Method and system for publishing bus arrival time with minimum expected off-board time cost

By obtaining passenger travel characteristics and off-car time value coefficient, establishing an off-car time cost model, optimizing the publication of bus arrival time, the problems of customization of bus services and passenger travel experience in the existing technology are solved, and the effective time outside the car is achieved Utilization and improvement of bus service quality.

CN114925930BActive Publication Date: 2024-11-22WUHAN UNIV OF TECH
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202210644980.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-06-08
Publication Date
2024-11-22
Estimated Expiration
2042-06-08

AI Technical Summary

Technical Problem

The existing bus arrival information APP fails to customize the station arrival information services according to the individual needs of passengers, resulting in poor quality of bus services and passenger travel experience, and passengers are unable to effectively use the time outside the car.

Method used

By obtaining passenger travel characteristic parameters and off-vehicle time value coefficient, a cost model for expected off-vehicle time is established, and the minimum cost arrival time is recommended, combining passenger walking distance and additional activity selection, matching the expected trips, and optimizing the arrival time release.

Benefits of technology

Customized arrival time releases have been achieved, guiding passengers to make reasonable decisions, saving time and cost outside the vehicle, and improving the quality of bus services and passenger experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114925930B_ABST
    Figure CN114925930B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and system for publishing a bus arrival time with the minimum expected off-board time cost, wherein the method comprises the following steps: S1, obtaining the travel parameters of passengers, specifically including: the passenger's travel purpose, destination location, departure location, and boarding station location information; S2, obtaining the passenger's travel time value coefficient; S3, obtaining the arrival time prediction interval of each shift upstream of the boarding station of the travel route; S4, matching the expected boarding shift in combination with the passenger's walking distance; S5, establishing an expected off-board time cost model; S6, publishing the arrival time with the minimum expected cost. The present invention mainly addresses the problem of passengers taking additional activities after receiving the bus arrival time information, and adopts a method for publishing the bus arrival time with the minimum expected off-board time cost to guide passengers to make reasonable additional activity decisions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of urban public transportation systems, and in particular to a method and system for publishing bus arrival time with minimum expected off-vehicle time cost. Background Art

[0002] In recent years, the urbanization process has accelerated significantly, the number of motor vehicles has increased year by year, the contradiction between urban transportation demand and supply has become more prominent, and traffic congestion has become an urgent problem to be solved. Public bus and tram transportation (abbreviated as "public transportation"), as a large-capacity and affordable urban transportation mode, plays an important role in alleviating urban traffic congestion.

[0003] In recent years, the volume of public transportation has been declining year by year, and optimizing the quality of public transportation services is an important task at present. In public transportation operation services, the expected arrival time information directly affects the passengers' additional activity selection behavior. Reasonable additional activity selection behavior can effectively optimize the utilization of off-board time and thus optimize the service quality of public transportation operations.

[0004] After market research, it was found that commonly used real-time bus arrival information APPs include "The Bus is Coming" and "Bus on Palm", etc., which all focus on pushing the real-time location and estimated arrival time information of the nearest bus to users, and do not customize bus arrival information services based on the individual needs of bus passengers.

[0005] Therefore, in order to optimize the quality and attractiveness of bus services, it is crucial to seek a bus arrival time release strategy that can customize arrival information services according to passengers' travel characteristics and make full use of the off-bus time to optimize the off-bus time utility. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a bus arrival time publishing system and method with the expected minimum off-board time cost, which can guide passengers to make reasonable additional activity decisions, realize the effective use of off-board time, improve the quality and attractiveness of bus services from the economic effect level, and improve the travel experience of bus passengers.

[0007] The technical problem to be solved by the present invention is to provide a method for publishing bus arrival time with the minimum expected off-board time cost, comprising the following steps:

[0008] S1. Obtaining the passenger's travel characteristic parameters, including: the passenger's travel purpose, destination location, departure location, and boarding station location information;

[0009] S2. Obtain the off-board time value coefficients of the passengers in each stage of the process of catching the bus. The process of catching the bus refers to the stage from the departure of the passengers from the starting point to the arrival at the station to take the bus and leave. The time value coefficients in this process include: the time value coefficient β of the passengers walking to the station, the time value coefficient α of the passengers' additional activities, the time value coefficient γ of the passengers waiting for the expected bus at the station, and the time value coefficient δ of the passengers waiting for the expected bus at the station;

[0010] S3, obtaining the predicted arrival time interval of each bus service upstream of the boarding station of the travel route;

[0011] S4, matching the expected bus frequency with the passenger's walking distance;

[0012] S5. Establishing the expected off-vehicle time cost model;

[0013] S6. Publish the arrival time with the minimum expected cost.

[0014] According to the above scheme, the expected off-board time cost of the bus in step S5 is composed of the walking time cost of the passenger to the station, the time cost of the passenger's additional activities, and the waiting time cost of the passenger at the station. The waiting time cost of the passenger at the station depends on whether the passenger can take the expected bus after selecting the additional activities. The expected off-board time cost C of the passenger has two expressions:

[0015] 1) When the passenger completes the additional activities and leaves on the expected bus, the passenger's waiting time at the station is the actual arrival time of the expected bus, t r The time t when the passenger completes the additional activities and arrives at the station f +t s The difference between the passenger's expected off-vehicle time cost C 1 Expressed as:

[0016] C 1 =βt f +αt s +γ(t r -t f -t s )

[0017] 2) When a passenger completes the additional activities and misses the expected bus, and leaves on the next bus, the passenger's waiting time at the station is the actual arrival time of the next bus, t r +h, h is the vehicle arrival interval, and the time t when the passenger completes the additional activities and arrives at the station f +t s The difference between the passenger's expected off-vehicle time cost C 2 Expressed as:

[0018] C 2 =βt f +αt s +γ(tr +ht f -t s ).

[0019] According to the above scheme, the expected off-board time cost of the passenger in step S5 is the sum of the product of the off-board time cost and its occurrence probability in two cases: when the passenger completes the additional activities and leaves on the expected bus, and when the passenger completes the additional activities and misses the expected bus and leaves on the next bus. The calculation method is as follows:

[0020]

[0021] In the formula, f(t s ) is the probability density function of the additional activity time of bus passengers, f(t r ) is the probability density function of the expected actual arrival time, t l is the lower limit of the confidence interval of the expected flight arrival time, t u is the upper limit of the confidence interval of the expected arrival time of the train.

[0022] According to the above scheme, the recommended value of arrival time in step S6 is solved according to the following steps:

[0023] S601. Determine the value range of the arrival time release value t based on the prediction accuracy of the arrival information service system and the expected flight arrival time prediction value, that is, t∈(t l ,t u );

[0024] S602: Substitute the minute integer value within the value range into the public transport passenger's expected off-vehicle time cost function, calculate the passenger's expected off-vehicle time cost, and select the arrival time prediction value corresponding to the minimum expected cost as the arrival time recommendation value.

[0025] The present invention also provides a bus arrival time publishing system with the minimum expected off-board time cost, comprising a passenger travel parameter acquisition module, a passenger off-board time value coefficient acquisition module, a travel route boarding station upstream bus arrival time interval prediction module, an expected bus schedule matching module, an expected off-board time cost calculation module and an arrival time publishing module;

[0026] The passenger travel parameter acquisition module is used to obtain the travel destination of the traveling passenger, the location of the passenger's travel destination, the location of the passenger's departure point, and the location information of the boarding station;

[0027] The passenger off-vehicle time value coefficient acquisition module is used to obtain the passenger off-vehicle time value coefficient at each stage of the process of catching the bus, and pass the passenger travel characteristic parameters to the expected off-vehicle time cost calculation module;

[0028] The travel route upstream bus station arrival time interval prediction module is used to obtain the arrival time interval information of each bus upstream of the travel route station, and pass it to the expected travel schedule matching module;

[0029] The expected bus schedule matching module matches the expected bus schedule in combination with the passenger's walking distance, and transmits the expected bus schedule to the expected off-board time cost calculation module;

[0030] The expected off-vehicle time cost calculation module determines the off-vehicle time cost corresponding to each integer release value in the prediction interval, and transmits the arrival time release value with the minimum off-vehicle time cost to the arrival time release module;

[0031] The arrival time publishing module is used to provide the recommended arrival time value to passengers.

[0032] According to the above scheme, the passenger travel parameter acquisition module passes the passenger travel purpose to the passenger off-vehicle time value coefficient acquisition module, passes the boarding station location and destination location information to the upstream bus arrival time interval prediction module of the boarding station of the travel route, and passes the departure location and boarding station location to the expected bus schedule matching module.

[0033] According to the above scheme, the passenger off-vehicle time value coefficient acquisition module determines the off-vehicle time value coefficients of the passengers at each stage of the process of catching the bus in combination with the passenger's travel purpose information, including: the time value coefficient β of the passenger walking to the station, the time value coefficient α of the passenger's additional activities, the time value coefficient γ of the passenger waiting at the station for the expected bus, and the time value coefficient δ of the passenger waiting at the station when the passenger misses the expected bus, and passes it to the expected off-vehicle time cost calculation module.

[0034] According to the above scheme, the module for predicting the arrival time interval of the upstream trains at the boarding station determines the passenger's travel route based on the passenger's destination location and the boarding station location, predicts the arrival time intervals of each train upstream of the boarding station, and passes it to the expected travel schedule matching module.

[0035] According to the above scheme, the expected off-board time cost calculation module constructs an expected off-board time cost function with the expected shift arrival time release value as the independent variable, determines the off-board time cost corresponding to each integer release value in the prediction interval by trial and error, and passes the arrival time release value with the smallest off-board time cost to the arrival time release module.

[0036] The system and method for publishing bus arrival time with the minimum expected off-board time cost according to the present invention have the following beneficial effects:

[0037] 1. The bus arrival time publishing method proposed in the present invention can combine passenger travel parameters, including passenger travel purpose, destination location, departure point location, boarding station location and other information, to provide passengers with customized arrival time, which can guide passengers to make reasonable additional activity decisions and improve the time utility outside the bus;

[0038] 2. The present invention quantifies the off-vehicle time cost of passengers and finds that optimizing the arrival time publishing value to guide passengers to choose additional activities can effectively save the expected off-vehicle time cost of passengers from the departure point to the departure stage. Through case analysis, it is found that compared with directly publishing the predicted expected value (point estimate), publishing the recommended value can effectively save the expected off-vehicle time cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A flowchart of a method for publishing bus arrival time with minimum expected off-vehicle time cost according to the present invention;

[0040] Figure 2 The bus arrival time prediction error distribution of the bus arrival time publishing method with the minimum expected off-vehicle time cost of the present invention;

[0041] Figure 3 The additional activity duration distribution of passengers in the method for publishing bus arrival time with the minimum expected off-vehicle time cost of the present invention;

[0042] Figure 4 It is the changing trend of the passenger's expected off-vehicle time cost with the published value of the expected arrival time of the present invention. DETAILED DESCRIPTION

[0043] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.

[0044] like Figure 1-4 In the embodiment of the bus arrival time publishing system that takes into account the minimum expected off-board time cost of the present invention, it includes a passenger travel parameter acquisition module, a passenger off-board time value coefficient acquisition module, a travel route boarding station upstream bus arrival time interval prediction module, an expected bus schedule matching module, an expected off-board time cost calculation module and an arrival time publishing module.

[0045] The passenger travel parameter acquisition module is used to obtain the travel purpose, destination location, departure location and boarding station location information of the traveling passengers, and pass the passenger travel purpose to the passenger off-board time value coefficient acquisition module, pass the boarding station location and destination location information to the travel route boarding station upstream arrival time interval prediction module, and pass the departure location and boarding station location to the expected boarding schedule matching module. The passenger off-board time value coefficient acquisition module determines the off-board time value coefficient of the passenger's bus rush process at each stage in combination with the passenger travel purpose information, and passes it to the expected off-board time cost calculation module. The travel route boarding station upstream arrival time interval prediction module determines the passenger's bus route in combination with the destination location and boarding station location, predicts the arrival time interval of each bus upstream of the boarding station, and passes it to the expected bus schedule matching module. The expected bus schedule matching module matches the expected bus schedule in combination with the passenger's walking distance, and passes it to the expected off-board time cost calculation module. The expected off-board time cost calculation module, combined with the above information, constructs an expected off-board time cost function with the expected flight arrival time release value as the independent variable, determines the off-board time cost corresponding to each integer release value in the prediction interval by trial and error, and passes the arrival time release value (recommended arrival time value) with the minimum off-board time cost to the arrival time release module. The arrival time release module provides the recommended arrival time value to the passengers.

[0046] The present invention also provides a method for publishing bus arrival time with the minimum expected off-board time cost. The method is suitable for bus passengers who have flexible demands for additional activities. Such passengers have the potential demand to take additional activities according to the published arrival time information to optimize the off-board time utility. Specifically, the method includes the following steps:

[0047] S1. Obtaining the passenger's travel characteristic parameters, including the passenger's travel purpose (leisure travel, commuting travel, business travel), destination location, departure location, and boarding station location;

[0048] S2. Obtain the off-board time value coefficients of the passengers in each stage of the process of catching the bus. The process of catching the bus refers to the stage from the departure of the passengers from the starting point of the trip to the arrival at the station and the departure of the bus. The time value coefficients in this process include: the time value coefficient β of the passengers walking to the station, the time value coefficient α of the passengers' additional activities (additional activities refer to a series of short-term activities that passengers can choose during the off-board time to enhance the utility of the off-board time, such as shopping, withdrawing money, buying breakfast, etc.), the time value coefficient γ of the passengers waiting for the expected bus at the station, and the time value coefficient δ of the passengers waiting for the expected bus at the station when they miss the expected bus. It should be noted that the time value coefficients of the passengers waiting for the expected bus at the station when they miss the expected bus are different for passengers with different travel purposes. Specifically, the time value coefficients of the passengers waiting for the expected bus at the station are the largest, followed by commuting travel, and the smallest for leisure travel.

[0049] S3, obtaining the predicted arrival time interval of each bus service upstream of the boarding station of the travel route, wherein the predicted arrival time interval of the bus service upstream of the boarding station of the travel route is its 99% confidence interval;

[0050] S4. Match the expected bus schedule based on the passenger's walking distance. The matched expected bus schedule is the earliest bus schedule that can ensure that the passenger can take the bus immediately after obtaining the information. That is, the lower bound of the confidence interval of the expected bus schedule arrival time prediction is higher than the walking time of the passenger to the station.

[0051] S5. Establish an expected off-board time cost model; the expected off-board time cost model for public transportation is a definite integral function. The independent variable is the published value of the expected arrival time of the bus, which is an integer minute. The value range is the 99% confidence interval of the expected arrival time of the bus. In particular, if the minimum integer value in the prediction interval is greater than the lower limit of the prediction interval, there is still a risk of passengers missing the expected bus when the minimum integer value in the prediction interval is published. At this time, the minimum integer published value should be an integer value that is less than and close to the lower limit of the prediction interval.

[0052] The expected off-board time cost of a bus is composed of the walking time cost of the passenger to the station, the time cost of the passenger's additional activities, and the waiting time cost of the passenger at the station. The waiting time cost of the passenger at the station depends on whether the passenger can take the expected bus after choosing the additional activities. Therefore, the expected off-board time cost C of the passenger can be expressed in two forms:

[0053] 1) After completing the additional activities, the passenger takes the desired bus and leaves. The passenger's waiting time at the station is the actual arrival time of the desired bus, t r The time t when the passenger completes the additional activities and arrives at the station f +t s The difference between the two is that the passenger expects the time cost outside the vehicle to be C 1 It can be expressed as:

[0054] C 1 =βt f +αt s +γ(t r -t f -t s )

[0055] 2) After completing the additional activities, the passenger misses the expected bus and leaves on the next bus. The passenger's waiting time at the station is the actual arrival time of the next bus, t r +h (h is the vehicle arrival interval) and the time t when the passenger completes the additional activities and arrives at the station f +t s The difference between the two is that the passenger expects the time cost outside the vehicle to be C 2 It can be expressed as:

[0056] C 2 =βtf +αt s +γ(t r +ht f -t s )

[0057] As a preferred embodiment of the above scheme, the passenger's expected off-vehicle time cost in step S5 is the sum of the product of the off-vehicle time cost in the above two cases and the probability of occurrence thereof, and the calculation method is as follows:

[0058]

[0059] In the formula, f(t s ) is the probability density function of the additional activity time of bus passengers, f(t r ) is the probability density function of the expected actual arrival time, t l is the lower limit of the confidence interval of the expected flight arrival time, t u is the upper limit of the confidence interval of the expected arrival time of the train.

[0060] S6. Publish the expected arrival time with the lowest cost.

[0061] The recommended value of the arrival time in step S6 is solved according to the following steps:

[0062] 1) Combine the prediction accuracy of the arrival information service system and the expected flight arrival time prediction value to determine the value range of the arrival time release value t, that is: t∈(t l ,t u );

[0063] 2) Substitute the integer value of the minute in the value range into the bus passenger's expected off-board time cost function, calculate the passenger's expected off-board time cost, and select the arrival time prediction value corresponding to the minimum expected cost as the recommended arrival time value.

[0064] Taking the urban residents of a certain city as an example, the expected bus arrival time release value in a virtual bus travel scenario was determined through the proposed bus arrival time release strategy, and the expected off-board time cost was compared with the traditional arrival time release strategy.

[0065] A method for publishing bus arrival time with the minimum expected off-board time cost, the specific implementation steps are as follows:

[0066] S1. Obtain the travel parameters of the target passenger. In this embodiment, the travel characteristic parameters of the passenger are:

[0067] ① The walking time for passengers to get to the station (the walking distance is about 500 meters) is about 6 minutes;

[0068] ② The purpose of the passenger's trip is shopping, which belongs to leisure travel;

[0069] ③ Such as Figure 2 As shown in the figure, assuming that the predicted error of bus arrival time follows a normal distribution with an expected value of 0, then the actual bus arrival time follows an expected value of t p According to the normal distribution principle of "3sigma" of normal distribution, the probability of the value being distributed in (μ-3σ,μ+3σ) is 0.9974. It can be considered that the actual arrival time is within the following prediction range:

[0070] t r ∈(t p -3σ,t p +3σ)

[0071] Previous studies have shown that the predicted error of arrival time increases with the increase of distance or time. Here, it is further assumed that the standard deviation σ of the arrival time error is proportional to the predicted value of the arrival time, that is:

[0072] σ=λt p

[0073] In this embodiment, it is assumed that the prediction accuracy (MAPE) of the bus arrival information system is 10%. Approximately equal to 0.125, t r ∈(12.5min, 27.5min), the corresponding expected train arrival time release value takes the interval t∈[12min,27.5min)(t is an integer in the interval), and the arrival interval between the expected train and its adjacent train is set to 10min, that is, h=10min.

[0074] S2. Obtain the off-board time value coefficients of the target passengers at each stage during the period from departure from the starting point to arrival at the station, including the value coefficient β of the passenger's walking time to the station, the value coefficient α of the passenger's additional activity time, the value coefficient γ of the passenger's waiting time at the station when waiting for the expected bus, and the value coefficient δ of the passenger's waiting time at the station when he misses the expected bus. In the case of the present invention, α=0.855CNY / min, β=1.425CNY / min, γ=1.425CNY / min, and δ=2.28CNY / min.

[0075] S3, obtaining the predicted arrival time interval of each bus service upstream of the boarding station of the travel route; Figure 2 As shown in the figure, assuming that the predicted error of bus arrival time follows a normal distribution with an expected value of 0, then the actual bus arrival time follows an expected value of t p According to the normal distribution principle of "3sigma" of normal distribution, the probability of the value being distributed in (μ-3σ,μ+3σ) is 0.9974. It can be considered that the actual arrival time is within the following prediction range:

[0076] t r ∈(t p -3σ,t p +3σ)

[0077] Previous studies have shown that the predicted error of arrival time increases with the increase of distance or time. Here, it is further assumed that the standard deviation σ of the arrival time error is proportional to the predicted value of the arrival time, that is:

[0078] σ=λt p

[0079] In this embodiment, it is assumed that the prediction accuracy (MAPE) of the bus arrival information system is 10%. Approximately equal to 0.125, for example, t p =20min, t r ∈(12.5min, 27.5min), the corresponding expected train arrival time release value takes the interval t∈[12min,27.5min)(t is an integer in the interval), and the arrival interval between the expected train and its adjacent train is set to 10min, that is, h=10min.

[0080] S4. Match the expected bus schedule based on the passenger's walking distance. According to the principle that the lower bound of the confidence interval of the expected bus schedule arrival time prediction is higher than the passenger's walking time to the station, match the earliest bus schedule that meets the conditions as the expected bus schedule. In this embodiment, the expected bus schedule arrival time is set as t p = 20min, and the corresponding arrival time release value interval t∈[12min,27.5min); S5, establish the expected off-vehicle time cost model; obtain the expected off-vehicle time cost function of the bus passengers from the departure point to the arrival station. Among them, if Figure 3 As shown in the figure, assuming that the additional activity duration of the passenger is within the selectable range (tt f ) obeys a uniform distribution, that is, passengers randomly select the additional activity duration within their selectable interval.

[0081] S5, such as Figure 4 As shown, the expected off-vehicle time cost corresponding to each integer published value in the arrival time prediction interval is solved. The arrival time prediction value (recommended value) with the minimum expected off-vehicle time cost is given, corresponding to the arrival time recommended value t*=18min in the case of the present invention.

[0082] Practical evaluation: From the perspective of optimization effect, compared with publishing the recommended value (18min) and the forecast lower limit (12min) and the integer value near the forecast expected value (20min) within the forecast interval, the expected off-vehicle time cost saved by passengers in the above travel scenario is 26.72-25.54=1.18CNY (saving 4.6%) and 25.96-25.54=0.42CNY (saving 1.6%), respectively. The optimization effect is significant.

[0083] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A method for publishing bus arrival time with the minimum expected off-board time cost. It is characterized in that The following steps are involved: S1. Obtaining the passenger's travel characteristic parameters, including: the passenger's travel purpose, destination location, departure location, and boarding station location information; S2. Obtain the off-board time value coefficients at each stage of the passenger's process of catching the bus. The process of catching the bus refers to the stage from the passenger's departure from the starting point to the arrival at the station and the departure of the bus. The time value coefficients in this process include: the time value coefficient of the passenger walking to the station β , the time value coefficient of passenger additional activities α , the value coefficient of the waiting time of passengers waiting for the expected bus at the station γ , the value coefficient of the waiting time at the station when the passenger misses the expected bus δ ; S3, obtaining the predicted arrival time interval of each bus service upstream of the boarding station of the travel route; S4, matching the expected bus frequency with the passenger's walking distance; S5. Establishing the expected off-vehicle time cost model; The expected off-board time cost of the bus in step S5 is composed of the walking time cost of the passenger to the station, the time cost of the passenger's additional activities, and the waiting time cost of the passenger at the station. The waiting time cost of the passenger at the station depends on whether the passenger can take the expected bus after selecting the additional activities. The expected off-board time cost C of the passenger has two expressions: 1) When a passenger completes the additional activities and leaves on the desired bus, the passenger's waiting time at the station is the actual arrival time of the desired bus. t r The time to arrive at the destination after completing additional activities with the passenger t f +t s The difference between the passenger's expected off-vehicle time cost C 1 Expressed as: 2) When a passenger completes the additional activities and misses the expected bus, and leaves on the next bus, the passenger's station waiting time is the actual arrival time of the next bus. t r +h , h The time between vehicle arrivals and the time it takes for passengers to complete additional activities and arrive at the station t f +t s The difference between the passenger's expected off-vehicle time cost C 2 Expressed as: ; The expected off-board time cost of the passenger in step S5 is the sum of the product of the off-board time cost and its occurrence probability in two cases: when the passenger completes the additional activities and leaves on the expected bus, and when the passenger completes the additional activities and misses the expected bus and leaves on the next bus. The calculation method is as follows: In the formula, f ( t s ) is the probability density function of the additional activity time of bus passengers, f ( t r ) is the probability density function of the expected actual arrival time, t l is the lower limit of the confidence interval of the expected flight arrival time, t u is the upper limit of the confidence interval of the expected flight arrival time; S6. Publish the arrival time with the minimum expected cost; The recommended value of the arrival time in step S6 is solved according to the following steps: S601: Determine the arrival time release value based on the prediction accuracy of the arrival information service system and the expected flight arrival time prediction value t The value range of is: t ∈( t l , t u ); S602: Substitute the minute integer value within the value range into the public transport passenger's expected off-vehicle time cost function, calculate the passenger's expected off-vehicle time cost, and select the arrival time prediction value corresponding to the minimum expected cost as the arrival time recommendation value.

2. A system for publishing bus arrival time with the minimum expected off-board time cost, the system implements the method for publishing bus arrival time with the minimum expected off-board time cost as claimed in claim 1, It is characterized in that It includes a passenger travel parameter acquisition module, a passenger off-board time value coefficient acquisition module, a travel route and bus stop upstream bus arrival time interval prediction module, an expected bus schedule matching module, an expected off-board time cost calculation module and an arrival time publishing module; The passenger travel parameter acquisition module is used to obtain the travel destination of the traveling passenger, the location of the passenger's travel destination, the location of the passenger's departure point, and the location information of the boarding station; The passenger off-vehicle time value coefficient acquisition module is used to obtain the passenger off-vehicle time value coefficient at each stage of the process of catching the bus, and pass the passenger travel characteristic parameters to the expected off-vehicle time cost calculation module; The travel route upstream bus station arrival time interval prediction module is used to obtain the arrival time interval information of each bus upstream of the travel route station, and pass it to the expected travel schedule matching module; The expected bus schedule matching module combines the passenger's walking distance to match the expected bus schedule and transmits it to the expected off-board time cost calculation module; The expected off-vehicle time cost calculation module determines the off-vehicle time cost corresponding to each integer release value in the prediction interval, and transmits the arrival time release value with the minimum off-vehicle time cost to the arrival time release module; The arrival time publishing module is used to provide the recommended arrival time value to passengers.

3. The bus arrival time publishing system with the minimum expected off-board time cost according to claim 2, It is characterized in that The passenger travel parameter acquisition module passes the passenger travel purpose to the passenger off-vehicle time value coefficient acquisition module, passes the boarding station location and destination location information to the upstream bus arrival time interval prediction module of the boarding station of the travel route, and passes the departure location and boarding station location to the expected bus schedule matching module.

4. The bus arrival time publishing system with the minimum expected off-board time cost according to claim 2, It is characterized in that The passenger's off-board time value coefficient acquisition module determines the off-board time value coefficients of each stage of the passenger's rushing process in combination with the passenger's travel destination information, including: the passenger's time value coefficient for walking to the station β , the time value coefficient of passenger additional activities α , the value coefficient of the waiting time of passengers waiting for the expected bus at the station γ , the value coefficient of the waiting time at the station when the passenger misses the expected bus δ, And pass it to the expected off-vehicle time cost calculation module.

5. The bus arrival time publishing system with the minimum expected off-board time cost according to claim 2, It is characterized in that The travel route and the upstream bus station arrival time interval prediction module determines the passenger's travel route in combination with the passenger's destination location and the bus station location, predicts the arrival time intervals of each bus upstream of the bus station, and transmits them to the expected travel schedule matching module.

6. The bus arrival time publishing system with the minimum expected off-board time cost according to claim 2, It is characterized in that The expected off-board time cost calculation module constructs an expected off-board time cost function with the expected shift arrival time release value as the independent variable, determines the off-board time cost corresponding to each integer release value in the prediction interval by trial and error, and passes the arrival time release value with the smallest off-board time cost to the arrival time release module.

Citation Information

Patent Citations

  • Conventional bus and variable line bus combined scheduling method

    CN113592215A

  • System and method for measuring perceived impact of schedule deviation in public transport

    US20170169373A1

  • Method for designing routine urban public transit network

    WO2017045294A1