Bus stop board content generation method in combination with pedestrian flow analysis

Abstract

The invention discloses a bus stop board content generation method combined with people flow analysis, and relates to the technical field of information processing, and the method comprises the steps: obtaining the bus operation line data of a target bus stop, monitoring and analyzing the station flow data of a certain line and the in-vehicle flow of a target vehicle, and predicting the in-vehicle flow when the vehicle arrives at the station; in combination with the route prediction flow of the platform, if the total flow exceeds a vehicle bearing threshold, activating a hitchhiking linkage platform, and generating and optimizing a hitchhiking route sequence; and mapping the optimized hitchhiking route information to each bus route according to a predetermined layout scheme, and dynamically displaying the optimized hitchhiking route information on a platform display screen. The technical problem that in an existing bus stop board content updating mode, passenger flow fluctuation is difficult to respond in time is solved, and the technical effects that the real-time performance and accuracy of bus stop board content updating are improved, so that passenger flow fluctuation is responded in time, and passenger distribution and traffic resource utilization are optimized are achieved.

Description

Technical Field

[0001] The present application relates to the field of information processing technology, and in particular to a method for generating bus stop content in combination with pedestrian flow analysis. Background Art

[0002] With the acceleration of urbanization, public transportation systems play an important role in the daily travel of urban residents. However, the content update method of traditional bus stops usually relies on fixed schedules or manual adjustments, which makes it difficult to reflect passenger flow fluctuations in a timely manner, especially during peak hours or special circumstances, which can easily lead to information lag and inefficient management. In recent years, with the development of computer technology, big data analysis and passenger flow monitoring technology, dynamic data processing has gradually been applied in bus information release systems. However, the existing system still has room for improvement in the response efficiency of real-time information collection, analysis and content generation, and it is difficult to meet actual operational needs. In this context, there is an urgent need for a bus stop content generation method that can be based on dynamic analysis of passenger flow to cope with complex passenger flow change scenarios and achieve more efficient information update and resource allocation.

[0003] At present, there is a technical problem in the relevant technologies that it is difficult to respond to passenger flow fluctuations in a timely manner in the way the bus stop content is updated. Summary of the invention

[0004] The present application solves the technical problem that it is difficult to respond to passenger flow fluctuations in a timely manner in existing bus stop content updating methods by providing a bus stop content generation method combined with passenger flow analysis.

[0005] This application provides a method for generating bus stop content combined with pedestrian flow analysis, including: Acquire multiple bus operation routes of a target bus stop, and randomly select a first operation route, monitor and acquire several station flow data of several bus stops in the first operation route, and in-vehicle flow data of a target vehicle, wherein the target vehicle is the bus closest to the target bus stop in the first operation route; predict the in-vehicle flow of the target vehicle when it arrives at the target bus stop based on the several station flow data and in-vehicle flow data, and acquire the predicted in-vehicle flow; acquire the first route predicted flow of the target bus stop, if the sum of the predicted in-vehicle flow and the first route predicted flow exceeds the on-board threshold value of the target vehicle, activate a ride-sharing linkage platform; acquire a first ride-sharing route set based on the ride-sharing linkage platform, evaluate the first ride-sharing route set according to the first operation route, generate a first ride-sharing sequence, and add it to multiple ride-sharing sequences; according to a predetermined layout plan, map and distribute the multiple ride-sharing sequences on the multiple bus operation routes, and dynamically display them on the display screen of the target bus stop.

[0006] The method for generating bus stop content combined with passenger flow analysis proposed in this application first obtains the bus operation route data of the target bus stop, monitors and analyzes the station flow data of a certain route and the in-vehicle flow of the target vehicle, and predicts the in-vehicle flow when the vehicle arrives at the station; combined with the predicted flow of the platform's routes, if the total flow exceeds the vehicle carrying threshold, activates the ride-sharing linkage platform, generates and optimizes the ride-sharing route sequence; according to the predetermined layout plan, maps the optimized ride-sharing route information to each bus route, and dynamically displays it on the platform display screen, thereby improving the real-time and accuracy of bus stop content updates, thereby achieving timely response to passenger flow fluctuations and optimizing passenger diversion and transportation resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0008] Figure 1 A flowchart of a method for generating bus stop content in combination with pedestrian flow analysis provided in an embodiment of the present application.

[0009] Figure 2 A schematic diagram of the first ride-sharing sequence production process of the bus stop content generation method combined with passenger flow analysis provided in an embodiment of the present application. DETAILED DESCRIPTION

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0011] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0012] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0013] The present application embodiment provides a method for generating bus stop content in combination with pedestrian flow analysis, such as Figure 1 As shown, the method includes: Step S100, obtain multiple bus operation routes of the target bus station, and randomly select the first operation route, monitor and obtain several station flow data of several bus stops in the first operation route, and the in-vehicle flow data of the target vehicle, wherein the target vehicle is the bus closest to the target bus station in the first operation route. Specifically, the system first connects to the bus operation management database or related data interface to obtain multiple bus operation route information associated with the target bus station, which covers the route number, starting station, passing stations and operation time, so as to build a basic data framework. Then, a random algorithm is used to select the first operation route from these routes to ensure a comprehensive and objective assessment of the platform operation status. For this route, with the help of flow monitoring equipment such as infrared sensing counters or computer vision-based cameras deployed at the entrance and exit of the station, the flow data of the stations between the target vehicle and the target station are monitored, and the data are collected and recorded at specific time intervals to form a station flow data set. At the same time, the target vehicle closest to the target platform in the first operating line is located, and its in-vehicle flow data is monitored in real time by relying on the passenger counting sensor installed at the door or the combination of the on-board camera and the image recognition algorithm to grasp the vehicle's passenger status and provide a starting data point for predicting the change trend of the in-vehicle flow when it arrives at the target platform.

