by the following formula (7):
(7) (7)
其中，如果在选择情境t下，出行策略选项i被受访者n选择，则等于1，否则等于0。Among them, if in the choice situationt , the travel strategy optioni is selected by the respondentn , then equals 1, otherwise equal to 0.
对对数似然函数求解，获得模型的拟合度R2和个人属性、出行行为属性以及集成出行策略属性所对应的参数γ、φ和β以及各参数对应的P值。Solve the log-likelihood function to obtain the model's fitting degree R2 and the parametersγ ,φ andβ corresponding to personal attributes, travel behavior attributes and integrated travel strategy attributes, as well as theP values corresponding to each parameter.
步骤S6中，对参数估计结果进行评估，当P值大于预设阈值时，判断参数估计值在置信区间上显著，当P值小于或等于预设阈值时，判断需要对集成出行策略中与该P值对应的交通用具的使用额度进行优化。In step S6, the parameter estimation results are evaluated. Whenthe P value is greater than the preset threshold, it is judged that the parameter estimate is significant in the confidence interval. Whenthe P value is less than or equal to the preset threshold, it is judged that the integrated travel strategy and the The usage amount of transportation equipment corresponding tothe P value is optimized.
具体的，可以将以上步骤S5~S6在Nlogit中编写代码并运行，得出Mixed Logit模型的参数估计结果。对模型的参数估计结果进行评估，首先查看模型拟合度R2，其值在0.2及以上可认为模型对SP调查数据的拟合效果是良好的，这表明了影响因素选取的有效性与模型的解释程度。Specifically, the above steps S5~S6 can be written and run in Nlogit to obtain the parameter estimation results of the Mixed Logit model. To evaluate the parameter estimation results of the model, first check the model fitting degree R2 . If the value is 0.2 and above, it can be considered that the model has a good fitting effect on the SP survey data. This shows that the effectiveness of the selection of influencing factors is consistent with the model. degree of explanation.
查看各个影响因素的估计参数，数值的正负分别代表了对相应MaaS出行策略选择的正向与负向影响，通过P值可以看出影响的显著性，P<0.1、P<0.05、P<0.01分别表示在90%、95%、99%的置信水平下显著。Check the estimated parameters of each influencing factor. The positive and negative values represent the positive and negative impacts on the corresponding MaaS travel strategy selection respectively. The significance of the impact can be seen through theP value,P <0.1,P <0.05,P < 0.01 indicates significance at the 90%, 95%, and 99% confidence levels respectively.
通过模型的参数估计结果可以得出目前MaaS的应用潜力，识别出MaaS的目标人群画像和对出行策略内所含出行方式的偏好并计算出出行者对于MaaS出行策略的支付意愿，可以为相关的规划者、运营者在优化出行策略内包含的出行方式及其水平以及定价方面提供一些政策启示与理论支撑。Through the parameter estimation results of the model, we can derive the current application potential of MaaS, identify the target group portrait of MaaS and the preferences for the travel modes included in the travel strategy, and calculate the travelers' willingness to pay for the MaaS travel strategy, which can provide relevant Planners and operators provide some policy inspiration and theoretical support in optimizing the travel modes and levels included in the travel strategy, as well as pricing.
作为一种可选的实现方式，本申请实施例还提供一种集成出行策略设计系统，该系统采用本申请实施例提供的集成出行策略设计方法。As an optional implementation manner, embodiments of this application also provide an integrated travel strategy design system, which adopts the integrated travel strategy design method provided by embodiments of this application.
