Travel service recommendation method and device, electronic equipment and storage medium

Abstract

The application provides a travel service recommendation method and device, an electronic device and a storage medium. The method comprises the following steps: performing regional clustering based on station position information of a traffic station to obtain a plurality of candidate travel region pairs, and constructing a candidate travel chain; obtaining first historical travel data in a preset historical period, and extracting corresponding second historical travel data for a preset operation time interval; for each candidate travel region pair, determining a maximum number of transfers, and obtaining a target historical passenger flow in a continuous historical time interval; combining a preset screening condition to screen a target travel region pair; extracting third historical travel data, generating a travel inappropriateness evaluation index of each operation time interval based on the first and third historical travel data for each target travel region pair, and recommending a travel service to a travel object according to the travel inappropriateness evaluation index. The application can provide personalized traffic services for travel objects with the same travel demand and provide a good travel experience.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation technology, and in particular to a method, apparatus, electronic device and storage medium for recommending travel services. Background Technology

[0002] With the acceleration of urbanization and the continuous expansion of urban space, the separation of work and residence has become increasingly prominent, and daily commuting has become an important part of urban residents' daily lives. As the artery of urban operation, public transportation carries the function of large-scale passenger travel, and the coverage of its service network and the transportation organization mode are directly related to travel efficiency and comfort.

[0003] The travel services in related technologies rely on passenger flow statistics at fixed stations for network planning and supply-demand matching. This method can meet large-scale travel demand, but it is difficult to fully reflect the real travel difficulties of passengers, resulting in a poor travel experience for passengers.

[0004] Therefore, how to provide personalized transportation services for travelers with the same travel needs, thereby providing a good travel experience, has become an urgent problem to be solved. Summary of the Invention

[0005] The main objective of this application is to provide a travel service recommendation method, apparatus, electronic device, and storage medium, which aims to provide personalized transportation services for travelers with the same travel needs, thereby providing a better travel experience.

[0006] To achieve the above objectives, a first aspect of this application proposes a travel service recommendation method, the method comprising:

[0007] Based on the location information of transportation stations in the target area, regional clustering is performed to obtain multiple candidate travel area pairs, and a candidate travel chain is constructed according to the starting station set and ending station set of each candidate travel area pair.

[0008] Obtain first historical travel data of the target area in a preset historical period. For each preset operating time interval, extract second historical travel data of multiple historical time intervals that are the same as the operating time interval from the first historical travel data. The second historical travel data represents the data of the travel object entering and exiting from the stations in the set of starting stations within the historical time interval and belonging to the corresponding set of ending stations.

[0009] For each candidate travel region pair, determine the maximum number of transfers for the candidate travel chain, and obtain the target historical passenger flow within a preset number of consecutive historical time intervals from the second historical travel data;

[0010] Based on the maximum number of transfers, the target historical passenger flow, and preset filtering conditions, a target travel region pair is obtained by filtering from multiple candidate travel region pairs.

[0011] Extract third historical travel data from the second historical travel data. The third historical travel data represents the data of stations entering and exiting from the target origin station set of the target travel area within the historical time interval, which belong to the corresponding target end station set.

[0012] For each target travel region pair, a travel inappropriateness evaluation index is generated for each operating time interval based on the first historical travel data and the third historical travel data, and travel services are recommended to the travel target based on the travel inappropriateness evaluation index.

[0013] To achieve the above objectives, a second aspect of this application provides a travel service recommendation device, the device comprising:

[0014] The construction unit is used to perform regional clustering based on the station location information of transportation stations in the target area to obtain multiple candidate travel area pairs, and to construct a candidate travel chain based on the starting station set and ending station set of each candidate travel area pair.

[0015] The first extraction unit is used to obtain the first historical travel data of the target area in a preset historical period, and for each preset operating time interval, extract the second historical travel data of multiple historical time intervals that are the same as the operating time interval from the first historical travel data. The second historical travel data represents the data of the travel object entering and exiting the station from the station in the starting station set within the historical time interval and the station belonging to the corresponding ending station set.

[0016] The acquisition unit is used to determine the maximum number of transfers of the candidate travel chain for each candidate travel area pair, and to acquire the target historical passenger flow within a preset number of consecutive historical time intervals from the second historical travel data.

[0017] The filtering unit is used to filter out a target travel area pair from multiple candidate travel area pairs based on the maximum number of transfers, the target historical passenger flow, and preset filtering conditions.

[0018] The second extraction unit is used to extract third historical travel data from the second historical travel data. The third historical travel data represents the data of stations entering and exiting from the target starting station set of the target travel area within the historical time interval, which belong to the corresponding target ending station set.

[0019] The recommendation unit is used to generate a travel inadequacy evaluation index for each of the target travel area pairs based on the first historical travel data and the third historical travel data, and to recommend travel services to the travel objects based on the travel inadequacy evaluation index.

[0020] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0021] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0022] The travel service recommendation method, apparatus, electronic device, and storage medium proposed in this application organize discrete transportation stations into meaningful travel area pairs through regional clustering. This elevates the analysis of travel demand from a point-to-point level to a region-to-region level, better reflecting actual commuting patterns. Subsequently, historical travel data is extracted and filtered hierarchically to aggregate travel information at different granularities. When filtering target travel area pairs, not only passenger flow but also the maximum number of transfers is considered, making the selected area pairs more representative and accurately reflecting potential travel service demand. Furthermore, a travel inconvenience evaluation index is used to quantitatively assess the travel experience. This index is generated based on a comprehensive analysis of first and third historical travel data, providing a more objective reflection of the traveler's comfort level within a specific operating time period. Compared to related technologies that rely on fixed-station passenger flow statistics, this embodiment can comprehensively evaluate the travel experience from multiple dimensions, accurately identify specific travel pain points for specific area pairs within a specific time period, and more comprehensively reflect the traveler's real travel difficulties. This makes the recommendation service more targeted and personalized, effectively solving the problem of providing a good commuting experience in related technologies and improving traveler satisfaction. Attached Figure Description

[0023] Figure 1 This is a flowchart of the travel service recommendation method provided in the embodiments of this application;

[0024] Figure 2 yes Figure 1 The flowchart of step S106 in the process;

[0025] Figure 3 yes Figure 2 Flowchart of step S203;

[0026] Figure 4 It generates a flowchart that takes an excessively long time.

[0027] Figure 5 This is a flowchart for generating a transfer mismatch process;

[0028] Figure 6 It is a flowchart for generating excessive congestion;

[0029] Figure 7 yes Figure 1 Another flowchart of step S106 in the process;

[0030] Figure 8 This is a schematic diagram of the travel service recommendation device provided in the embodiments of this application;

[0031] Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0033] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0035] With the acceleration of urbanization and the continuous expansion of urban space, the separation of work and residence has become increasingly prominent, and daily commuting has become an important part of urban residents' daily lives. As the artery of urban operation, public transportation carries the function of large-scale passenger travel, and the coverage of its service network and the transportation organization mode are directly related to travel efficiency and comfort.

[0036] The travel services in related technologies rely on macro indicators such as passenger flow statistics at fixed stations and the proportion of long-distance commutes for network planning and supply-demand matching. While this conventional approach can meet the large-scale travel planning of urban public transportation, it has technical defects at the micro-level of customized services. It is difficult to fully reflect the real travel difficulties of passengers, and it cannot accurately identify and aggregate similar travel needs. As a result, it cannot recommend travel services to travelers in a personalized way, leading to a poor travel experience.

[0037] Based on this, embodiments of this application provide a travel service recommendation method, apparatus, electronic device, and storage medium, aiming to provide personalized transportation services for travelers with the same travel needs, thereby providing a better travel experience.

[0038] The travel service recommendation method, apparatus, electronic device, and storage medium provided in this application are specifically described through the following embodiments. First, the travel service recommendation method in this application embodiment is described.

[0039] The travel service recommendation method provided in this application relates to the field of intelligent transportation technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the travel service recommendation method, but is not limited to the above forms.

[0040] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0041] Figure 1 This is an optional flowchart of the travel service recommendation method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S101 to S106.

[0042] Step S101: Based on the station location information of transportation stations in the target area, perform regional clustering to obtain multiple candidate travel area pairs, and construct candidate travel chains according to the starting station set and ending station set of each candidate travel area pair.