[0014] Step S200, based on the several station flow data and in-vehicle flow data, predict the in-vehicle flow of the target vehicle when it arrives at the target bus stop, and obtain the predicted in-vehicle flow. Specifically, firstly, the target vehicle passes through the stations according to its real-time position and the target bus stop position, and integrates the flow data of the passing stations and the in-vehicle flow data. For each passing station, multiple passenger flow fluctuation factors such as surrounding commercial activities, weather, time period, etc. are collected and input into the passenger flow fluctuation analyzer (a machine learning model trained based on a large amount of historical data) to obtain multiple passenger flow fluctuation coefficients. According to the order in which vehicles arrive at the passing stations, gradient weights (the sum of 1 and N, N is positively correlated with the arrival order) are configured to compensate for the passenger flow fluctuation coefficient to obtain multiple compensated fluctuation coefficients. Based on this, the flow data of the passing stations are compensated to obtain the predicted flow of multiple passing stations. The bus operation log is queried to obtain the average proportion of boarding and alighting flow at the passing stations. The additional in-vehicle flow is calculated based on the predicted flow of the passing stations. The additional in-vehicle flow is summed with the initial in-vehicle flow data, and finally the predicted in-vehicle flow when the target vehicle arrives at the target bus stop is obtained. This result can provide an important reference for bus operation management, help to judge in advance whether the vehicle is overloaded and take corresponding dispatching measures to improve operational efficiency and service quality, and ensure passengers' travel experience.

[0015] In a possible implementation, according to the several station flow data and the in-vehicle flow data, the in-vehicle flow of the target vehicle when arriving at the target bus station is predicted, and the predicted in-vehicle flow is obtained. Step S200 further includes step S210, screening the several bus stations according to the real-time position of the target vehicle and the position of the target bus station, determining multiple passing stations, and obtaining the flow data of multiple passing stations. Specifically, first, the system will track the location information of the target vehicle in real time, and at the same time clarify the fixed position of the target bus station. By comparing the relative position relationship between the two, the stations that the target vehicle will pass through in the process of driving to the target station are screened from the several bus station data previously obtained, and the screened stations are multiple passing stations. In order to obtain the flow data of the passing stations, corresponding flow monitoring equipment is deployed at each station. For example, high-precision infrared sensors are installed at the entrance and exit of the station, which can accurately count the changes in the number of passengers getting on and off the bus; or cameras based on image recognition technology are used to shoot and analyze the flow of people in the station area, so as to obtain the flow data of the station in different time periods. The data is collected and stored in real time for subsequent in-depth analysis and processing, providing basic data support for predicting the flow of traffic inside the target vehicle when it arrives at the target station.

[0016] Step S220, collect multiple crowd flow fluctuation factors of the multiple passing sites, and input the multiple crowd flow fluctuation factors into the crowd flow fluctuation analyzer respectively, and output multiple crowd flow fluctuation coefficients, wherein the crowd flow fluctuation analyzer is a machine learning model. Specifically, for the determined multiple passing sites, start to collect multiple crowd flow fluctuation factors. Factors cover many aspects, such as the "commuting" factor. During the morning and evening peak hours on weekdays, the commuting of employees in office buildings, factories and other areas around the site will cause a significant increase in the site's passenger flow; while during non-peak hours, the passenger flow is relatively stable. "Weather" is also a key factor. In bad weather such as heavy rain, heavy snow, high temperature or severe cold, people's willingness to travel will be suppressed, and the passenger flow of the site will be significantly reduced; on the contrary, in clear and comfortable weather, the number of people traveling tends to increase. The "going to and from school" factor is also not to be ignored. During the school going and going home time, the bus stops near the school will usher in the peak of student flow, and the passenger flow will fluctuate significantly. The collected multiple crowd fluctuation factors, such as the proportion of office workers in the surrounding office buildings of a certain station, the quantitative indicators of the current weather conditions, the matching degree between the school start and end time and the current time, are input into the pre-built and trained crowd fluctuation analyzer. The crowd fluctuation analyzer is a model trained by machine learning algorithms based on a large amount of historical data. It can output multiple crowd fluctuation coefficients based on the characteristics and data of various crowd fluctuation factors input. The coefficient can quantitatively reflect the degree to which the crowd flow of each station is affected by different fluctuation factors. For example, a coefficient of 0.8 may indicate that the station is greatly affected by a certain current fluctuation factor, while a coefficient of 0.2 indicates that the impact is relatively small.

[0017] Step S230, configure the gradual weight according to the order in which the vehicle arrives at the multiple passing stations, compensate the multiple passenger flow fluctuation coefficients, and determine multiple compensation fluctuation coefficients, wherein the gradual weight is the sum of 1 and N, and N is positively correlated with the order of vehicle arrival. Specifically, after obtaining multiple passenger flow fluctuation coefficients, considering that different passing stations are at different distances from the target bus station, the degree of influence on the in-vehicle flow when the target vehicle finally arrives at the target station is also different, so it is necessary to configure the gradual weight according to the order in which the vehicle arrives at the multiple passing stations. Specifically, the gradual weight is set to the sum of 1 and N, where N is positively correlated with the order of vehicle arrival, that is, the station that the vehicle is about to arrive at (i.e., the station closer to the target station) has a larger corresponding N value and a larger weight. For example, for the station closest to the target station, N may be 5, and its weight is 1+5=6; while for the station slightly farther away, N is 3, and the weight is 4. Using these configured gradual weights, the compensation calculation of the multiple passenger flow fluctuation coefficients obtained previously is performed. The calculation method is to multiply the passenger flow fluctuation coefficient of each station by its corresponding gradient weight to obtain multiple compensation fluctuation coefficients. The purpose of this is to highlight the importance of passenger flow fluctuations at stations closer to the target platform to the final vehicle flow prediction, because the passenger flow changes at these stations are more likely to directly affect the number of people in the target vehicle when it arrives at the target platform.