以上所揭露的仅为本申请的较佳实施例而已，然其并非用以限定本申请之权利范围，本领域普通技术人员可以理解：在不脱离本申请及所附的权利要求的精神和范围内，改变、修饰、替代、组合、简化，均应为等效的置换方式，仍属于发明所涵盖的范围。What is disclosed above is only the preferred embodiment of the present application, but it is not used to limit the scope of rights of the present application. Those of ordinary skill in the art can understand that: without departing from the spirit and scope of the present application and the appended claims, Within the scope of the invention, changes, modifications, substitutions, combinations, and simplifications should all be equivalent substitutions and still fall within the scope of the invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310910715.XACN116628527B (en) | 2023-07-24 | 2023-07-24 | Design method and system for integrated travel strategy |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310910715.XACN116628527B (en) | 2023-07-24 | 2023-07-24 | Design method and system for integrated travel strategy |
| Publication Number | Publication Date |
|---|---|
| CN116628527A CN116628527A (en) | 2023-08-22 |
| CN116628527Btrue CN116628527B (en) | 2023-11-10 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310910715.XAActiveCN116628527B (en) | 2023-07-24 | 2023-07-24 | Design method and system for integrated travel strategy |
| Country | Link |
|---|---|
| CN (1) | CN116628527B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108564396A (en)* | 2018-03-23 | 2018-09-21 | 同济大学 | Multimode traffic trip questionnaire survey design method based on D-error design effectivelies |
| CN110390415A (en)* | 2018-04-18 | 2019-10-29 | 北京嘀嘀无限科技发展有限公司 | A kind of method and system carrying out trip mode recommendation based on user's trip big data |
| CN111737601A (en)* | 2020-06-08 | 2020-10-02 | 北京奇虎科技有限公司 | Recommended method, device, device and storage medium for travel strategy |
| CN112632374A (en)* | 2020-12-18 | 2021-04-09 | 东南大学 | Resident travel mode selection analysis method considering customized bus |
| CN113269358A (en)* | 2021-05-19 | 2021-08-17 | 兆边(上海)科技有限公司 | Planning method based on multi-mode integrated travel |
| CN113780808A (en)* | 2021-09-10 | 2021-12-10 | 西南交通大学 | Vehicle service attribute decision optimization method based on flexible bus connection system line |
| CN115049217A (en)* | 2022-05-17 | 2022-09-13 | 中国平安财产保险股份有限公司 | Method and device for generating travel strategy, computer equipment and storage medium |
| CN115412857A (en)* | 2022-08-24 | 2022-11-29 | 浙江大学 | A method for predicting residents' travel information |
| WO2023273292A1 (en)* | 2021-06-30 | 2023-01-05 | 深圳市城市交通规划设计研究中心股份有限公司 | Resident trip chain generation method based on multi-source data fusion, and vehicle-sharing query method |
| CN116431988A (en)* | 2023-03-22 | 2023-07-14 | 浙江大学 | Resident trip activity time sequence generation method based on activity mode-Markov chain |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10649747B2 (en)* | 2015-10-07 | 2020-05-12 | Andreas Voellmy | Compilation and runtime methods for executing algorithmic packet processing programs on multi-table packet forwarding elements |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108564396A (en)* | 2018-03-23 | 2018-09-21 | 同济大学 | Multimode traffic trip questionnaire survey design method based on D-error design effectivelies |
| CN110390415A (en)* | 2018-04-18 | 2019-10-29 | 北京嘀嘀无限科技发展有限公司 | A kind of method and system carrying out trip mode recommendation based on user's trip big data |
| CN111737601A (en)* | 2020-06-08 | 2020-10-02 | 北京奇虎科技有限公司 | Recommended method, device, device and storage medium for travel strategy |
| CN112632374A (en)* | 2020-12-18 | 2021-04-09 | 东南大学 | Resident travel mode selection analysis method considering customized bus |
| CN113269358A (en)* | 2021-05-19 | 2021-08-17 | 兆边(上海)科技有限公司 | Planning method based on multi-mode integrated travel |
| WO2023273292A1 (en)* | 2021-06-30 | 2023-01-05 | 深圳市城市交通规划设计研究中心股份有限公司 | Resident trip chain generation method based on multi-source data fusion, and vehicle-sharing query method |
| CN113780808A (en)* | 2021-09-10 | 2021-12-10 | 西南交通大学 | Vehicle service attribute decision optimization method based on flexible bus connection system line |
| CN115049217A (en)* | 2022-05-17 | 2022-09-13 | 中国平安财产保险股份有限公司 | Method and device for generating travel strategy, computer equipment and storage medium |