[0043] Step S102: Obtain the first historical travel data of the target area in a preset historical period. For each preset operating time interval, extract the second historical travel data of multiple historical time intervals that are the same as the operating time interval from the first historical travel data. The second historical travel data represents the data of the travel object entering and exiting from the stations in the starting station set within the historical time interval and belonging to the corresponding ending station set.

[0044] Step S103: For each candidate travel area pair, determine the maximum number of transfers for the candidate travel chain, and obtain the target historical passenger flow within a preset number of consecutive historical time intervals from the second historical travel data.

[0045] Step S104: Based on the maximum number of transfers, the target historical passenger flow, and preset filtering conditions, the target travel area pair is obtained from multiple candidate travel area pairs.

[0046] Step S105: Extract the third historical travel data from the second historical travel data. The third historical travel data represents the data of stations entering and exiting from the target origin station set of the target travel area within the historical time interval, which belong to the corresponding target end station set.

[0047] Step S106: For each target travel area pair, generate a travel inappropriateness evaluation index for each operating time interval based on the first historical travel data and the third historical travel data, and recommend travel services to the travel target based on the travel inappropriateness evaluation index.

[0048] In step S101, the target region can be a city or a specific area within a city. The target region includes multiple transportation stations, such as bus stops and subway stations. A candidate travel region pair refers to a pair consisting of two travel regions obtained through region clustering, where one travel region is the starting region and the other is the ending region. The starting station set refers to the set of all transportation stations in the candidate travel region pair that serve as the starting point of a trip, and the ending station set refers to the set of all transportation stations in the candidate travel region pair that serve as the ending point of a trip. A candidate travel chain refers to a path in the candidate travel region pair that starts from any transportation station in the starting station set and ultimately reaches any transportation station in the ending station set.

[0049] First, transportation stations with similar geographical locations can be categorized into different travel areas. Optionally, the Bron-Kerbosch clustering algorithm can be used to group transportation stations whose location information satisfies the condition that the walking distance is less than a preset distance threshold M into the same travel area, where the distance threshold M can be set to 1km. In the area division process, this application allows the same transportation station to belong to multiple different travel areas simultaneously, forming flexible travel areas with overlapping characteristics. For example, station G1 can belong to travel area area1 (including stations G1, G2, and G3) or travel area area2 (including stations G1, G4, and G5).

[0050] Taking a target area containing five transportation stations A, B, C, D, and E as an example, the regional clustering process in step S101 is explained in detail. It is assumed that the distances between stations A and E, B and E, C and D, C and E, and D and E are all greater than a preset distance threshold (e.g., greater than 1 kilometer), while the distances between all other station pairs are no more than 1 kilometer. When performing regional clustering, it is required that the distance between any two transportation stations within the same area does not exceed 1 kilometer. Finally, regional clustering is performed on the above five stations to obtain three flexible travel areas: {A, B, C}, {A, B, D}, and {E}. It should be noted that for a single station E, which is more than 1 kilometer away from all other transportation stations, it is separately classified into a travel area {E}.

[0051] Existing technologies only focus on fixed-station-to-fixed-station demand, with each station belonging to only one fixed area. Travelers with the same destination and close proximity are scattered across isolated station data, leading the decision-making system to deem passenger flow insufficient to support route opening. This application constructs flexible travel areas, aggregating scattered passenger demand around transportation hubs and maximizing the aggregation of similar travel passenger flow from nearby stations, thus making it easier to meet the conditions for carpooling or minibus operation. Simultaneously, this application imposes distance constraints, ensuring that regardless of where the carpooling or minibus stop is located within the area, the walking distance to the boarding point is within an acceptable range, allowing travelers to enjoy direct express service without incurring excessive walking costs.

[0052] Furthermore, for flexible travel areas without shared transportation stops, each flexible travel area is used as a starting station set (areaO) and paired with other flexible travel areas without shared transportation stops. That is, the transportation stops in the remaining flexible travel areas are used as an ending station set (areaD), constructing candidate travel area pairs (areaO, areaD). Then, based on the starting and ending station sets of each candidate travel area pair, candidate travel chains are constructed. For example, if the starting station set areaO contains stations G1 and G2, and the ending station set areaD contains stations G3 and G4, candidate travel chains are constructed for G1 to G3, G1 to G4, G2 to G3, and G2 to G4. This provides data support for subsequently quantifying the transfer patterns of travelers on specific routes.

[0053] In step S102, the preset historical period refers to the time range used to collect historical travel data, such as the past week, month, or year. The first historical travel data consists of complete travel records for all transportation stations in the target area within the preset historical period. This data can be exported from the traffic management department's database and may include information such as the traveler's arrival time, departure time, arrival station, departure station, and transfer stations. To achieve a more refined measurement of travel characteristics, the traffic operating hours of the target area can be divided into finer time granularities based on preset time slices, resulting in multiple consecutive operating time intervals (Time_in). For example, if the traffic operating hours are from 6:00 to 23:00, using 5-minute time slices, several consecutive operating time intervals are obtained: 6:00-6:05, 6:05-6:10, ..., 22:55-23:00. Travel data within each operating time interval is used to analyze travel characteristics in different time periods.

[0054] Then, second historical travel data can be extracted from the first historical travel data, covering multiple historical time intervals that are the same as the aforementioned operating time intervals. The second historical travel data represents travel data where, within the corresponding historical time interval, the traveler enters from a station in the originating station set and exits from a station in the corresponding ending station set. It should be noted that the second historical travel data represents travel data where, within the historical time interval, the traveler enters from a station in the originating station set (areaO) and exits from a station in the corresponding ending station set (areaD), but the exit time does not need to meet the requirement of being within the historical time interval. For example, for one operating time interval, 6:00-6:05, travel data from the previous month (historical period) between 6:00-6:05 that entered from stations in areaO and exited from stations in areaD (exit time does not need to meet 6:00-6:05) can be collected. Corresponding travel data can be collected for each day of the previous month between 6:00-6:05 (multiple historical time intervals), thus obtaining the second historical travel data.

[0055] In step S103, to reduce computational load, trip area pairs that meet preset conditions can be selected from multiple candidate trip area pairs. Specifically, parameters for selecting trip area pairs can be obtained first. For each candidate trip area pair, the maximum number of transfers (TransBin) in the corresponding candidate trip chain is determined. Specifically, the number of transfers for each candidate trip chain can be calculated first. Since the starting and ending stations of the candidate trip chain are determined, the number of transfers can be directly determined through route planning, which can be obtained by subtracting 1 from the number of segments of the trip path. For example, a trip path with 1 segment represents 0 transfers, and 2 segments represent 1 transfer. Then, the maximum number of transfers can be determined from the transfer counts of all candidate trip chains. For example, in the aforementioned candidate trip area pair containing four candidate trip chains, three of the candidate trip chains have 1 transfer, while the transfer count from G2 to G4 is 3. Therefore, the maximum number of transfers (TransBin) for this area pair is determined to be 3.

[0056] Meanwhile, to obtain the target historical passenger flow within a preset number (e.g., 3, 4, etc.) consecutive historical time intervals from the second historical travel data, it is necessary to ensure that the set passenger flow examination time is an integer multiple of the aforementioned time slices. For example, assuming the preset number is 3, the historical passenger flow corresponding to the three consecutive historical time intervals of 6:00-6:05, 6:05-6:10, and 6:10-6:15 can be extracted from the second historical travel data. It should be noted that the historical passenger flow represents the passenger flow that enters the station from the starting station set of the candidate travel area pair within the three historical time intervals of 6:00-6:05, 6:05-6:10, and 6:10-6:15, and the exit station belongs to the corresponding ending station set (the exit time does not need to meet the above three historical time intervals).

[0057] Furthermore, the passenger flow corresponding to each historical time interval within a continuous historical time interval can be extracted. ,in Represents the first in a continuous historical time interval The system can calculate a target historical passenger flow by summing the results over a predetermined number of historical passenger flow intervals. As mentioned above, the historical passenger flow count from 6:00 to 6:05 is 1,000 people, the historical passenger flow count from 6:05 to 6:10 is 1,500 people, the historical passenger flow count from 6:10 to 6:15 is 1,500 people, and the final calculated target historical passenger flow is 4,000 people.

[0058] Thus, by accumulating passenger flow over consecutive time periods, this application can more objectively reflect whether candidate travel areas have continuous, concentrated, and similar passenger travel demand within a specific time period.

[0059] In step S104, in order to pre-screen travel area pairs that meet the basic group size requirements and have inconvenient transportation, and reduce the computational workload of subsequent complex model calculations, the maximum number of transfers (Transbin) and the target historical passenger flow determined in the aforementioned steps can be used as a basis. The system identifies target travel area pairs that meet preset screening criteria from multiple candidate travel area pairs. Specifically, a transfer number threshold K can be set, for example, K=1, and a passenger flow threshold N can be set, for example, N=5000 people. The preset screening criteria include that the maximum number of transfers for the candidate travel area pairs is not less than K, and the target historical passenger flow is not less than N. This allows the system to screen target travel area pairs from multiple candidate travel area pairs where the maximum number of transfers is not less than the transfer number threshold, and the target historical passenger flow is not less than the passenger flow threshold. In this way, the present application can screen target travel area pairs from candidate travel area pairs that have at least one complicated route requiring transfers, and whose passenger flow within a continuous time period reaches the preset passenger flow scale requirement. The screened target travel area pairs can more accurately reflect the similarity of travel users in terms of travel service needs or travel characteristics.

[0060] In step S105, subsequent travel inappropriateness evaluation indicators can be calculated for the selected target travel area pairs. Specifically, third historical travel data can be extracted from the second historical travel data. The third historical travel data represents the travel data of travelers entering and exiting stations from the target origin station set of the target travel area pair within a specific historical time interval, where the exit station belongs to the corresponding target end station set. For example, for one operating time interval, 6:00-6:05, the travel data of travelers entering stations from the target origin station set of the target travel area pair within 6:00-6:05 (historical time interval) in the previous month (historical period) and exiting stations from the corresponding target end station set (exit time does not need to meet 6:00-6:05) in the previous month (multiple historical time intervals) can be collected, thus obtaining the third historical travel data.

[0061] This application first uses regional clustering to organize discrete transportation stations into meaningful pairs of travel regions, elevating the analysis of travel demand from a point-to-point level to a region-to-region level, which is more in line with actual commuting patterns. Then, by performing stratified extraction and filtering of historical travel data, it achieves a focus on travel information at different time granularities, thus better identifying travelers with similar travel needs.

[0062] In step S106, for each target travel area pair, a travel inconvenience evaluation index for each operating time interval can be generated based on the first historical travel data and the third historical travel data. The travel inconvenience evaluation index is a comprehensive indicator used to quantify the comfort level of a traveler's travel experience from the target originating station set to the target ending station set within a specific operating time interval. Furthermore, travel services can be recommended to travelers based on the travel inconvenience evaluation index. For example, when the travel inconvenience evaluation index of a target travel area pair reaches a preset inconvenience condition within a specific operating time interval, suitable travel services can be recommended to travelers within the corresponding target travel area pair who are traveling within the operating time interval. These travel services may include carpooling services or dedicated bus line services.

[0063] Steps S101 to S106 of this application, through regional clustering, organize discrete transportation stations into meaningful travel area pairs, elevating the analysis of travel demand from point-to-point to region-to-region level, which is more in line with actual commuting patterns. Subsequently, historical travel data is extracted and filtered hierarchically to aggregate travel information at different granularities. When filtering target travel area pairs, not only passenger flow is considered, but also the maximum number of transfers, making the selected area pairs more representative and more accurately reflecting potential travel service demand. In addition, the travel experience is quantitatively evaluated through a travel inadequacy evaluation index. The generation of this index is based on a comprehensive analysis of first and third historical travel data, which can more objectively reflect the travel comfort of the traveler within a specific operating time interval. Compared with related technologies that rely on fixed-station passenger flow statistics, this embodiment can comprehensively evaluate the travel experience from multiple dimensions, accurately identify specific travel pain points of specific area pairs within a specific time period, and more comprehensively reflect the real travel difficulties of the traveler, making the recommendation service more targeted and personalized. This can effectively solve the problem of providing a good commuting experience in related technologies and improve the satisfaction of the traveler.

[0064] In some embodiments, a target travel region pair includes target travel chains constructed from a set of target originating stations and a set of target ending stations. A target travel chain refers to a specific travel path within the target travel region pair, from a station (gridO) in a target originating station set to a station (gridD) in a corresponding target ending station set. This allows for data drill-down extraction from a macro-level regional pair to a micro-level single target travel chain, providing a data foundation for subsequent accurate and independent calculation of multiple dimensions of travel inconvenience. For a single travel link, once the origin and destination are fixed, the number of transfers obtained from path planning is usually deterministic, and the time difference is relatively small. However, travel experiences such as vehicle congestion vary at different times. Therefore, as mentioned above, it is necessary to extract third historical travel data corresponding to each target travel chain combined with different operating time intervals. This third historical travel data can include basic operating parameters such as the number of transfer segments, in-vehicle time for each segment, and vehicle density, thereby providing data support for the calculation of multi-dimensional travel inconvenience evaluation indicators.

[0065] In this embodiment, please refer to Figure 2 Based on the first and third historical travel data, travel inappropriateness evaluation indicators are generated for each operating time interval, including:

[0066] Step S201: For each target travel chain of the target travel area pair, generate the transfer inappropriateness, time inappropriateness, and congestion inappropriateness corresponding to each operating time interval of the target travel chain based on the first historical travel data and the third historical travel data.

[0067] Step S202: For each target travel chain of the target travel area pair, extract the historical number of passengers corresponding to the target travel chain in each operating time interval from the third historical travel data.

[0068] Step S203: For each target travel area pair, generate travel inappropriateness evaluation indicators for each operating time interval based on transfer inappropriateness, time inappropriateness, congestion inappropriateness, and historical passenger numbers.

[0069] In step S201, for each target travel chain of the target travel area pair, a transfer dissatisfaction corresponding to the target travel chain in each operating time interval is generated based on the first historical travel data and the third historical travel data. Transfer dissatisfaction characterizes the degree of experience degradation caused by excessive transfers on a specific travel route. Specifically, transfer benchmark features can be obtained from the first historical travel data, which characterizes the macro-level travel situation of the target area. Simultaneously, the actual number of transfer segments of the target travel chain within a specific operating time interval is extracted from the third historical travel data, which represents a micro-level single travel link. Then, the actual micro-level transfer situation of the target travel chain is compared and evaluated with the macro-level transfer benchmark features to objectively quantify the dissatisfaction of the target travel chain in the transfer dimension.

[0070] Furthermore, based on the first and third historical travel data, time-consuming inadequacy is generated for each operating time interval of the target travel chain. Time-consuming inadequacy reflects the degree of experience degradation caused by the total time spent by the traveler on public transportation. Similarly, macro-level time-consuming benchmark features at the target region level can be extracted from the first historical travel data, and the actual benchmark time of the target travel chain within a specific operating time interval can be obtained from the third historical travel data. Then, by comparing the difference between the actual benchmark time and the macro-level benchmark time, time-consuming inadequacy of the target travel chain is generated.

[0071] Furthermore, it is necessary to generate the level of congestion in each operating time interval for the target travel chain based on historical travel data. The level of congestion comprehensively considers the travel duration of passengers and the density of passengers in the carriage, which can reproduce the differences in the actual travel experience in different operating time intervals. For the target travel chain, the in-vehicle dwell time and vehicle density data reflecting the degree of congestion can be extracted from the third-party historical travel data. Since the congestion situation of the same target travel chain varies at different times, this application can combine the vehicle density and the corresponding in-vehicle time in different time intervals to jointly assess the level of congestion.

[0072] In step S202, to avoid the technical problem of relying on objective path disadvantages to evaluate travel inadequacy and thus deviating from actual passenger demand, this application combines the objective inadequacy generated by the travel link with the actual population size to comprehensively solve the travel inadequacy evaluation index. Specifically, for each target travel chain of the target travel area pair, the historical number of passengers corresponding to the target travel chain in each operating time interval is extracted from the third historical travel data. The historical number of passengers refers to the actual number of people who choose a certain target travel chain to travel within a specific operating time interval, that is, the number of passengers who enter the station at the target starting station corresponding to the target travel chain and exit the station at the corresponding target ending station within the specific operating time interval (the exit time does not need to meet the specific operating time interval).

[0073] It should be noted that within a historical period (e.g., the last 30 days of the previous month), 30 samples of historical travel numbers are collected. These 30 samples can be sorted from smallest to largest, and the number of people corresponding to the preset quantiles is taken as the final historical travel number. For example, the 75th percentile. To ensure the accuracy of historical travel statistics, the selected quantile needs to be greater than the 50th percentile. This is because if the arithmetic mean of historical travel numbers is used, the high passenger flow during weekday peak hours of the target travel chain is often significantly reduced by the low passenger flow on weekends or holidays; if the maximum value is used, it is easily affected by abnormal interference from extreme weather or occasional events. Using the 75th percentile can effectively filter out extreme and occasional values, while also accurately reflecting the normal passenger flow level that the target travel chain can stably achieve on most weekdays.

[0074] While the aforementioned steps quantify the travel inadequacy of the target travel chain, a true assessment of whether a particular travel area meets the economic threshold for recommending carpooling or launching a micro-circulation bus service requires consideration of the actual scale of the travel population. Therefore, this application extracts the historical number of travelers for each target travel chain within the target travel area pair from third-party historical travel data. In this way, the travel intolerance evaluation index can be guaranteed to truly reflect the travel target group with large numbers of people and poor travel experience, and avoid interference in the calculation of the travel intolerance evaluation index by the data of individual travel chains with extremely poor experience but almost no passenger flow (such as a target travel chain with only a few passengers but complicated transfers, or several target travel chains with few travelers but low travel intolerance).

[0075] In step S203, for each target travel region pair, the comprehensive score of each target travel chain within it can be calculated first, and then combined with the historical number of travelers to generate the final travel inconvenience evaluation index representing the target travel region pair in each operating time interval. In this way, the travel inconvenience evaluation index not only considers the quality of the travel experience, but also the number of affected travelers, making the evaluation results more representative and practical.

[0076] This embodiment refines the assessment of travel intolerance from a macro-regional level to the travel chain level, comprehensively considering multiple dimensions such as transfers, travel time, and congestion, while also incorporating actual historical travel data. This assessment method solves the problem of inaccurate individual travel intolerance assessments caused by related technologies relying on single macro-indicators such as the proportion of long commutes. This allows the generated travel intolerance evaluation index to more accurately and comprehensively reflect the actual travel experience of travelers within a specific time period and along a specific route. Compared to generalized assessments of entire regions, this embodiment can identify which travel regions have poor travel experiences for people traveling within specific operating time periods, as well as the corresponding scale of these travelers. This effectively prevents the travel characteristics of high travel intolerance groups from being flattened by redundant travel chain data with fewer participants or relatively comfortable experiences, thus providing a more accurate basis for personalized travel service recommendations.

[0077] In some embodiments, please refer to Figure 3 Based on factors such as transfer inadequacy, travel time inadequacy, crowding inadequacy, and historical passenger numbers, a travel inadequacy evaluation index is generated for each operating time interval, including:

[0078] Step S301: Calculate the overall link inadequacy of the target travel chain in each operating time interval based on transfer inadequacy, time inadequacy, and congestion inadequacy.

[0079] Step S302: Sum the historical number of passengers for all target travel chains within the target travel area to obtain the total number of historical passengers;

[0080] Step S303: The link-based inadequacy of each target travel chain is weighted and summed with the corresponding historical number of travelers. The ratio between the weighted summation result and the total number of historical travelers is used to calculate the travel inadequacy evaluation index of the target travel area in each operating time interval.

[0081] In step S301, in practical applications, it is necessary to integrate the travel inadequacy evaluation indicators of a single target travel chain and the corresponding historical number of travelers to ensure that the evaluation indicators can reflect the actual passenger flow weight of different target travel chains, so as to comprehensively and accurately assess the travel inadequacy of the target travel area during a specific operating time period.

[0082] First, the overall inadequacy of the target travel chain can be calculated for each operating time interval. In real-world travel experiences, any unbearable level of cumbersome transfers, excessively long transfer times, or overcrowded carriages will directly lead to a poor overall travel experience. This application abandons the simple summation and averaging algorithm in related technologies and calculates the overall inadequacy of transfers (… ), inappropriate time consumption ( ) and overcrowding ( This involves using a specific nonlinear algorithm to accurately calculate the overall link mismatch. Specifically, the maximum value of the target travel link's mismatch across these three dimensions can be extracted first (i.e., The base is used as the basis for inadequacy. Then, the average of the remaining two relatively lower inadequacies is calculated (i.e., The average value is multiplied by a preset, configurable, adjustable scaling factor. (For example, you can set) The product is set to 0.2), and the calculated product is added as an additional increment to the base inequality. Finally, the result of the above superposition operation is compared with the full score of 100 and the smaller value is taken, thus limiting the final calculation result to the percentage mapping range of 0 to 100. The mathematical expression is as follows:

[0083] (1)

[0084] in, This indicates that the overall link performance of the target travel chain is inadequate in each operating time interval. This indicates an inappropriate transfer. This indicates that the time spent was inappropriate. This indicates that the crowding is inappropriate.

[0085] The method for calculating the overall travel inadequacy in this embodiment ensures the dominance of travel inadequacy in dimensions with a larger proportion, preventing it from being leveled off by other dimensions. For example, assuming the transfer inadequacy of the target travel chain exceeds 80 points within a specific operating time interval, even if its time-consuming and congestion inadequacies are very low, this embodiment will still extract transfer inadequacy as a baseline. The other two lower inadequacies only play a supplementary role by adjusting coefficients. Thus, as long as any one of the inadequacies in the target travel chain is greater than or equal to 80 points, the final calculated overall travel inadequacy will be at least 80 points. This embodiment objectively restores the actual travel characteristics where a single travel inadequacy determines a poor overall travel experience, laying a quantitative benchmark for subsequent evaluation of overall travel inadequacy indicators based on passenger flow and regional data.

[0086] In step S302, when assessing whether the target travel area has the potential to recommend carpooling services or direct bus lines, it is not enough to limit the overall inadequacy of a single target travel chain; the passenger flow scale of all target travel chains in the target travel area needs to be taken into account.

[0087] Specifically, the total number of historical travelers can be obtained by summing the historical travel numbers of all target travel chains within the target travel area extracted in the aforementioned steps. As described above, each target travel chain operates within a specific time interval ( Each of these corresponds to a historical number of travelers. The total number of historical travelers is obtained by summing up all historical travelers to the target travel area. .

[0088] For example, the target travel area includes four target travel chains. Within a specific operating time interval, the historical number of passengers corresponding to the four target travel chains is... The total number of historical trips to the target travel area within the specified operating time interval was calculated using the figures of 1000, 1500, 500, and 2000 people respectively. For 5,000 people.

[0089] In step S303, specifically, for each target travel chain in the same target travel area pair, the comprehensive link mismatch of the target travel chain within the same operating time interval is obtained. And calculate the overall inadequacy of the link and the corresponding historical number of travelers. The weighted travel inadequacy of the target travel chain is calculated by multiplying the products of the products. Then, the weighted travel inadequacy of all target travel chains for the target travel region pair is summed.

[0090] Furthermore, based on the ratio between the weighted summation result and the total number of historical trips, an evaluation index for travel inadequacy in the target travel area during each operating time interval is calculated. The specific mathematical formula is as follows: ,in The travel inadequacy evaluation index accurately reflects the real travel experience of travelers within a specific operating time period and the target travel area.

[0091] This embodiment first integrates the inadequacy of transfers, time consumption, and congestion for a single target travel chain to calculate the overall inadequacy of the chain, thus unifying multi-dimensional inadequacy information into a single indicator. Simultaneously, the historical passenger numbers for all target travel chains within the target travel area are summed to obtain the total historical passenger numbers, providing a passenger flow benchmark for subsequent weighted calculations. Then, the overall inadequacy of each target travel chain is weighted and summed with its corresponding historical passenger numbers. The ratio of the weighted summation to the total historical passenger numbers is used to calculate the travel inadequacy evaluation index for the target travel area in each operating time interval. This calculation method ensures that target travel chains with higher passenger volume have a greater impact on the overall travel inadequacy evaluation index, allowing the final evaluation index to more accurately reflect the actual travel experience of most travelers within the target travel area. It also avoids the bias that may arise from simple averaging, enabling travel service recommendations to more accurately match travelers with similar travel needs in reality.

[0092] In some embodiments, please refer to Figure 4The method for generating time-consuming data includes the following steps:

[0093] Step S401: Extract the first historical travel time from the first historical travel data, determine the reference historical travel time based on the median of the first historical travel time, determine the first benchmark time based on the first quantile of the first historical travel time, and determine the second benchmark time based on the second quantile of the first historical travel time, wherein the first quantile is greater than the second quantile.

[0094] Step S402: For each target travel chain in the target travel area pair, extract the second historical travel time corresponding to each operating time interval from the third historical travel data, and determine the target benchmark time based on the median of the second historical travel time.

[0095] Step S403: For each target travel chain in the target travel area pair, generate the time inappropriateness corresponding to the target travel chain in each operating time interval based on the target baseline time, the reference historical travel time, the first baseline time, and the second baseline time.

[0096] In step S401, in actual travel scenarios, travel time is often affected by various factors, such as traffic conditions and weather changes, resulting in uncertainty and volatility in its distribution. Simply using the average travel time may not accurately reflect the traveler's true feelings and tolerance for travel time, nor can it effectively distinguish between normal fluctuations and abnormal interference, thus affecting the accuracy of the evaluation of inappropriate travel time and the accuracy of recommended travel services.

[0097] First, a baseline characteristic of travel time distribution within the target region can be determined, addressing the lack of a unified comparison benchmark for travel time across single-target travel chains, and the problem that manually set benchmarks are too subjective and fail to reflect the overall travel time of the target region. To construct this baseline characteristic, the first historical travel time for all travel chains within the target region can be extracted from the first historical travel data. This first historical travel time represents the travel time corresponding to all historical travel records included in the first historical travel data. Then, a reference historical travel time can be determined based on the median of the first historical travel time. Choosing the median, rather than the arithmetic mean, as the reference standard can effectively eliminate the interference of outliers in travel time caused by extreme weather or occasional traffic accidents, thus objectively reflecting the travel time of the vast majority of travelers in the target area.

[0098] Furthermore, to quantify the dispersion and distribution range of time-travel data in the target region, it is necessary to determine the upper and lower boundary parameters of the time-travel distribution. Therefore, a first baseline time can be determined based on the first quantile of the first historical travel time, and a second baseline time can be determined based on the second quantile of the first historical travel time, where the first quantile is greater than the second quantile. Specifically, the first quantile can be the 75th quantile of the time-travel distribution, thereby determining the corresponding first baseline time. Correspondingly, the second quantile can be the 25th quantile of the time consumption distribution, thereby determining the corresponding second baseline time consumption. It should be noted that the first quantile can also be the 70th quantile, and correspondingly, the second quantile can be the 30th quantile. The above example does not impose limitations on this example. In this way, by using the first and second benchmark travel times, the travel time range of the core travel population in the target area can be accurately defined, and a macro-level travel time evaluation benchmark that is not affected by local road condition fluctuations can be constructed.

[0099] In step S402, for each target travel chain in the target travel area pair, the second historical travel time corresponding to each operating time interval of the target travel chain can be extracted from the third historical travel data. The second historical travel time refers to the set of all historical actual travel time samples within the historical period that enter the station at the target originating station corresponding to the target travel chain, and whose entry time falls within the specified operating time interval.

[0100] Furthermore, the target baseline time can be determined based on the median of the second historical travel time. Specifically, the actual travel times of all target travel chains generated within a specific operating time interval during a historical period can be sorted from smallest to largest, and the median can be selected as the corresponding target baseline travel time. For example, for a target travel chain, the actual historical travel times of each day of the previous month within the period of 07:30-07:45 can be extracted, and then the target baseline travel time of the target travel chain within that operating time interval can be determined based on the median of the actual historical travel times (e.g., 48 minutes).

[0101] In step S403, since the distances of different regions and different target travel chains vary greatly, it is necessary to consider the inappropriate travel time of the target travel chain in the overall traffic time distribution environment of the target region.

[0102] Specifically, the time calculation based on the target baseline time of the target travel chain, the reference historical travel time, the first baseline time, and the second baseline time is inappropriate. Firstly, the difference between the first baseline time and the second baseline time can be calculated, i.e. Then, for each target travel chain within the target travel area pair, the difference between its target baseline travel time within a specific operating time interval and the macro-level reference historical travel time of the target region can be calculated, i.e. Furthermore, the time offset of the target travel chain can be calculated. The specific calculation formula is as follows: .

[0103] Furthermore, in real-world commuting experiences, the sensitivity of travelers to travel time changes non-linearly. When the time exceeding the average is small, the increase in traveler distress is relatively slow, but when the travel time exceeds a certain threshold, the traveler's anxiety and distress increase sharply. To closely fit this pattern of subjective human psychological perception, the Sigmoid function can be used to calculate and transform the time offset. The specific mapping formula is as follows: ,in This indicates the preset curve slope control parameter (for example, a value of 0.7; the larger the value, the steeper the curve of the increase in travel pain as the travel time increases).

[0104] This embodiment provides a more refined and accurate method for assessing time-inappropriate travel. By introducing global reference historical travel times, first baseline travel times, and second baseline travel times, and combining them with the target baseline travel times of a specific target travel chain within a specific operating time interval, it effectively avoids the limitations of single-indicator assessment. This allows time-inappropriate travel to more objectively reflect the traveler's true perception of travel time, distinguish between normal fluctuations and abnormal delays, and improve the accuracy of time-inappropriate travel assessment. As a result, it can recommend more accurate and personalized travel services to travelers.

[0105] In some embodiments, please refer to Figure 5 The method for generating transfer inadequacy includes the following steps:

[0106] Step S501: Determine the reference number of transfers based on the median number of transfers for all travel chains in the target area, determine the first benchmark number of transfers corresponding to the third quantile of the number of transfers for all travel chains, and determine the second benchmark number of transfers corresponding to the fourth quantile of the number of transfers for all travel chains, wherein the third quantile is greater than the fourth quantile.

[0107] Step S502: For each target travel chain in the target travel area pair, obtain the target number of transfers for the target travel chain, and calculate the transfer inadequacy corresponding to the target travel chain in each operating time interval based on the target number of transfers, the reference number of transfers, the first benchmark number of transfers, and the second benchmark number of transfers.

[0108] In step S501, a baseline for the distribution of transfer times within the target area can be determined first. Specifically, the actual number of transfers for all travel chains within the target area can be obtained. Then, a reference number of transfers can be determined based on the median of all actual transfer times. .

[0109] Specifically, the number of transfers for all travel chains can be sorted by size and quantiles can be used to determine the first baseline number of transfers corresponding to the third quantile of the number of transfers for all travel chains. And determine the second baseline number of transfers corresponding to the fourth quantile of the number of transfers for all travel chains. The third quantile is greater than the fourth quantile. For example, the 75th quantile can be selected as the third quantile, and correspondingly, the 25th quantile can be selected as the fourth quantile. It should be noted that the above example does not limit this embodiment.

[0110] In step S502, the number of transfers for a single target travel chain can be compared with the overall transfer frequency distribution of the transportation network in the target area to generate objective and consistent quantitative evaluation indicators. For each target travel chain within the target travel area pair, the target number of transfers for that target travel chain is obtained. .

[0111] To achieve the aforementioned standardized measurement, the relative offset can be calculated based on the target number of transfers, the reference number of transfers, the first benchmark number of transfers, and the second benchmark number of transfers. Specifically, the difference between the first benchmark number of transfers and the second benchmark number of transfers can be calculated first, i.e. Then, the difference between the target number of transfers for the target travel chain and the macro-reference number of transfers for the target region can be calculated, i.e. Furthermore, the offset of the target travel chain along the transfer dimension can be calculated, i.e. ,in, This represents the offset of the target travel chain in terms of the number of transfers.

[0112] Furthermore, considering that travelers are quite sensitive to increases in the number of transfers in real-world travel experiences, the Sigmoid function can be used to perform a non-linear mapping of the above-mentioned offset to obtain the transfer inconvenience, specifically expressed as follows: ,in, The preset curve slope control parameter (e.g., a value of 0.7) is used. After the mapping transformation by the Sigmoid function, the final transfer incongruity is output.

[0113] This embodiment analyzes the number of transfers across all travel chains in the target area to obtain representative reference transfer numbers, a first benchmark transfer number, and a second benchmark transfer number. Then, for each target travel chain to be evaluated, its transfer number is obtained and compared with preset statistical benchmarks to calculate the transfer inadequacy of the target travel chain within a specific operating time interval. This method considers the transfer situation of the target travel chain within the context of the entire travel network of the target area, making the assessment of transfer inadequacy no longer isolated but relative and comparable. Thus, this embodiment can more accurately reflect the transfer experience of travelers on different travel chains, thereby improving the accuracy of travel service recommendations.

[0114] In some embodiments, please refer to Figure 6 The method for generating congestion inadequacy includes the following steps:

[0115] Step S601: For each target travel area pair, divide the corresponding target travel chain according to the target number of transfers to obtain multiple corresponding target travel sub-chains;

[0116] Step S602: For each target travel sub-chain, extract the in-vehicle dwell time and passenger congestion coefficient corresponding to each operating time window from the third historical travel data. The operating time window includes a preset number of consecutive historical time intervals.

[0117] Step S603: For each target travel chain in the target travel area pair, calculate the congestion inequality corresponding to each operating time window based on the in-vehicle dwell time of the target travel chain, the vehicle congestion coefficient, and the preset reference in-vehicle dwell time. The reference in-vehicle dwell time is determined based on the fifth quantile of the in-vehicle dwell time of all targets in the third historical travel data.

[0118] In step S601, because the riding environment varies along different segments of a single target travel chain in a complex urban public transportation network—that is, the passenger density and dwell time in different segments of the same target travel chain are drastically different—it is difficult to uniformly measure the degree of congestion. If congestion is assessed broadly from a macro perspective of the target travel chain, the assessment of individual congestion levels will be inaccurate. Therefore, it is necessary to break down the target travel chain into multiple independent sub-chains with relatively uniform riding environments.

[0119] Specifically, for each pair of target travel areas, a corresponding target travel chain is divided based on the number of target transfers, resulting in multiple target travel sub-chains. A target travel sub-chain refers to dividing the target travel chain according to its internal transfer points, forming several continuous travel segments that do not include internal transfers. For example, a target travel chain containing two transfers can be divided into three target travel sub-chains. This division helps to analyze the congestion situation at different stages of the journey within the target travel chain in a more refined manner.

[0120] In step S602, to refine and quantify the actual experience of each travel environment and address the problem of inaccurate assessment results due to significant differences in passenger congestion levels across different travel segments and operating time intervals, this embodiment divides the time dimension by introducing an operating time window. An operating time window is a set containing a preset number of consecutive historical time intervals. For example, as described above, each 5-minute interval represents an operating time interval. To smooth out extreme data fluctuations and accurately match actual travel peaks and troughs, every 3 consecutive historical time intervals can be combined to form a 15-minute operating time window (e.g., 07:00 to 07:15, 07:15 to 07:30, etc.). The operating time window is then used as the time dimension for evaluating congestion discomfort. It should be noted that the above example does not limit this embodiment.

[0121] Furthermore, for each target travel sub-chain, the in-vehicle dwell time corresponding to each operating time window is extracted from the third historical travel data. (Units are usually minutes). Simultaneously, the congestion coefficient corresponding to each operating time window can be extracted from the third-party historical travel data. Specifically, the vehicle density of the target travel sub-chain within a specific operating time window can be extracted first. (Unit: per person per square meter). Then, based on the set density threshold standard, the vehicle density can be converted into the corresponding passenger congestion coefficient. Specifically, when vehicle density At that time, the riding environment of the target travel sub-chain within the current operating time window is defined as comfortable, and its riding congestion coefficient is set accordingly. Set to 1.0 (meaning no additional discomfort from overcrowding at this density); when the vehicle density is... When the interval is specified, the riding environment of the target travel sub-chain within the current operating time window is characterized as slightly crowded, and its riding congestion coefficient is set accordingly. The corresponding increase is 1.2; while when the vehicle density When the travel environment of the target travel sub-chain within the current operating time window is extremely crowded, the travel congestion coefficient is set accordingly. Set to 1.5.

[0122] In step S603, in actual complex public transportation commutes, the congestion discomfort of travelers has a cumulative effect over time and a cumulative effect across road segments. If the congestion level of each target travel sub-chain cannot be scientifically assessed in conjunction with time and mapped to a unified benchmark for the target area, it is impossible to accurately reflect the overall congestion discomfort of the target travel chain at a specific time period.

[0123] Therefore, we can first determine the reference in-vehicle dwell time. Specifically, we can extract the in-vehicle dwell time of all subjects within the entire target area from the third historical travel data. Then, we can determine the reference in-vehicle dwell time based on the fifth quantile of the aforementioned subjects' in-vehicle dwell times. The fifth quantile can be the 95th percentile. This is because if some travelers stay in the subway for a long time to enjoy the air conditioning, taking the highest time would result in an outlier caused by such an event. The 95th percentile can more accurately reflect the higher time of normal travel.

[0124] Furthermore, for each target travel sub-chain contained in the target travel chain, the corresponding in-vehicle dwell time is extracted. With the congestion coefficient Simultaneously, a preset time adjustment coefficient is introduced. (For example The value is 0.02, used to quantitatively characterize the increase in discomfort caused by each additional minute of time spent inside the vehicle. (The value represents the time spent inside the vehicle.) With the congestion coefficient Nonlinear fusion is performed to calculate the time-congestion joint coefficient of the target travel subchain. The specific expression is .

[0125] Furthermore, based on the time each subject spends inside the vehicle... As a weighting parameter, for all the aforementioned time-congestion joint coefficients The specific mathematical formula for performing chain-level weighted average is as follows: This allows us to derive a weighted joint coefficient that characterizes the overall congestion status of the target travel chain. Then, based on the reference in-vehicle dwell time and the highest passenger congestion coefficient benchmark (e.g., 1.5), the weighted joint coefficient can be mapped to a standard score using a normalization formula. The specific mathematical expression is as follows:

[0126] (2)

[0127] in, This indicates the time spent inside the vehicle, thus allowing for the calculation of congestion levels for the target travel chain within each operating time window. .

[0128] This embodiment refines the target travel chain into multiple target travel sub-chains and independently assesses congestion discomfort for each sub-chain. For each sub-chain, the in-vehicle dwell time and congestion coefficient within a specific operating time window are extracted from third-party historical travel data. This data reflects the actual experience of travelers within the carriage and the congestion situation. Subsequently, combined with a reference in-vehicle dwell time, the congestion discomfort for each target travel chain within the corresponding operating time window is calculated. This calculation method comprehensively considers both the actual dwell time of passengers within the carriage and the degree of congestion, enabling a more accurate quantification of discomfort during the travel process. Thus, this embodiment fully incorporates the key factor of carriage congestion when generating travel discomfort evaluation indicators, making the final travel service recommendation more closely aligned with users' actual feelings and needs.

[0129] In some embodiments, the evaluation indicators for travel intolerance include a comprehensive score for travel intolerance; please refer to [reference needed]. Figure 7 Based on the travel inappropriateness assessment index, travel service recommendations are made for travelers, including:

[0130] Step S701: When the comprehensive score of travel inappropriateness exceeds the preset score threshold, and the number of travel objects in the target travel area within the operating time interval corresponding to the comprehensive score of travel inappropriateness exceeds the first preset threshold, recommend carpooling service to the corresponding travel objects.

[0131] Step S702: Obtain the number of carpooling orders and the number of carpooling partners within a preset number of days. When the number of carpooling orders for each day within the preset number of days exceeds a preset order quantity threshold and the number of carpooling partners for each day exceeds a second preset threshold, recommend the bus route service to the corresponding travel partners.

[0132] In this embodiment, the evaluation index for travel inadequacy includes a comprehensive travel inadequacy score, which is a quantitative indicator of the degree of poor travel experience in a specific travel area within a specific operating time period. The comprehensive travel inadequacy score comprehensively considers multiple dimensions such as transfer inadequacy, time-consuming inadequacy, congestion inadequacy, and historical passenger numbers, and can be obtained through weighted summation or other aggregation algorithms.

[0133] Current public transportation planning methods lack effective intervention mechanisms for small-scale travelers whose total passenger volume has not yet reached the standard for regular bus routes but who experience poor travel experiences. Directly opening bus routes for this small group can easily lead to resource waste and high empty-running rates; while ignoring the issue will fail to improve their commuting difficulties. Therefore, this embodiment uses the travel inadequacy evaluation index calculated through the aforementioned steps to utilize socialized small-scale transportation capacity to accurately and flexibly meet the needs of this group of travelers with similar travel characteristics and smaller passenger volumes.

[0134] In step S701, the comprehensive score of travel inappropriateness for the target travel area within a specific operating time interval can be obtained first (i.e., the score in the aforementioned step). The system then determines whether the overall travel inconvenience score exceeds a preset threshold (e.g., 80 points) to confirm whether travelers in the target travel area are experiencing poor commuting conditions. Simultaneously, it obtains the number of travelers in the target travel area within the corresponding operating time interval and determines whether this number exceeds a first preset threshold. Since the purpose of this stage is to trigger basic small-scale transportation services, the first preset threshold can be set as a small group formation baseline (e.g., 3 people).

[0135] Furthermore, when the comprehensive score for travel inappropriateness exceeds a scoring threshold, and the number of travelers exceeds a first preset threshold, carpooling services can be recommended to the corresponding travelers. Specifically, the above judgment result indicates that within the corresponding operating time interval, there are indeed several (e.g., two or more) travelers facing commuting difficulties (e.g., multiple transfers, crowded transportation) in the target travel area. In this case, a direct service recommendation for small-scale social transportation (usually limited to four people or less), such as Didi carpooling, can be sent to the terminals of these travelers. Therefore, this embodiment can meet local travel needs through a carpooling platform without consuming a large amount of public transportation resources.

[0136] In step S702, this embodiment can accurately identify, based on the feedback results of the carpooling service, which target travel areas have only occasional fluctuations in carpooling demand and which target travel areas actually gather enough normal passenger flow to support medium-sized transportation capacity (such as 7-9 seat micro-circulation buses).

[0137] The process begins by acquiring the number of carpooling orders and carpooling participants within a preset number of days. Detecting this data filters out abnormal data such as sudden surges in daily passenger flow. Specifically, it involves acquiring and statistically analyzing the actual number of carpooling orders and carpooling participants completed within the same operating time interval for the target travel area over consecutive preset days. Carpooling participants represent the cumulative number of successful carpooling trips within the preset number of days. Then, it determines whether the number of carpooling orders for each day within the preset number of days exceeds a preset order threshold (e.g., the number of successful carpooling orders per day must be greater than or equal to 3). Simultaneously, it determines whether the number of carpooling participants per day exceeds a second preset threshold. This second preset threshold represents the minimum capacity for a minibus route; since the target capacity is being upgraded to a larger minibus route, the second preset threshold must be greater than the aforementioned first preset threshold for carpooling (e.g., setting the daily cumulative number of carpooling participants to be greater than or equal to 9).

[0138] When the number of carpooling orders for each day within a preset number of days exceeds a preset order threshold, and the number of carpooling participants for each day exceeds a second preset threshold, it can be determined that the large-scale direct demand for the target travel area has become normalized. Then, dedicated bus line services can be recommended to the corresponding travelers. Specifically, information on the demand for 7-9 seat minibus lines (including vehicle allocation, route, and departure time) can be generated, upgrading the capacity from the original 4-person carpooling to customized dedicated bus lines operated by public transport companies. In contrast to related technologies where carpooling platforms only look at real-time order numbers and public transport companies only look at macro-level totals without sufficient data for line opening, leading to data gaps and technical problems such as minibuses being unable to operate, this embodiment not only ensures that the deployed dedicated bus lines have a stable and data-verified passenger base, improving the success rate of bus line opening, but also fundamentally breaks down the system barriers between carpooling and public transport, achieving precise and efficient deployment of public transport resources.

[0139] Furthermore, after the minibuses are launched, they will first enter a 4-week trial operation observation window (or other durations, which are not limited here). During this period, the comprehensive score of travel inconvenience for the target travel area pairs can be continuously obtained. If, within these 4 weeks, the comprehensive score of travel inconvenience for the target travel area pairs remains greater than or equal to the preset score threshold, it indicates that the poor travel experience of the travelers is due to structural and long-term defects caused by the lack of coverage of the transportation network. This confirms from a data perspective that the travelers in the target travel area pairs have an extremely strong demand for direct services, rather than short-term fluctuations caused by occasional events. At the same time, the actual passenger flow performance of the minibuses during the trial operation observation window can be obtained and extracted, with the occupancy rate as the evaluation index. When the actual occupancy rate of the minibuses exceeds 70%, it indicates that the minibus routes accurately meet the needs of travelers with similar travel characteristics, solving the technical problem of long-term empty runs caused by the blind opening of traditional bus routes. When both the comprehensive score for travel inappropriateness and the actual occupancy rate meet the requirements, the minibus can be converted from a trial to a fixed micro-circulation service. This means that the corresponding service route is incorporated into the city's regular public transport network scheduling system, with fixed vehicle allocation, departure timetable, and stops.

[0140] This embodiment introduces a comprehensive travel inconvenience score and combines it with multiple dimensions such as the number of travelers and carpooling service operation data to achieve refined and personalized recommendations for travel services. When the comprehensive travel inconvenience score exceeds a preset score threshold, it indicates that the travel experience in the target travel area is poor during that operating time period. Furthermore, if the number of travelers in the target travel area exceeds a first preset threshold, it indicates that a large number of similar travel demands are not being fully met. In this case, recommending carpooling services can effectively alleviate travel pressure, improve vehicle utilization, and provide travelers with more flexible travel options. The recommendation mechanism of this embodiment can accurately identify travel pain points and provide immediate and effective solutions. Furthermore, to cope with long-term and stable high demand, this embodiment also introduces continuous monitoring of carpooling service operation data. This involves obtaining the number of carpooling orders and the number of carpooling participants within a preset number of days and comparing them with a preset order number threshold and a second preset threshold. When the number of carpooling orders and the number of carpooling participants each day consistently exceed the corresponding thresholds, it indicates that the travel demand in the area has reached a certain scale, and the existing carpooling service may be saturated or unable to provide an optimal solution. At this point, recommending dedicated bus lines can provide a more scalable, stable, and efficient public transportation solution, thereby further optimizing overall travel efficiency and user experience. This tiered recommendation strategy can accurately respond to different levels of travel discomfort and demand, enabling the travel service recommendation system to provide progressive solutions, from flexible carpooling to large-scale dedicated bus lines, based on different demand intensities and persistence. This more comprehensively meets travel needs and effectively utilizes urban transportation resources.

[0141] Please see Figure 8 This application also provides a travel service recommendation device that can implement the above-described travel service recommendation method. The device includes:

[0142] The construction unit 810 is used to perform regional clustering based on the station location information of transportation stations in the target area, obtain multiple candidate travel area pairs, and construct candidate travel chains based on the starting station set and ending station set of each candidate travel area pair.

[0143] The first extraction unit 820 is used to obtain the first historical travel data of the target area in a preset historical period. For each preset operating time interval, it extracts the second historical travel data of multiple historical time intervals that are the same as the operating time interval from the first historical travel data. The second historical travel data represents the data of the travel object entering the station from the station in the starting station set and the exit station belonging to the corresponding ending station set within the historical time interval.

[0144] The acquisition unit 830 is used to determine the maximum number of transfers for each candidate travel area pair and to acquire the target historical passenger flow within a preset number of consecutive historical time intervals from the second historical travel data.

[0145] The filtering unit 840 is used to filter the target travel area pair from multiple candidate travel area pairs based on the maximum number of transfers, the target historical passenger flow and preset filtering conditions.

[0146] The second extraction unit 850 is used to extract the third historical travel data from the second historical travel data. The third historical travel data represents the data of the stations entering and exiting from the target origin station set of the target travel area within the historical time interval, which belong to the corresponding target end station set.

[0147] Recommendation unit 860 is used to generate a travel inadequacy evaluation index for each operating time interval based on the first historical travel data and the third historical travel data for each target travel area pair, and to recommend travel services to the travel target based on the travel inadequacy evaluation index.

[0148] The specific implementation of this travel service recommendation device is basically the same as the specific embodiment of the travel service recommendation method described above, and will not be repeated here.

[0149] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned travel service recommendation method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0150] Please see Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0151] The processor 901 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0152] The memory 902 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 using the travel service recommendation method of the embodiments of this application.

[0153] The input / output interface 903 is used to implement information input and output;

[0154] The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0155] Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904);

[0156] The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0157] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described travel service recommendation method.

[0158] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0159] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0160] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0161] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0162] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0163] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0164] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0165] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0166] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0167] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0168] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0169] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for recommending travel services, characterized in that, The method includes: Based on the location information of transportation stations in the target area, regional clustering is performed to obtain multiple candidate travel area pairs, and a candidate travel chain is constructed according to the starting station set and ending station set of each candidate travel area pair. Obtain first historical travel data of the target area in a preset historical period. For each preset operating time interval, extract second historical travel data of multiple historical time intervals that are the same as the operating time interval from the first historical travel data. The second historical travel data represents the data of the travel object entering and exiting from the stations in the set of starting stations within the historical time interval and belonging to the corresponding set of ending stations. For each candidate travel region pair, determine the maximum number of transfers for the candidate travel chain, and obtain the target historical passenger flow within a preset number of consecutive historical time intervals from the second historical travel data; Based on the maximum number of transfers, the target historical passenger flow, and preset filtering conditions, a target travel region pair is obtained by filtering from multiple candidate travel region pairs. Extract third historical travel data from the second historical travel data. The third historical travel data represents the data of stations entering and exiting from the target origin station set of the target travel area within the historical time interval, which belong to the corresponding target end station set. For each pair of target travel areas, a travel inappropriateness evaluation index is generated for each of the operating time intervals based on the first historical travel data and the third historical travel data, and travel services are recommended to the travel target based on the travel inappropriateness evaluation index. The target travel area includes a target travel chain constructed from the target originating station set and the target ending station set. The step of generating travel inadequacy evaluation indicators for each operating time interval based on the first historical travel data and the third historical travel data includes: For each target travel chain of the target travel area pair, based on the first historical travel data and the third historical travel data, the transfer inappropriateness, time inappropriateness, and congestion inappropriateness of the target travel chain in each operating time interval are generated; For each target travel chain of the target travel area pair, extract the historical number of passengers corresponding to the target travel chain within each operating time interval from the third historical travel data; For each target travel area pair, a travel inappropriateness evaluation index is generated for each operating time interval based on the transfer inappropriateness, the time inappropriateness, the congestion inappropriateness, and the historical number of travelers.

2. The travel service recommendation method according to claim 1, characterized in that, The process of generating travel inconvenience evaluation indicators for each operating time interval based on the transfer inconvenience, the time consumption inconvenience, the congestion inconvenience, and the historical passenger numbers includes: The overall link inadequacy of the target travel chain in each of the operating time intervals is calculated based on the transfer inadequacy, the time inadequacy, and the congestion inadequacy. The total number of historical travelers is obtained by summing the historical travel numbers of all target travel chains within the target travel area. The link-based inadequacy of each target travel chain is weighted and summed with the corresponding historical number of travelers. The ratio between the weighted sum and the total number of historical travelers is used to calculate the travel inadequacy evaluation index of the target travel area in each operating time interval.

3. The travel service recommendation method according to claim 1, characterized in that, The method for generating the aforementioned time-consuming inappropriate method includes the following steps: First historical travel time is extracted from the first historical travel data. Reference historical travel time is determined based on the median of the first historical travel time. First baseline time is determined based on the first quantile of the first historical travel time. Second baseline time is determined based on the second quantile of the first historical travel time. The first quantile is greater than the second quantile. For each target travel chain in the target travel area pair, the second historical travel time corresponding to each operating time interval is extracted from the third historical travel data, and the target baseline time is determined based on the median of the second historical travel time. For each target travel chain in the target travel area pair, a time inappropriateness corresponding to the target travel chain in each operating time interval is generated based on the target baseline time, the reference historical travel time, the first baseline time, and the second baseline time.

4. The travel service recommendation method according to claim 1, characterized in that, The method for generating the aforementioned transfer discomfort includes the following steps: The reference number of transfers is determined based on the median number of transfers for all travel chains in the target area. The first benchmark number of transfers is determined based on the third quantile of the number of transfers for all travel chains. The second benchmark number of transfers is determined based on the fourth quantile of the number of transfers for all travel chains. The third quantile is greater than the fourth quantile. For each target travel chain in the target travel area pair, the target number of transfers for the target travel chain is obtained, and the transfer inadequacy corresponding to the target travel chain in each operating time interval is calculated based on the target number of transfers, the reference number of transfers, the first benchmark number of transfers, and the second benchmark number of transfers.

5. The travel service recommendation method according to claim 4, characterized in that, The method for generating the aforementioned congestion inadequacy includes the following steps: For each pair of target travel regions, the target travel chain is divided according to the target number of transfers to obtain multiple target travel sub-chains. For each target travel subchain, the in-vehicle dwell time and the passenger congestion coefficient corresponding to each operating time window are extracted from the third historical travel data. The operating time window includes a preset number of consecutive historical time intervals. For each target travel chain in the target travel area pair, the congestion inequality corresponding to the target travel chain in each operating time window is calculated based on the in-vehicle dwell time of the target, the vehicle congestion coefficient, and a preset reference in-vehicle dwell time. The reference in-vehicle dwell time is determined based on the fifth quantile of the in-vehicle dwell time of all targets in the third historical travel data.

6. The travel service recommendation method according to claim 1, characterized in that, The travel inadequacy assessment index includes a comprehensive travel inadequacy score, and the step of recommending travel services to the travel subject based on the travel inadequacy assessment index includes: When the comprehensive score of travel inappropriateness exceeds a preset score threshold, and the number of travelers in the target travel area within the operating time interval corresponding to the comprehensive score of travel inappropriateness exceeds a first preset threshold, carpooling service is recommended to the corresponding travelers. The system obtains the number of carpooling orders and the number of carpooling partners within a preset number of days. When the number of carpooling orders exceeds a preset order number threshold for each day within the preset number of days, and the number of carpooling partners exceeds a second preset threshold for each day, the system recommends a dedicated bus line service to the corresponding traveler.

7. A travel service recommendation device, characterized in that, The device includes: The construction unit is used to perform regional clustering based on the station location information of transportation stations in the target area to obtain multiple candidate travel area pairs, and to construct a candidate travel chain based on the starting station set and ending station set of each candidate travel area pair. The first extraction unit is used to obtain the first historical travel data of the target area in a preset historical period, and for each preset operating time interval, extract the second historical travel data of multiple historical time intervals that are the same as the operating time interval from the first historical travel data. The second historical travel data represents the data of the travel object entering and exiting the station from the station in the starting station set within the historical time interval and the station belonging to the corresponding ending station set. The acquisition unit is used to determine the maximum number of transfers of the candidate travel chain for each candidate travel area pair, and to acquire the target historical passenger flow within a preset number of consecutive historical time intervals from the second historical travel data. The filtering unit is used to filter out a target travel area pair from multiple candidate travel area pairs based on the maximum number of transfers, the target historical passenger flow, and preset filtering conditions. The second extraction unit is used to extract third historical travel data from the second historical travel data. The third historical travel data represents the data of stations entering and exiting from the target starting station set of the target travel area within the historical time interval, which belong to the corresponding target ending station set. The recommendation unit is used to generate a travel inappropriateness evaluation index for each of the operating time intervals based on the first historical travel data and the third historical travel data for each of the target travel area pairs, and to recommend travel services to the travel objects based on the travel inappropriateness evaluation index. The target travel area includes a target travel chain constructed from the target originating station set and the target ending station set. The step of generating travel inadequacy evaluation indicators for each operating time interval based on the first historical travel data and the third historical travel data includes: For each target travel chain of the target travel area pair, based on the first historical travel data and the third historical travel data, the transfer inappropriateness, time inappropriateness, and congestion inappropriateness of the target travel chain in each operating time interval are generated; For each target travel chain of the target travel area pair, extract the historical number of passengers corresponding to the target travel chain within each operating time interval from the third historical travel data; For each target travel area pair, a travel inappropriateness evaluation index is generated for each operating time interval based on the transfer inappropriateness, the time inappropriateness, the congestion inappropriateness, and the historical number of travelers.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the travel service recommendation method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the travel service recommendation method according to any one of claims 1 to 6.

Images