[0018] Step S240, compensating the flow data of the multiple passing sites according to the multiple compensation fluctuation coefficients, and obtaining the predicted flow of the multiple passing sites. Specifically, based on the multiple compensation fluctuation coefficients calculated above, compensatory processing is performed on the flow data of the multiple passing sites to obtain the predicted flow of the multiple passing sites. The specific operation is to multiply the original flow data of each passing site with the corresponding compensation fluctuation coefficient or calculate other specific compensation algorithms. For example, the original boarding flow data of a passing site is 20 people, and its compensation fluctuation coefficient is 1.2, then after compensation, the predicted boarding flow of the site becomes 20*1.2=24 people. By performing such compensation calculations on the boarding flow and alighting flow data of each passing site, a series of predicted flow data of multiple passing sites that are more in line with the actual situation and take into account the influence of the fluctuation factors of the flow of people are obtained. Compared with the original flow data, these predicted flow data can more accurately reflect the possible actual flow changes when the vehicle passes through each site after considering various fluctuation factors of the flow of people, and provide a more reliable data basis for the final prediction of the in-vehicle flow when the target vehicle arrives at the target platform.

[0019] Step S250, based on the predicted flow and in-vehicle flow data of the multiple passing sites, the in-vehicle flow of the target vehicle when it arrives at the target bus station is predicted, and the predicted in-vehicle flow is output. Specifically, the predicted flow of the multiple passing sites and the in-vehicle flow data of the target vehicle initially monitored are used for comprehensive calculation to predict the in-vehicle flow of the target vehicle when it arrives at the target bus station and output the result. First, query the bus operation log to obtain the average proportion of the boarding flow and the average proportion of the alighting flow of each passing site on the first operating line. For example, the average proportion of the boarding flow of a passing site is 30%, and the average proportion of the alighting flow is 20%. Combined with the predicted boarding flow and alighting flow data of the site, the net increase or decrease of the in-vehicle flow of the site can be calculated. The impact of each passing site on the in-vehicle flow is cumulatively calculated, and the initial in-vehicle flow data is added to obtain the predicted in-vehicle flow of the target vehicle when it arrives at the target bus station. For example, initially there are 30 people in the bus, and after calculating the flow changes at multiple stations, the net increase is 15 people, so the final predicted flow in the bus is 30+15=45 people. This prediction result can provide an important reference for bus operation and scheduling, such as judging in advance whether the vehicle will be overloaded, so as to take corresponding measures in time, such as adjusting the departure frequency, allocating empty vehicle resources, or coordinating with other transportation modes, so as to improve the efficiency and service quality of bus operations and better meet the travel needs of passengers.

[0020] In a possible implementation, based on the predicted traffic and in-vehicle traffic data of the multiple passing sites, the in-vehicle traffic of the target vehicle is predicted when it arrives at the target bus stop, and the predicted in-vehicle traffic is output. Step S250 further includes step S251, querying the bus operation log to obtain the average of multiple boarding traffic proportions and the average of multiple alighting traffic proportions of the first operating route of the multiple passing sites. Specifically, the system will establish a connection with the bus operation log database, which stores a large amount of historical data on bus operating routes, including the number of boardings and alightings at each station in different time periods and vehicle operation information. For the multiple passing sites previously determined and the first operating route currently concerned, the average of the boarding traffic proportions and the average of the alighting traffic proportions of these passing sites on the first operating route during previous operations are extracted from the database. For example, for a certain stop A, the system will count the proportion of the number of passengers boarding each time the vehicle stops at the stop during the operation period of the first operating line in the past month or longer, and calculate the average of these proportions, which is the mean proportion of boarding flow at the stop. Similarly, the mean proportion of alighting flow at the stop will also be calculated. These mean data reflect the average level of passenger boarding and alighting flow relative to the total passenger flow of the vehicle under normal operation of the stop, providing an important historical reference for the subsequent accurate prediction of changes in vehicle flow.

[0021] Step S252, based on the average of the multiple boarding flow ratios, the average of the multiple alighting flow ratios and the predicted flow of the multiple passing stations, the additional in-vehicle flow is calculated to obtain the additional in-vehicle flow. Specifically, after obtaining the average of the boarding flow ratios and the average of the alighting flow ratios of the multiple passing stations, the additional in-vehicle flow is calculated in combination with the predicted flow of the multiple passing stations that have been obtained before. For each passing station, according to its predicted boarding flow and the corresponding average of the boarding flow ratio, the actual additional number of boardings that the station may bring to the vehicle can be calculated. For example, if a certain passing station predicts a boarding flow of 30 people and its average boarding flow ratio is 40%, then the actual number of new boardings that may be added to the vehicle at the station is 30×40%=12 people. Similarly, according to the predicted alighting flow and the average of the alighting flow ratio, the number of alightings that may actually be reduced at the station can be calculated. Subtract the number of alightings from the number of new boardings at each station to obtain the net increase in the in-vehicle flow at the station. The net increase of all the stations passed through is accumulated and calculated to get the total additional in-vehicle flow. For example, the net increase of station A is 5 people, the net increase of station B is -2 people (indicating that the number of people getting off the bus is greater than the number of people getting on the bus, and the in-vehicle flow is reduced), and the net increase of station C is 8 people, so the total additional in-vehicle flow is 5+(-2)+8=11 people.

[0022] Step S253, summing the newly added in-vehicle flow and the in-vehicle flow data to obtain the predicted in-vehicle flow. Specifically, summing the newly added in-vehicle flow calculated above with the in-vehicle flow data of the target vehicle initially monitored. Assuming that the in-vehicle flow data initially monitored is 20 people and the newly added in-vehicle flow is 11 people, the predicted in-vehicle flow is 20+11=31 people. This predicted in-vehicle flow result intuitively reflects the approximate number of passengers of the target vehicle when it arrives at the target bus stop. Based on this prediction result, the bus operation management department can make corresponding scheduling arrangements in advance. For example, when the predicted in-vehicle flow is close to or exceeds the full load capacity of the vehicle, it can consider increasing the departure frequency of the line, deploying empty vehicles to the station for passenger diversion, or linking with other modes of transportation (such as taxis, online car-hailing, etc.) to guide passengers to choose other travel modes, thereby avoiding vehicle overloading and ensuring the safety and efficiency of bus operations and the travel comfort of passengers.

[0023] Step S300, obtain the predicted flow rate of the first route of the target bus station. If the sum of the predicted in-vehicle flow rate and the predicted flow rate of the first route exceeds the on-board threshold value of the target vehicle, activate the ride-sharing linkage platform. Specifically, first extract the historical operation data of the target bus station on the first operating route from the bus operation system database, and combine special factors such as the current time period such as weekdays, holidays, and large-scale events in the surrounding area. Use specific data analysis algorithms and models, such as time series analysis methods, to predict the average proportion of the target station's boarding and alighting flow on the route based on historical passenger flow patterns, and combine the predicted flow data of the target station. The predicted flow of the first route is calculated by the predetermined formula. Then the predicted in-vehicle flow and the predicted flow of the first route are obtained and summed, and the sum is compared with the vehicle threshold value determined according to the vehicle design passenger capacity, safety regulations and operating experience. If the sum of the two is greater than the vehicle threshold value, it means that the target vehicle faces the risk of overloading when it arrives at the target platform, which will make the passenger experience poor and affect safety. At this time, the ride-sharing linkage platform is automatically activated to integrate the surrounding ride-sharing resources to provide passengers with additional travel options, so as to optimize the allocation of transportation resources, achieve the coordinated operation of buses and ride-sharing, and improve the overall efficiency of the urban public transportation system.

[0024] In a possible implementation, the predicted flow of the first route of the target bus station is obtained. If the sum of the predicted in-vehicle flow and the predicted flow of the first route exceeds the vehicle threshold of the target vehicle, the ride-sharing linkage platform is activated. Step S300 further includes step S310, obtaining the average of the target boarding flow ratio and the target alighting flow ratio of the first operating route of the target bus station, and selecting the target station predicted flow of the target bus station from the predicted flow of the multiple passing stations. Specifically, the system will deeply mine the historical data resources of bus operation, and the data covers the detailed flow information of each station of the first operating route over a long period of time in the past. For the target bus station, the average of the target boarding flow ratio and the target alighting flow ratio of the station in the first operating route are calculated by statistically analyzing the historical data. For example, in the operating data of the past month, the proportion of the number of passengers who get on the station every day to the total passenger flow change of the line on that day (the number of passengers getting on the bus minus the number of passengers getting off the bus) is counted, and the average value is calculated to obtain the average of the target boarding flow ratio; similarly, the average of the target alighting flow ratio is calculated. At the same time, the predicted traffic data of the target station corresponding to the target bus station is accurately selected from the previously obtained predicted traffic data of multiple passing stations. This data is obtained based on the collection and analysis of the fluctuation factors of the flow of people around the target station, combined with the coefficients output by the flow fluctuation analyzer and the configuration of gradual weight compensation. It can more accurately reflect the actual traffic situation of the target station when the target vehicle arrives under various current influencing factors.

[0025] Step S320, calculate the predicted flow of the first route according to the target mean of the proportion of boarding flow, the target mean of the proportion of alighting flow and the predicted flow of the target station. Specifically, after obtaining the above key data, the predicted flow of the first route can be calculated. Assume that the target mean of the proportion of boarding flow is , the average proportion of target alighting flow is , the target site predicted traffic is Since the boarding flow and the alighting flow will have different effects on the final passenger flow in the car, in order to accurately predict the flow change of the vehicle at the target station, these factors need to be considered comprehensively. The calculation process is as follows: First, calculate the actual impact of the number of people boarding at the target station on the flow in the car, that is, (Here is the remaining proportion after considering the impact of the alighting flow, so that the proportion of boarding flow can be reasonably reflected in the change of total passenger flow). This calculation result is the possible increase in passenger flow of the vehicle at the target station after considering the characteristics of the alighting and alighting flow at the target station and the predicted flow, that is, the predicted flow of the first line. Through this calculation method, it is possible to more accurately estimate the passenger flow situation of the first operating line itself when the target vehicle arrives at the target bus station, and the trend of the number of passengers in the vehicle, which provides important data basis for subsequent judgments on whether additional traffic allocation measures (such as activating the ride-sharing linkage platform, etc.) are needed.

[0026] Step S400, based on the ride-sharing linkage platform, obtain the first ride-sharing route set, evaluate the first ride-sharing route set according to the first operating route, generate a first ride-sharing sequence, and add it to multiple ride-sharing sequences. Specifically, when the bus operation system activates the ride-sharing linkage platform because the sum of the predicted in-vehicle flow and the predicted flow of the first route exceeds the on-board threshold value of the target vehicle, the platform uses the intelligent matching algorithm in its ride-sharing resource database to select a batch of ride-sharing routes with starting points or passing points near the platform and end points close to the end point of the first operating route based on the location of the target bus stop and the key information of the first operating route, forming the first ride-sharing route set. Then, for each route in the route set, the similarity between the start and end points is calculated with the target bus station as the starting point and the end point of the first operating route as the reference, which is negatively correlated with the sum of the distances between the start and end points; the path overlap with the first operating route, that is, the proportion of the overlapping path length, and the time overlap are analyzed, and the departure and travel time of the ride and the bus are comprehensively considered; at the same time, the vehicle credibility is obtained, and a comprehensive evaluation is made based on the driver's historical order records, the basic and personal information of the vehicle, etc. The route fitness is obtained based on the start and end point similarity, path overlap, time overlap and vehicle credibility evaluation, and the first ride sequence is generated by arranging the first ride route from large to small according to the fitness, and finally it is added to multiple ride sequences corresponding to all operating routes of the bus station, so as to provide passengers with comprehensive travel selection information, help the bus operation management department to allocate resources, and improve the efficiency and service quality of the transportation system.

[0027] In one possible implementation, Figure 2As shown, based on the hitchhiking linkage platform, the first hitchhiking route set is obtained, the first hitchhiking route set is evaluated according to the first operating route, a first hitchhiking sequence is generated, and added to multiple hitchhiking sequences. Step S400 further includes step S410, obtaining the target end point of the first operating route, taking the target bus stop as the target starting point, performing similarity analysis on the first hitchhiking route set according to the target starting point and the target end point, and obtaining multiple start and end point similarities, wherein the start and end point similarity is negatively correlated with the sum of the start and end point distances. Specifically, the target end point of the first operating route is clarified, which is the final destination of the passengers on the route. At the same time, the target bus stop is set as the target starting point, representing the current departure point of the passenger. For each route in the first hitchhiking route set, the sum of the distances between its starting point and the target starting point and between its end point and the target end point is calculated through a geographic information system (GIS) or a special path calculation algorithm. For example, if the starting point of a hitchhiking route is 2 kilometers away from the target bus stop and the end point is 5 kilometers away from the target end point of the first operating route, then the sum of the start and end point distances is 7 kilometers. The similarity between the start and end points is associated with the sum of the distances according to a specific mathematical model. Generally speaking, the smaller the sum of the distances, the higher the similarity between the start and end points. For example, using the inverse proportional function model, the similarity between the start and end points = 1 / (1 + the sum of the start and end point distances). When the sum of the distances is 7 kilometers, the corresponding start and end point similarity value is calculated. This calculation is performed for each route in the first ride-sharing route set to obtain multiple start and end point similarity values. These values ​​will intuitively reflect the degree of fit between each ride-sharing route and the start and end point needs of bus passengers.

[0028] Step S420, analyze the path overlap and time overlap of the first ride-sharing route set according to the first operating route, and obtain multiple path overlaps and multiple time overlaps. Specifically, for each route in the first ride-sharing route set, its driving path is compared in detail with the path of the first operating route. Using map data and path analysis technology, calculate the proportion of the length of the overlapping part of the two routes to the total length of the ride-sharing route, which is the path overlap. For example, if the total length of a ride-sharing route is 10 kilometers, of which the part overlapping with the first operating route is 4 kilometers, then its path overlap is 4 / 10=0.4. In terms of time overlap analysis, it is necessary to integrate various time information. First, determine the estimated time for the first operating route vehicle to arrive at the target bus stop, which can be predicted based on the real-time position of the vehicle, the driving speed and the traffic conditions of the route. Then obtain the estimated departure time, driving time and arrival time of the ride-sharing route in an area close to the first operating route. By comparing these time information, it is determined whether the ride-sharing can be connected with the travel time of bus passengers at the appropriate time node. For example, if the bus is expected to arrive at the target station in 10 minutes, and a rideshare route is expected to pass by the station in 8-12 minutes and the subsequent travel time has a large overlap with the time range of the bus arrival at the destination, then this rideshare route has a good performance in terms of time overlap. By analyzing each rideshare route, multiple path overlap and multiple time overlap values ​​are obtained.

[0029] Step S430, obtain the credibility of multiple vehicles in the first ride-sharing route set, evaluate the first ride-sharing route set according to the similarities of the multiple start and end points, the overlap of multiple paths, the overlap of multiple times and the credibility of multiple vehicles, and obtain multiple route fitness. Specifically, the acquisition of vehicle credibility is based on the historical number of operations and the evaluation of operational effects. For each vehicle in the first ride-sharing route set, query its historical operation data. If a vehicle has a large number of historical operations, for example, more than 50 orders in the past month, it means that the vehicle is relatively active in the market and has certain operating experience. At the same time, analyze its operational effect evaluation, including passenger ratings, complaints, etc. If the average passenger rating is above 4.5 points (out of 5 points) and the complaint rate is less than 5%, it indicates that the vehicle has a good operational effect. Combining these factors, a vehicle credibility value is assigned to each vehicle by weighted average and other methods, for example, the weight of the number of operations is 0.4, and the weight of the operational effect evaluation is 0.6, and the vehicle credibility is calculated. After obtaining multiple start-end similarities, multiple path overlaps, multiple time overlaps, and multiple vehicle credibility, the first ride-sharing route set is evaluated using a comprehensive evaluation model. For example, the start-end similarity weight can be set to 0.3, the path overlap weight to 0.3, the time overlap weight to 0.2, and the vehicle credibility weight to 0.2. According to the formula: route fitness = start-end similarity × 0.3 + path overlap × 0.3 + time overlap × 0.2 + vehicle credibility × 0.2, the route fitness value of each ride-sharing route is calculated, thereby obtaining multiple route fitness.

[0030] Step S440, arrange the multiple first ride-sharing routes in the first ride-sharing route set according to the route fitness from large to small, and generate the first ride-sharing sequence. Specifically, all the first ride-sharing routes in the first ride-sharing route set are sorted according to the calculated route fitness value. Starting from the route with the highest route fitness, arrange them in descending order. For example, if there are three ride-sharing routes, and their route fitness are 0.8, 0.6, and 0.4 respectively, then the order after arrangement is that the route with a fitness of 0.8 ranks first, 0.6 ranks second, and 0.4 ranks third, thus generating the first ride-sharing sequence. This sequence will serve as an important basis for subsequent recommendations to bus passengers. Passengers can give priority to ride-sharing with high route fitness according to this sequence to improve the convenience, comfort and efficiency of travel.

[0031] In a possible implementation, multiple first ride-sharing routes in the first ride-sharing route set are arranged from large to small according to route fitness, and the first ride-sharing sequence is generated. Step S440 further includes step S441, based on the bus operation log of the target bus station, analyzing and obtaining the average proportion of pedestrian collaborative travel when the target vehicle arrives at the target bus station. Specifically, the massive data information contained in the bus operation log of the target bus station is deeply mined. The log records in detail the travel characteristics of passengers who took the bus at the target bus station in different time periods in the past, including multi-dimensional data such as the time of passengers getting on and off the bus, the number of passengers getting on and off the bus, and whether there are companions. By performing a detailed statistical analysis on these data, the relevant data records of the target vehicle when it arrives at the target bus station are screened out. Then, the proportion of the number of pedestrian collaborative travel (i.e., riding with companions) in the total number of travels is calculated, and the average value is further calculated, so as to accurately determine the average proportion of pedestrian collaborative travel when the target vehicle arrives at the target bus station. For example, in the past month's operation data, there were 100 trip records when the target vehicle arrived at the station, and 30 of them involved pedestrians. The average proportion of pedestrians involved was 30%. This average data reflects a behavioral pattern and trend of passengers traveling at the station, and provides an important reference for the subsequent allocation of ride-sharing resources.

[0032] Step S442, set the hitchhiking compensation weight according to the mean value of the pedestrian collaborative travel ratio, and optimize and adjust the first hitchhiking sequence according to the hitchhiking compensation weight, wherein the hitchhiking compensation weight is positively correlated with the remaining space in the car. Specifically, after obtaining the mean value of the pedestrian collaborative travel ratio, set the hitchhiking compensation weight based on this. If the mean value of the pedestrian collaborative travel ratio is high, for example, it exceeds the predetermined index (assuming the predetermined index is 40%), this means that there is a high probability that multiple people will travel together at this station. At this time, a higher compensation weight is given to the hitchhiking with more remaining space in the car. Specifically, weights are allocated according to the number of remaining spaces in each hitchhiking car. For example, one hitchhiking car has 4 remaining spaces and the other has 2 remaining spaces. When the mean value of the pedestrian collaborative travel ratio is high, the hitchhiking compensation weight of the former will be higher than that of the latter. Then, the first hitchhiking sequence generated previously is optimized and adjusted based on these set hitchhiking compensation weights. In the sorting process, the compensation weight is taken into consideration, so that the position of the ride-sharing cars whose interior space can better meet the needs of collaborative travel can be improved in the sequence. For example, a ride-sharing car with a larger interior space that was originally ranked at the back of the first ride-sharing sequence may be moved to a higher position after considering the ride-sharing compensation weight because it is more suitable for traveling with multiple people, thereby providing more suitable travel options for possible collaborative travel passengers, further improving the matching degree between ride-sharing resources and public transportation passenger needs, and optimizing the entire transportation service system.

[0033] Step S500, according to the predetermined layout plan, the multiple hitchhiking sequences are mapped and distributed on the multiple bus operating routes, and dynamically displayed on the display screen of the target bus stop. Specifically, based on the overall flow conditions of multiple bus operating routes, the flow difference of each line station is calculated (the sum of the in-vehicle flow and the line predicted flow minus the on-board threshold value), and the parameters such as the number of hitchhiking, display duration and screen share are set accordingly to generate a predetermined layout plan. For example, routes with large flow differences are allocated with more hitchhiking information, longer display duration and larger screen share. Then, multiple hitchhiking sequences are accurately matched and associated with the corresponding bus operating routes according to the plan, and the hitchhiking route information with high similarity to the starting and ending points of the routes, good path overlap and time overlap is selected, and the display area is allocated as required. Finally, the target bus stop display screen will display the ride-sharing information of each route in a cycle according to the scheduled duration and proportion, and use animation or color to highlight key content. For example, high-credibility vehicles are marked in green, and vehicles with many remaining seats have the number of seats displayed in large fonts. This allows passengers to intuitively understand bus congestion forecasts and ride-sharing resources, facilitate travel planning, and improve the collaborative operation service effect of buses and ride-sharing and the level of public transportation intelligence.

[0034] In a possible implementation, after the multiple hitchhiking sequences are mapped and distributed on the multiple bus operation routes according to a predetermined layout plan, and dynamically displayed on the display screen of the target bus station, step S500 further includes step S510, calculating the flow difference of multiple stations of the multiple bus operation routes, wherein the flow difference of the station is the sum of the in-vehicle flow and the line predicted flow minus the flow difference of the on-board threshold value. Specifically, for each bus operation route, its related in-vehicle flow data, line predicted flow data and corresponding target vehicle on-board threshold value data must be obtained. The in-vehicle flow data can be obtained by real-time monitoring of the passenger counting device installed on the bus, and the line predicted flow data is calculated based on the historical operation data of the line, the characteristics of the current time period (such as working days, holidays, peak and flat peaks, etc.) and the passenger flow prediction of the passing stations. The on-board threshold value is a fixed value set according to the design passenger capacity of the vehicle and combined with safety and operational comfort requirements. Then, for each station, according to the calculation formula of the station flow difference, that is, the station flow difference = the sum of the in-vehicle flow and the line predicted flow minus the flow difference of the on-board threshold value, accurate calculation is performed. For example, the in-vehicle flow of a station on a certain line is 30 people, the line predicted flow is 15 people, and the on-board threshold value is 40 people, then the flow difference of the station is (30+15)-40=5 people. By performing such calculations one by one for each station of multiple bus operating lines, multiple station flow difference data can be obtained, which can intuitively reflect the passenger flow tension and vehicle load pressure at each station of each line.

[0035] Step S520, the predetermined layout scheme is obtained according to the flow difference analysis of the multiple sites. Specifically, after obtaining a large amount of site flow difference data, a predetermined layout scheme is formulated based on this. A comprehensive statistical analysis is performed on the flow differences of these sites, focusing on the size distribution of the flow difference. Since the larger the flow difference, the more likely it is that the line where the site is located will be congested at the site, and the higher the demand for ride-sharing resources. Therefore, according to the association rule between the flow difference and the ride-sharing display resource, that is, the larger the flow difference, the longer the ride-sharing display time on the screen and the more the number, the specific layout scheme is determined. For example, for the line where the site has a large flow difference, a longer ride-sharing information display time is planned on the bus station display screen, such as 2-3 minutes, and a large number of ride-sharing routes, such as 5-8, are displayed at the same time; and for the line site with a small flow difference, the ride-sharing display time can be shortened to about 1 minute, and the number of displays is correspondingly reduced to 2-3. In addition, other layout details such as the allocation of display areas and the arrangement of display order can be further refined according to different levels of flow difference. Through such analysis and planning, we will eventually determine a predetermined layout plan that can reasonably allocate bus stop display screen resources so that the display of ride-sharing information closely matches the actual passenger flow demand of each bus operation route, so as to improve the efficiency and service quality of the coordinated operation of buses and ride-sharing, better meet the travel needs of passengers and optimize the allocation of urban public transportation resources.

[0036] In a possible implementation, the predetermined layout plan is obtained based on the analysis of the multiple site traffic differences. Step S520 further includes step S521 of setting a plurality of hitchhiking car quantities, a plurality of display durations, and a plurality of screen ratios according to the multiple site traffic differences, where the site traffic differences are positively correlated with the hitchhiking car quantities, display durations, and screen ratios. Specifically, the calculated multiple site traffic differences are sorted and classified, and arranged in descending order to clearly understand the traffic pressure conditions of each site. Corresponding hitchhiking car resource allocation strategies are set for different traffic difference ranges. For example, when the site traffic difference is in a relatively high range (such as greater than a certain threshold T1), it indicates that the passenger flow pressure on the line where the site is located is extremely high. At this time, a relatively large number of hitchhiking cars are set, such as 8 - 10, a relatively long display duration, which can be set to 3 - 5 minutes, and a relatively large screen ratio, which may reach 60% - 80% of the display screen. This is because under such high traffic pressure, more hitchhiking car options are needed to relieve the bus transportation pressure. The relatively long display duration ensures that passengers have enough time to understand the hitchhiking car information in detail, and the relatively large screen ratio makes it more prominent among numerous pieces of information and easier for passengers to notice. If the site traffic difference is in a medium range (T2 < difference < T1), the number of hitchhiking cars is moderately reduced to 4 - 7, the display duration is adjusted to 2 - 3 minutes, and the screen ratio is set between 30% - 60%. For sites with relatively small traffic differences (difference < T2), correspondingly fewer hitchhiking cars are configured, such as 1 - 3, a relatively short display duration, about 1 minute, and the screen ratio is between 10% - 30%. Through such hierarchical settings, the hitchhiking car resources can be reasonably allocated according to the actual traffic demands of the sites, giving full play to their role in relieving the bus passenger flow pressure and improving the resource utilization efficiency.

[0037] Step S522, based on the number of hitchhikers, the display durations and the screen ratios, the predetermined layout scheme is generated. Specifically, after determining the number of hitchhikers, the display duration and the screen ratio corresponding to the flow difference of different stations, the predetermined layout scheme is constructed. First, the overall layout architecture of the bus stop display screen is determined, and different areas are divided for displaying bus information and hitchhiker information of different routes. For each bus operation line, the resource allocation of the line on the display screen is determined according to the number of hitchhikers, the display duration and the screen ratio corresponding to the largest station flow difference in each station. For example, if the flow difference of a station in line A is in the high interval, then the hitchhiker display area allocated to line A on the display screen will be set according to the parameters corresponding to the high interval, that is, 8-10 hitchhiker information will be displayed, the display duration is 3-5 minutes, and a larger proportion of the screen space will be occupied. In terms of the display order, it can be arranged according to the importance of the line, the size of the passenger flow or the alphabetical order to ensure that passengers can easily and quickly find the hitchhiker information of the line they are concerned about. At the same time, the frequency of information updates must also be considered to ensure that the ride-sharing information can reflect the actual situation in a timely manner, such as changes in the location of the vehicle, whether it is full, etc. Through such detailed planning and arrangement, a complete scheduled layout plan is generated, which can effectively display the ride-sharing information on the bus station display screen, and cooperate with the bus operation line information to provide passengers with comprehensive, accurate and easy-to-obtain travel references, thereby improving the overall service quality and efficiency of the coordinated operation of buses and ride-sharing.

[0038] The embodiment of the present application obtains the bus operation route data of the target bus stop, monitors and analyzes the station flow data of a certain line and the in-vehicle flow of the target vehicle, and predicts the in-vehicle flow when the vehicle arrives at the station; combined with the predicted flow of the platform line, if the total flow exceeds the vehicle carrying threshold, activates the hitchhiking linkage platform, generates and optimizes the hitchhiking route sequence; according to the predetermined layout plan, maps the optimized hitchhiking route information to each bus route, and dynamically displays it on the platform display screen, thereby improving the real-time and accuracy of the bus stop content update, thereby achieving timely response to passenger flow fluctuations and optimizing passenger diversion and transportation resource utilization. Technical effect.