| CN115412857A (en)* | 2022-08-24 | 2022-11-29 | 浙江大学 | A method for predicting residents' travel information |
| CN116431988A (en)* | 2023-03-22 | 2023-07-14 | 浙江大学 | Resident trip activity time sequence generation method based on activity mode-Markov chain |
| Title |
|---|
| Research on Travel Mode Choice Behavior Under Integrated Multi-modal Transit Information Service;HU Hua，J Teng，YF Gao，XM Zhou;《Transport Policy》;全文* |
| Transport decarbonization in big cities: An integrated environmental co-benefit analysis of vehicles purchases quota-limit and new energy vehicles promotion policy in Beijing;Xi Yang，Wanqi Lin，Cecilia Springer;《Sustainable Cities and Society》;全文* |
| Yilin Sun ; Zhiyi Huang and Kitamura, R..The Automobility Cohort as a Tool in the Study of Urban Travel.《 2011 International Conference on Electric Technology and Civil Engineering (ICETCE)》.2011,全文.* |
| 基于Two-step Cluste的出行选择行为研究;温惠英;任倩;;《公路工程》(第02期);全文* |
| 基于两阶段聚类的电动自行车出行者选择研究;胡郁葱;陈杰;邹小健;陈枝伟;;《广西师范大学学报(自然科学版)》(第03期);全文* |
| 基于聚类分析的铁路出行旅客类别划分;吕红霞;王文宪;蒲松;余大本;;《交通运输系统工程与信息》(第01期);全文* |
| 家庭收入差异对出行方式选择的影响分析;范琪;王炜;杨洋;华雪东;;《交通信息与安全》(第06期);全文* |
| 殷复莲.《数据分析与数据挖掘实用教程》.中国传媒大学出版社,2017,第249页.* |
| 活动―出行决策行为与TDM策略互动关系研究的贝叶斯方法;隽志才;宗芳;栾琨;;《交通运输系统工程与信息》(第04期);全文* |
| 钱兵等.《智能运维之道 基于AI技术的应用实践》.机械工业出版社,2021,第111-113页.* |
| Publication number | Publication date |
|---|---|
| CN116628527A (en) | 2023-08-22 |
| Publication | Publication Date | Title |
|---|---|---|
| Jin et al. | Mode choice analysis in urban transport with shared battery electric vehicles: A stated-preference case study in Beijing, China | |
| Circella et al. | The adoption of shared mobility in California and its relationship with other components of travel behavior | |
| Fang et al. | Exploring satisfaction of choice and captive bus riders: An impact asymmetry analysis | |
| Li et al. | Improving service quality with the fuzzy TOPSIS method: a case study of the Beijing rail transit system | |
| Ye et al. | Competition or cooperation: Relationship between carsharing and other travel modes | |
| CN116070033A (en) | Novel shared public transportation transfer demand estimation method based on mobile phone signaling data | |
| Tao et al. | Research on travel behavior with car sharing under smart city conditions | |
| Meng et al. | Modeling heterogeneous parking choice behavior on university campuses | |
| Deneke et al. | Transportation mode choice behavior with multinomial logit model: work and school trips | |
| Guzman et al. | Exploring the effects of public transport subsidies on satisfaction and ridership | |
| CN108564396A (en) | Multimode traffic trip questionnaire survey design method based on D-error design effectivelies | |
| Alemi | What Makes Travelers Use Ridehailing? Exploring the Latent Constructs behind the Adoption and Frequency of Use of Ridehailing Services, and Their Impacts on the Use of Other Travel Modes | |
| Ngcobo et al. | Evaluating the measures to promote sustainable transport infrastructure: A case of city of Johannesburg, South Africa | |
| CN116628527B (en) | Design method and system for integrated travel strategy | |
| Rajagopalan et al. | Integrating household-level mode choice and modal expenditure decisions in a developing country: multiple discrete–continuous extreme value model | |
| Luo et al. | Understanding the behavioral intentions of MaaS during public health events | |
| Rodier et al. | A before and after evaluation of shared mobility projects in the San Joaquin Valley | |
| Roukouni et al. | Mind the Gap: A Comparative Study of Low-Car Policy Acceptance | |
| Ong | The Adoption, Use, and Impacts of Ride-sourcing Services in the Metro Vancouver Area | |
| Trivedi et al. | Application-based cab services in India: Commuters’ barriers due to COVID-19 | |
| Huang et al. | Does the information provider choose to promote public transport use or protect public health? A case study in China | |
| Rui | Modeling route choice behaviour in public transport network | |
| Dong | A Philadelphia ride-share story: An Investigation of ride-share’s impact on transit | |
| Firmansyah et al. | Exploring the transition to electric motorcycles and its role in multimodal travel in the Jakarta metropolitan area | |
| Mapfurira | Modelling public transport mode choice for low-income residential suburbs in Harare, Zimbabwe |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |