A passenger flow perception method and system based on big data

Abstract

This application relates to the field of passenger flow sensing technology, specifically to a passenger flow sensing method and system based on big data. The method includes: defining the time interval between the start of ticket sales for trains on the departure date and the start of ticket sales for buses as the pre-sale period; calculating the number of tickets sold for each train at each time point during the pre-sale period; obtaining the number of waitlist orders for each train at each time point after bus ticket sales begin; dividing the pre-sale period into multiple time periods and calculating the diversion index for each train in each time period; determining the diversion pressure index for each time period; obtaining the passenger pressure index for each empty train; determining the comprehensive passenger flow index for the current time point; determining the passenger flow assessment value; and adjusting bus schedules accordingly. This application, through dynamic sensing of passenger flow, can reasonably adjust bus schedules to alleviate passenger pressure in a timely manner, improve the effective diversion of passenger flow, and enhance transportation efficiency.

Description

Technical Field

[0001] This application relates to the field of passenger flow sensing technology, specifically to a passenger flow sensing method and system based on big data. Background Technology

[0002] Passenger flow sensing is a crucial component of smart transportation and urban planning. It aims to support traffic management, passenger service optimization, and resource allocation by monitoring and analyzing passenger flow changes at transportation hubs. Real-time monitoring of passenger flow through passenger flow sensing provides data support for improving transportation efficiency, enhancing passenger experience, and developing sound policies.

[0003] In the face of massive passenger flows during peak travel periods, increasing train frequency is quite challenging, as it significantly impacts the entire train network. Buses, on the other hand, can divert passenger traffic from trains, and increasing their frequency is relatively easy, thus reducing passenger pressure on trains. However, adjusting bus frequency requires accurate estimation of passenger flow pressure. If the passenger flow estimate is too low, insufficient bus frequency increases will not solve the peak-season passenger pressure problem. Conversely, if the passenger flow estimate is too high, excessive bus frequency will increase the operating costs and reduce the revenue of bus stations, leading to unreasonable bus frequency adjustments, hindering effective passenger diversion, and reducing transportation efficiency. Summary of the Invention

[0004] To address the aforementioned technical issues, a passenger flow perception method and system based on big data are provided to resolve existing problems.

[0005] The solution to the technical problem in this application is to provide a passenger flow perception method and system based on big data, including the following steps:

[0006] In a first aspect, embodiments of this application provide a passenger flow perception method based on big data, the method comprising the following steps:

[0007] The time interval between the time when train tickets go on sale and the time when bus tickets go on sale is recorded as the advance sale period; the number of tickets sold for each train at each time during the advance sale period is calculated; and the number of waitlist orders for each train at each time after bus tickets go on sale is obtained.

[0008] Based on the time when tickets for all train services sold out during the pre-sale period, the pre-sale period is divided into multiple time periods, and the corresponding sold-out train services are obtained.

[0009] Analyze the differences in the rate of change of ticket sales for each train in each time period and between adjacent time periods, and calculate the diversion index of each train in each time period; combine the differences in ticket sales for different trains in each time period with the differences in ticket sales for the corresponding sold-out trains in the time period before they sold out, and determine the diversion pressure index for each time period.

[0010] Based on the diversion index and diversion pressure index of each empty train for each train in all time periods before it is sold out, the passenger transport pressure index of each empty train for each train is obtained.

[0011] The bus routes for which tickets are sold out at the current time are denoted as the "bus routes with sold-out tickets". The changes in the number of waitlist orders for all train routes with sold-out tickets between the time when bus tickets went on sale and the current time are analyzed. Combined with the passenger transport pressure index, the comprehensive passenger flow index for the current time is determined. By analyzing the difference in departure times between the bus and train routes with sold-out tickets, the passenger flow assessment value for each bus route with sold-out tickets at the current time is determined, and the bus routes are adjusted accordingly.

[0012] Preferably, the step of dividing the pre-sale period into multiple time periods and obtaining the corresponding sold-out trains includes: recording the time when tickets for all trains within the pre-sale period are sold out as the sold-out time; dividing the pre-sale period into multiple time periods using the sold-out time as the dividing point; and recording the sold-out trains corresponding to each sold-out time as the corresponding sold-out trains.

[0013] Preferably, the calculation of the diversion index for each train service in each time period includes:

[0014] Starting from the time when each time period is sold out between it and the previous time period, number all the times in each time period and the time period before it along both sides.

[0015] For each time period and each moment in the preceding time period, the ticket sales rate is calculated for each time period and the preceding time period, respectively, based on the corresponding sequence number and ticket sales volume.

[0016] The difference in ticket sales rate between each train service and its preceding time period is recorded as the ticket sales difference; the maximum value between the ticket sales difference and a preset value is selected.

[0017] The diversion index is the result of normalizing the maximum value.

[0018] Preferably, the calculation process for the ticket sales rate is as follows:

[0019] The normalized results of the reciprocals of the sequence numbers corresponding to each train number in each time period and the time period preceding it are calculated and denoted as time weights.

[0020] Using the time weight corresponding to each moment within each time period as the weight, the ticket sales volume of each train service at all moments within each time period is weighted and summed, and recorded as the ticket sales rate of each time period.

[0021] Using the time weights corresponding to each moment within the previous time period as weights, the ticket sales volume of each train service at all moments within the previous time period is weighted and summed, and recorded as the ticket sales rate of the previous time period.

[0022] Preferably, determining the diversion pressure index for each time period includes:

[0023] For each time period that is sold out between itself and the previous time period, the average number of tickets sold for the corresponding train number that is sold out at all times in the previous time period is recorded as the first average.

[0024] Calculate the average number of tickets sold for each train at all times within each time period, and record it as the second average.

[0025] Calculate the normalized value of the difference between the first mean and the second mean, and perform a negative mapping on the normalized value;

[0026] The diversion pressure index is the sum of the product of the negative mapping result of all train services in each time period and the diversion index.

[0027] Preferably, the passenger transport pressure index is the sum of the products of the diversion index and the diversion pressure index for all time periods before the corresponding empty train time for each train.

[0028] Preferably, determining the comprehensive passenger flow index at the current moment includes:

[0029] The ratio of the number of waitlist orders for each empty train at the current moment to the number of waitlist orders for buses when they go on sale is recorded as a relative comparison.

[0030] The comprehensive passenger flow index is the sum of the products of the passenger pressure index of all sold-out train services corresponding to the current time and the relative comparison.

[0031] Preferably, determining the passenger flow assessment value for each sold-out bus trip at the current time includes:

[0032] Calculate the time interval between the departure times of any sold-out bus and any sold-out train at the current moment;

[0033] The product of the ratio of the passenger pressure index to the time interval for all sold-out train services is normalized, and the product of the normalized result and the comprehensive passenger flow index is used as the passenger flow assessment value for any sold-out bus service at the current time.

[0034] Preferably, the adjustment of bus routes includes: normalizing the passenger flow assessment value; if the normalization result is greater than or equal to a preset threshold, then adding a new bus route between each sold-out bus route corresponding to the bus and the next bus route; otherwise, not adding a new bus route.

[0035] Secondly, embodiments of this application also provide a passenger flow perception system based on big data, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the above-described passenger flow perception methods based on big data.

[0036] This application has at least the following beneficial effects:

[0037] This application divides the pre-sale period into multiple time periods, using the time when train tickets for a particular train are sold out as the dividing point. Based on the difference in ticket sales rate between adjacent time periods, it calculates the diversion index for each train in each time period. Its advantages lie in considering the ticket sales situation of other trains before and after the sold-out train at the dividing point, thus reflecting the diversion effect of other trains on the sold-out train. Secondly, it compares the ticket sales situation of other trains in each time period with the sold-out train... By analyzing the differences in ticket sales in the previous time period, the diversion pressure index for each time period is determined. Its beneficial effect lies in considering the ticket booking situation of other trains when diverting passengers from sold-out trains, thus reflecting the diversion pressure in each time period. The passenger transport pressure index for each sold-out train is obtained by considering the pressure changes of each sold-out train during all time periods before it sold out, thus reflecting the passenger flow borne by each sold-out train. This analysis aims to determine the extent to which buses are needed to divert passenger traffic for each available train. Furthermore, after bus tickets go on sale, the analysis examines the changes in the number of waitlist orders for all available trains between the time bus tickets went on sale and the current time. This analysis determines the current passenger flow index, which is beneficial because it assesses the passenger pressure on all buses by analyzing changes in waitlist orders, reflecting the effectiveness of buses in diverting passenger traffic from trains. Finally, it determines the passenger flow assessment value for each available bus at the current time, which is crucial for evaluating the bus's performance. Adjusting bus schedules has the advantage of assessing the passenger load on each empty bus route by analyzing the difference in departure times between sold-out bus and train routes. This reflects the diversion effect of each sold-out bus route on train routes, and helps determine whether additional bus routes should be added between the current sold-out bus route and the next available bus route. By dynamically sensing passenger flow, bus schedules can be adjusted rationally to alleviate passenger pressure in a timely manner, improve the effective diversion of passenger flow, and enhance transportation efficiency. Attached Figure Description

[0038] The following section provides a more detailed description of a big data-based passenger flow perception method according to this application, with reference to the accompanying drawings.

[0039] Figure 1 A flowchart illustrating the steps of a big data-based passenger flow perception method provided in this application embodiment;

[0040] Figure 2 A flowchart illustrating the steps of the method for obtaining the diversion pressure index for each time period provided in the embodiments of this application. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of a big data-based passenger flow perception method and system proposed in this application, in conjunction with the accompanying drawings and implementation examples, is provided. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0042] 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 pertains.

[0043] Please see Figure 1 The diagram illustrates a flowchart of a big data-based passenger flow perception method according to an embodiment of this application. The method includes the following steps:

[0044] Step 1: Describe the time interval between the start of ticket sales for train tickets on the departure date and the start of ticket sales for bus tickets as the pre-sale period; calculate the number of tickets sold for each train at each time during the pre-sale period; and obtain the number of waitlist orders for each train at each time after bus tickets go on sale.

[0045] Because bus ticket bookings and train ticket bookings complement each other, passenger flow for buses can be estimated by analyzing train ticket booking data, thereby obtaining train ticket booking data. Specifically:

[0046] Taking the journey from point A to point B as an example, since train tickets from point A to point B are generally sold in advance, tickets for all train services on the departure date are usually available for sale in advance. Since the sale time for train tickets is generally earlier than the sale time for bus tickets, the time period between the sale time for train tickets and the sale time for bus tickets on the departure date is called the advance sale period.

[0047] Statistical analysis of ticket sales at each time point during the aforementioned pre-sale period;

[0048] After the buses go on sale, obtain the number of waitlist orders for each train at each time slot;

[0049] In this embodiment, the number of tickets sold and the number of waitlist orders are counted for each hour. As for other implementation methods, the implementer can set them according to the actual situation.

[0050] Record the time when train tickets for each train service sold out during the pre-sale period, and denote it as each sell-out time; use the sell-out time as a dividing point to divide the pre-sale period into multiple time periods;

[0051] This gives us the number of tickets sold at each time during the pre-sale period, as well as the number of waitlist orders for each train at each time.

[0052] Step 2: Analyze the differences in the rate of change of ticket sales for each train service in each time period and between adjacent time periods, and calculate the diversion index of each train service in each time period.

[0053] During peak travel periods, tickets for multiple train services may sell out, but the timing of these sell-outs varies. Passengers tend to prefer trains with convenient departure times, leading to some trains becoming "hot" trains. Tickets for these hot trains sell out more quickly. When tickets for hot trains are sold out, passengers often choose other trains with similar departure times as alternatives and purchase tickets accordingly. The speed at which alternative trains sell out before and after the hot trains have sold out varies significantly.

[0054] Secondly, when tickets for popular trains are sold out, passengers will be diverted to other trains with similar departure times. Since the ticketing speed of other trains with similar departure times increases significantly when popular trains are sold out, the moment when the ticketing speed changes the most is near the moment when popular trains are sold out. Therefore, the possibility of being diverted can be judged by the changes in the ticketing speed of other trains before and after the moment when a sold-out train is sold out.

[0055] Since tickets for one train are sold out at each sell-out time, at least one train's tickets will be sold out between any two adjacent time periods. The difference in ticket sales rates for the same train between each time period and the previous time period is analyzed to calculate the diversion index, specifically:

[0056] Starting from the time when each time period is sold out between it and the previous time period, number all the times in each time period and the time period before it along both sides.

[0057] It should be noted that, for ease of understanding, we assume that for the th The time period and the first Taking the first time period as an example, if the first time period is... Each time point within a time period is denoted as follows: , , , No. Each time point within a time period is denoted as follows: , , No. The time period and the first The time slots between the time periods are as follows ,in, exist Time and Between moments, therefore, the time of sell-out Starting from the first, for the second All times within a time period , , If we number them according to the value 1, then The sequence number corresponding to the time is 3. The sequence number corresponding to the time is 2. The sequence number corresponding to time 1 is 1, for the first time All times within a time period , If we number them according to the value 1, then The sequence number corresponding to the time is 1. The sequence number corresponding to the time is 2.

[0058] The normalized result of the reciprocal of the sequence number corresponding to each train number at each time within each time period is calculated and denoted as the time weight.

[0059] The normalized result of the reciprocal of the sequence number corresponding to each time point in the preceding time period for each train number is calculated and denoted as the time weight.

[0060] In this embodiment, the softmax activation function is used for normalization. The softmax activation function is a well-known technology and will not be described in detail here. As other implementation methods, implementers can use other methods of the prior art, such as the sigmoid function, etc. This embodiment does not impose any special restrictions on this.

[0061] Using the time weights corresponding to each moment within each time period as weights, the ticket sales volume of each train at all moments within each time period is weighted and summed to obtain the ticket sales rate of each train in each time period.

[0062] Using the time weights corresponding to each moment in the previous time period as weights, the ticket sales volume of each train service at all moments in the previous time period is weighted and summed to obtain the ticket sales rate of each train service in the previous time period.

[0063] In this embodiment, the softmax activation function is used for normalization. The softmax activation function is a well-known technology and will not be described in detail here. As other implementation methods, implementers can use other methods of the prior art, such as the sigmoid function, etc. This embodiment does not impose any special restrictions on this.

[0064] It should be noted that the closer the time is to the sold-out time, the greater the time weight is assigned, reflecting that the ticket sales situation at the time closer to the sold-out time has a more important impact on the diversion of passengers.

[0065] The difference in ticket sales rate between each train service and the previous time period is recorded as the ticket sales difference.

[0066] In this embodiment, the difference in ticket sales rate between each train service in each time period and the previous time period is recorded as the ticket sales difference.

[0067] The maximum value between the ticket sales difference and the preset value is selected, and the maximum value is normalized to serve as the diversion index for each train in each time period.

[0068] In this embodiment, the preset value is 0. Next, the sigmoid function is used for normalization. The sigmoid function is a well-known technique and will not be elaborated upon here. As for other implementation methods, the implementer can use other methods from existing technologies, such as the tanh function, softmax activation function, etc. This embodiment does not impose any special restrictions on this. Furthermore, the formula for calculating the diversion index of each train number in each time period is as follows:

[0069]

[0070] in, For the first The train number is at Diversion index for each time period; For the first The train number is at Ticket sales rate for a given time period; For the first The train number is at Ticket sales rate for a given time period To find the maximum value function, This is the normalization function.

[0071] It should be noted that the ticket sales difference reflects the change in the ticket sales rate of other trains before and after a certain train's tickets are sold out. When the ticket sales rate changes significantly, that is, the larger the diversion index, the more significant the impact of the other trains on passenger diversion after the tickets for the train corresponding to the sold-out time are sold out.

[0072] Thus, the diversion index of each train service in each time period is obtained.

[0073] Step 3: By comparing the differences in ticket sales between different train numbers in different time periods and the time periods before the corresponding sold-out trains sold out, and combining this with the diversion index, determine the diversion pressure index for each time period; based on the diversion index and diversion pressure index of each sold-out train in all time periods before it sold out, obtain the passenger transport pressure index of each sold-out train.

[0074] Since selling out tickets for popular train routes generally alleviates passenger pressure, this manifests in train ticket bookings as a slowdown in booking speed for trains with high diversion indices after the diversion. However, if the booking speed for trains with high diversion indices doesn't change much or even increases, it indicates that passenger pressure remains high after the diversion. Therefore, based on the changes in ticket purchasing speed for trains with high diversion indices before and after the diversion, the diversion pressure in each time period is estimated, and a diversion pressure index is calculated. The flowchart of the method for obtaining the diversion pressure index for each time period provided in this application embodiment is shown below. Figure 2 As shown. Specifically:

[0075] Record the train numbers that are sold out at each sold-out time as the train numbers that are sold out.

[0076] For each time period that is sold out between itself and the previous time period, the average number of tickets sold for the corresponding train number at all times in the previous time period is recorded as the first average.

[0077] It should be noted that there is a dividing point between each time period and the previous time period. This dividing point is the time when tickets for a train are sold out. The average number of tickets sold for that train in the time period before the dividing point is denoted as the first average.

[0078] Calculate the average number of tickets sold for each train at all times within each time period, and record it as the second average.

[0079] Calculate the normalized value of the difference between the first mean and the second mean, and perform a negative mapping on the normalized value;

[0080] In this embodiment, the normalized value of the difference between the first mean and the second mean is calculated. Then, the sigmoid function is used for normalization. The sigmoid function is a well-known technique and will not be described in detail here. As other implementation methods, implementers can use other methods of the prior art, such as the tanh function, softmax activation function, etc. This embodiment does not impose any special restrictions on this. The negative mapping process is as follows: the normalized value is negatively mapped by the value 1, and the difference between the value 1 and the normalized value is taken as the result of the negative mapping.

[0081] The sum of the products of the negative mapping results of all train services in each time period and the diversion index is used as the diversion pressure index for each time period.

[0082] It should be noted that the larger the negative mapping result, the more the ticket sales of other trains increase after the tickets for the train corresponding to the sold-out time are sold out. This reflects that the other trains have diverted the passenger flow from the sold-out trains. The larger the diversion pressure index, the higher the ticket booking speed remains after the diversion by other trains during this period, indicating that there is still significant passenger pressure after the diversion.

[0083] Secondly, during peak travel periods, each train whose tickets are sold out represents a currently popular train, and the popularity is ranked according to the order in which tickets are sold out. If, within a short period after a particular train sells out, other trains with a high diversion index also sell out and still experience high diversion pressure, it indicates that the passenger pressure index of that particular train is higher. Therefore, analyzing the diversion index and the diversion pressure index to determine the passenger pressure index is as follows:

[0084] Arrange the sold-out train numbers corresponding to each sold-out time according to the sold-out time, and record them as the sold-out train numbers corresponding to each train.

[0085] The sum of the products of the diversion index and the diversion pressure index for each empty train service corresponding to a train in all time periods before its corresponding empty time is calculated as the passenger pressure index for each empty train service corresponding to the train.

[0086] It should be noted that the higher the passenger transport pressure index, the greater the passenger flow that the empty train is carrying, the more likely that the empty train is the train that passengers expect to take, and the greater the need for bus diversion for the empty train.

[0087] Thus, the passenger pressure index for each empty train route is obtained.

[0088] Step 4: Analyze the changes in the number of waiting orders for all empty train services between the time when bus tickets went on sale and the current time. Combine this with the passenger pressure index to determine the comprehensive passenger flow index for the current time. Then, based on the difference in departure times between the empty bus and train services, determine the passenger flow assessment value for each empty bus service at the current time and adjust the bus schedule accordingly.

[0089] Furthermore, currently, a waitlist option exists when booking train tickets. The passenger pressure index for each train service can accurately reflect passengers' main intended trains from the entire ticket order. The decrease in train waitlist orders after bus tickets go on sale reflects the degree to which buses divert passenger flow pressure from trains, thus allowing for an assessment of bus passenger flow after ticket sales begin. Therefore, analyzing the changes in the number of waitlist orders for corresponding sold-out train services after bus ticket sales begin determines the comprehensive passenger flow index, specifically:

[0090] The number of waitlist orders for each train service when bus tickets go on sale is recorded; and after bus tickets go on sale, all bus services that are currently sold out are recorded as the respective sold-out bus services.

[0091] The ratio of the number of waitlist orders for each empty train at the current moment to the number of waitlist orders for buses when they go on sale is recorded as a relative comparison.

[0092] The sum of the products of the passenger pressure index and the relative comparison for all sold-out trains corresponding to the current time is taken as the comprehensive passenger flow index for the current time.

[0093] It should be noted that the larger the relative ratio, the more waiting orders there are for the empty trains, even after the bus diversion, reflecting the poor diversion effect of the buses. The larger the passenger flow comprehensive index, the more remaining waiting orders there are for the empty trains and the greater the passenger flow pressure on the trains, indicating that the overall diversion effect of the buses is insufficient.

[0094] Secondly, the passenger flow for each sold-out bus route varies. Furthermore, when the bus departure time is close to the train departure time, it attracts more passengers who haven't been able to purchase train tickets. Therefore, the closer the departure time of a sold-out bus is to the departure time of a sold-out train with a higher passenger flow pressure index, the greater the passenger flow the bus will handle. By analyzing the passenger flow for each sold-out bus route and calculating the passenger flow assessment value, we can determine whether to add a new bus between the departure time of the sold-out bus and the departure time of the next bus to alleviate passenger flow pressure. Specifically:

[0095] Calculate the time interval between the departure times of any sold-out bus and any sold-out train at the current moment;

[0096] The ratio of the passenger pressure index to the time interval for all sold-out train services is normalized, and the product of the normalization result and the comprehensive passenger flow index is used as the passenger flow assessment value for any sold-out bus service at the current time.

[0097] It should be noted that, when calculating the ratio, to avoid the denominator being 0, a preset value greater than 0 is added to the denominator. The range of this preset value greater than 0 is [range missing]. In this embodiment, the preset value greater than 0 is 0.1. As for other implementation methods, the implementer can set it according to the actual situation. Secondly, the smaller the time interval, the more passengers tend to choose alternative transportation with similar departure times, reflecting the stronger diversion effect of the sold-out bus service on the train service. The larger the passenger pressure index, the more likely the sold-out train service is the train service that passengers want to take, and the greater the passenger flow pressure of the sold-out train service. The larger the passenger flow assessment value, the higher the passengers' intention to take the sold-out bus service with a similar departure time to the sold-out train service, and the larger the number of people who want to take the sold-out bus service.

[0098] The passenger flow assessment value is normalized. If the normalization result is greater than or equal to a preset threshold, a new bus is added between each sold-out bus and the next bus; otherwise, no new bus is added.

[0099] In this embodiment, the preset threshold value is 0.7. As for other implementation methods, the implementer can set it according to the actual situation.

[0100] It should be noted that if the normalization result is greater than or equal to the preset threshold, the passenger transport pressure can be alleviated by adding more bus trips. By dynamically sensing the passenger flow, the bus trips can be adjusted reasonably to alleviate the passenger transport pressure in a timely manner and improve the effective diversion of passenger flow and transportation efficiency.

[0101] Based on the same inventive concept as the above methods, this application also provides a big data-based passenger flow sensing system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described big data-based passenger flow sensing methods.

[0102] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0103] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0104] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application. Therefore, any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of this application, without departing from the content of the technical solution of this application, shall fall within the protection scope of the technical solution of this application.

Claims

1. A passenger flow perception method based on big data, characterized in that, The method includes the following steps: The time interval between the time when train tickets go on sale and the time when bus tickets go on sale is recorded as the advance sale period; the number of tickets sold for each train at each time during the advance sale period is calculated; and the number of waitlist orders for each train at each time after bus tickets go on sale is obtained. Based on the time when tickets for all train services sold out during the pre-sale period, the pre-sale period is divided into multiple time periods, and the corresponding sold-out train services are obtained. Analyze the differences in the rate of change of ticket sales for each train in each time period and between adjacent time periods, and calculate the diversion index of each train in each time period; combine the differences in ticket sales for different trains in each time period with the differences in ticket sales for the corresponding sold-out trains in the time period before they sold out, and determine the diversion pressure index for each time period. Based on the diversion index and diversion pressure index of each empty train for each train in all time periods before it is sold out, the passenger transport pressure index of each empty train for each train is obtained. The bus routes for which tickets are sold out at the current time are denoted as the sold-out bus routes. The changes in the number of waitlist orders for all sold-out train routes between the time when bus tickets went on sale and the current time are analyzed. Combined with the passenger transport pressure index, the comprehensive passenger flow index for the current time is determined. By analyzing the difference in departure times between the sold-out bus routes and train routes, the passenger flow assessment value for each sold-out bus route at the current time is determined, and the bus routes are adjusted accordingly. The step of dividing the pre-sale period into multiple time periods and obtaining the corresponding sold-out train services includes: recording the time when tickets for all train services within the pre-sale period are sold out as the sold-out time; dividing the pre-sale period into multiple time periods using the sold-out time as the dividing point; and recording the sold-out train services corresponding to each sold-out time as the corresponding sold-out train services. The calculation of the diversion index for each train service in each time period includes: Starting from the time when each time period is sold out between it and the previous time period, number all the times in each time period and the time period before it along both sides. For each time period and each moment in the preceding time period, the ticket sales rate is calculated for each time period and the preceding time period, respectively, based on the corresponding sequence number and ticket sales volume. The difference in ticket sales rate between each train service and the previous time period is recorded as the ticket sales difference; the maximum value between the ticket sales difference and a preset value is selected. The diversion index is the result of normalizing the maximum value. The calculation process for the ticket sales rate is as follows: The normalized results of the reciprocals of the sequence numbers corresponding to each train number in each time period and the time period preceding it are calculated and denoted as time weights. Using the time weight corresponding to each moment within each time period as the weight, the ticket sales volume of each train service at all moments within each time period is weighted and summed, and recorded as the ticket sales rate of each time period. Using the time weights corresponding to each moment in the previous time period as weights, the ticket sales volume of each train service at all moments in the previous time period is weighted and summed, and recorded as the ticket sales rate of the previous time period. The determination of the diversion pressure index for each time period includes: For each time period that is sold out between itself and the previous time period, the average number of tickets sold for the corresponding train number that is sold out at all times in the previous time period is recorded as the first average. Calculate the average number of tickets sold for each train at all times within each time period, and record it as the second average. Calculate the normalized value of the difference between the first mean and the second mean, and perform a negative mapping on the normalized value; The diversion pressure index is the sum of the products of the negative mapping results of all train services in each time period and the diversion index. The passenger transport pressure index is the sum of the products of the diversion index and the diversion pressure index for all time periods before the corresponding empty train time for each train; The determination of the current passenger flow comprehensive index includes: The ratio of the number of waitlist orders for each empty train at the current moment to the number of waitlist orders for buses when they go on sale is recorded as a relative comparison. The passenger flow comprehensive index is the sum of the products of the passenger pressure index of all sold-out trains corresponding to the current time and the relative comparison. The determination of the passenger flow assessment value for each sold-out bus at the current time includes: Calculate the time interval between the departure times of any sold-out bus and any sold-out train at the current moment; The product of the ratio of the passenger pressure index to the time interval for all sold-out train services is normalized, and the product of the normalization result and the comprehensive passenger flow index is used as the passenger flow assessment value for any sold-out bus service at the current time. The adjustment of bus routes includes: normalizing the passenger flow assessment value; if the normalization result is greater than or equal to a preset threshold, then adding a new bus route between each sold-out bus route and the next bus route; otherwise, not adding a new bus route.

2. A passenger flow sensing system based on big data, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the big data-based passenger flow perception method as described in claim 1.

Images