[0039] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for generating bus stop content in combination with pedestrian flow analysis, characterized in that: Methods include: Acquire multiple bus operation routes of a target bus stop, randomly select a first operation route, monitor and acquire a plurality of station flow data of a plurality of bus stops in the first operation route, and in-vehicle flow data of a target vehicle, wherein the target vehicle is a bus in the first operation route that is closest to the target bus stop; Predicting the in-vehicle flow of the target vehicle when it arrives at the target bus stop based on the plurality of station flow data and in-vehicle flow data, and obtaining the predicted in-vehicle flow; Obtain the predicted flow rate of the first route of the target bus stop, and if the sum of the predicted in-vehicle flow rate and the predicted flow rate of the first route exceeds the on-board threshold value of the target vehicle, activate the ride-sharing linkage platform; Acquire a first ride-sharing route set based on the ride-sharing linkage platform, evaluate the first ride-sharing route set according to the first operating route, generate a first ride-sharing sequence, and add it to multiple ride-sharing sequences; According to a predetermined layout plan, the multiple hitchhiking sequences are mapped and distributed on the multiple bus operation routes, and dynamically displayed on the display screen of the target bus stop.

2. The method for generating bus stop content in combination with pedestrian flow analysis according to claim 1, characterized in that: Predicting the in-vehicle flow of the target vehicle when it arrives at the target bus stop according to the plurality of station flow data and in-vehicle flow data, and obtaining the predicted in-vehicle flow, including: The plurality of bus stops are screened according to the real-time position of the target vehicle and the position of the target bus stop, a plurality of passing stops are determined, and traffic data of the plurality of passing stops are obtained; Collecting multiple crowd flow fluctuation factors of the multiple passing sites, and inputting the multiple crowd flow fluctuation factors into a crowd flow fluctuation analyzer respectively, and outputting multiple crowd flow fluctuation coefficients, wherein the crowd flow fluctuation analyzer is a machine learning model; According to the order in which the vehicles arrive at the multiple passing stations, a gradual weight is configured, the multiple passenger flow fluctuation coefficients are compensated, and multiple compensation fluctuation coefficients are determined, wherein the gradual weight is the sum of 1 and N, and N is positively correlated with the order in which the vehicles arrive; Compensating the flow data of the plurality of passing sites according to the plurality of compensation fluctuation coefficients to obtain predicted flow of the plurality of passing sites; Based on the predicted traffic flow and in-vehicle traffic flow data of the multiple passing sites, the in-vehicle traffic flow of the target vehicle is predicted when it arrives at the target bus stop, and the predicted in-vehicle traffic flow is output.

3. The method for generating bus stop content in combination with pedestrian flow analysis according to claim 2, characterized in that: According to the predicted traffic flow and in-vehicle traffic flow data of the plurality of passing stations, predicting the in-vehicle traffic flow of the target vehicle when it arrives at the target bus station, and outputting the predicted in-vehicle traffic flow, including: Query the public transportation operation log to obtain multiple average boarding flow ratios and multiple average alighting flow ratios of the first operating route passing through the multiple stations; Based on the average of the multiple boarding flow rates, the average of the multiple alighting flow rates and the predicted flow rates of the multiple passing stations, additional in-vehicle flow rate is calculated to obtain additional in-vehicle flow rate; The newly added in-vehicle flow rate and the in-vehicle flow rate data are summed to obtain the predicted in-vehicle flow rate.

4. The method for generating bus stop content in combination with pedestrian flow analysis according to claim 3 is characterized in that: Obtaining the predicted flow of the first route of the target bus stop, including: Obtaining the average of the target boarding flow ratio and the target alighting flow ratio of the first operating route of the target bus stop, and selecting the target station predicted flow of the target bus stop from the plurality of passing station predicted flows; The predicted flow of the first route is calculated based on the average target boarding flow ratio, the average target alighting flow ratio and the predicted flow of the target site.

5. The method for generating bus stop content combined with pedestrian flow analysis according to claim 1, characterized in that: The first rideshare route set is evaluated according to the first operating route to generate a first rideshare sequence, including: Obtain a target destination of the first operating route, take a target bus stop as the target starting point, perform similarity analysis on the first rideshare route set according to the target starting point and the target destination, and obtain multiple start and end point similarities, wherein the start and end point similarities are negatively correlated with the sum of the start and end point distances; Performing path overlap and time overlap analysis on the first rideshare route set according to the first operating route to obtain multiple path overlaps and multiple time overlaps; Acquire multiple vehicle credibility of the first ride-sharing route set, evaluate the first ride-sharing route set according to the multiple start-end similarities, multiple path overlaps, multiple time overlaps and multiple vehicle credibility, and obtain multiple route fitness; Arrange multiple first ride-sharing routes in the first ride-sharing route set according to route fitness from large to small to generate the first ride-sharing sequence.

6. The method for generating bus stop content in combination with pedestrian flow analysis according to claim 5, characterized in that: Generating the first ride-sharing sequence includes: Based on the bus operation log of the target bus stop, analyzing and obtaining the average proportion of pedestrian cooperative travel when the target vehicle arrives at the target bus stop; A ride-sharing compensation weight is set according to the average proportion of pedestrian collaborative travel, and the first ride-sharing sequence is optimized and adjusted according to the ride-sharing compensation weight, wherein the ride-sharing compensation weight is positively correlated with the remaining space in the car.

7. The method for generating bus stop content in combination with pedestrian flow analysis according to claim 1, characterized in that: Get a pre-defined layout plan, including: Calculating the flow difference of multiple stations of the multiple bus operation routes, wherein the flow difference of the station is the flow difference obtained by subtracting the vehicle threshold value from the sum of the in-vehicle flow and the route predicted flow; The predetermined layout plan is obtained based on the flow difference analysis of the multiple sites.

8. The method for generating bus stop content in combination with pedestrian flow analysis according to claim 7, characterized in that: The predetermined layout scheme is obtained according to the flow difference analysis of the plurality of sites, including: According to the plurality of site traffic differences, a plurality of numbers of rideshares, a plurality of display durations, and a plurality of screen proportions are set, wherein the site traffic difference is positively correlated with the number of rideshares, the display duration, and the screen proportion; The predetermined layout plan is generated based on the multiple numbers of hitchhiking vehicles, multiple display durations, and multiple screen proportions.

Images