Empty bus scheduling method and system for demand response type buses

Abstract

The embodiment of the invention discloses an empty bus scheduling method and system for a demand response type bus. The method comprises the steps that S1, station information and vehicle information of the demand response type bus are determined; s2, constructing an empty vehicle scheduling matching model; s3, constructing a solution algorithm of an empty bus scheduling problem of the demand response type buses, and solving the empty bus scheduling problem of the demand response type buses; and S4, obtaining an optimal scheduling decision of each vehicle, scheduling an empty vehicle according to the decision, and updating new vehicle information, station information and parking station state information. According to the invention, the operation efficiency of the demand response type public transportation system can be obviously improved, so that the system can better cope with the station demand. The method can help to optimize an empty vehicle scheduling strategy, and enables the empty vehicles to be reasonably matched with station demands.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to an empty vehicle dispatching method and system for demand-responsive public transportation. Background Art

[0002] Demand-responsive bus, also known as on-demand bus, is a public transportation service mode that is flexibly adjusted according to passenger demand. This service can effectively improve the efficiency of bus resource utilization and alleviate traffic pressure during peak hours. In this mode, buses dynamically adjust their routes and times according to real-time passenger demand to achieve higher service coverage and passenger satisfaction. In recent years, with the development of data science and artificial intelligence technology, demand-responsive bus services have been promoted in many cities. In the process of demand-responsive bus operation, vehicle scheduling algorithm is the core strategy to guide operation.

[0003] Existing dispatching algorithms mainly focus on the problem of matching people with vehicles, but rarely involve the study of empty vehicle dispatching strategies. In the actual operation process, if empty vehicles can be reasonably dispatched and empty vehicle coverage sites can be reasonably allocated, the goal of responding to demand faster can be achieved and the system's operational efficiency can be improved. Summary of the invention

[0004] The technical problem to be solved by the embodiments of the present invention is to provide an empty vehicle dispatching method and system for demand-responsive public transportation, so as to improve the operating efficiency of the demand-responsive public transportation system.

[0005] In order to solve the above technical problems, an embodiment of the present invention proposes an empty vehicle dispatching method for demand-responsive public transportation, comprising the following steps: S1: Determine the site information and vehicle information of the demand-responsive bus, wherein the site information includes station information, stop site information and demand site information, and the vehicle information includes information of vehicles in different service states; S2: Build an empty car dispatch matching model to match demand stations with parking stations according to the demand volume of demand stations; S3: Construct an algorithm to solve the empty vehicle dispatching problem of demand-responsive public transportation and solve the empty vehicle dispatching problem of demand-responsive public transportation; S4: Obtain the best dispatch decision for each vehicle, dispatch empty vehicles according to the decision, and update new vehicle information, station information, and parking station status information.

[0006] Accordingly, an embodiment of the present invention further provides an empty vehicle dispatching system for demand-responsive public transportation, comprising: Information determination module: determining the station information and vehicle information of the demand-responsive bus, wherein the station information includes station information, stop station information and demand station information, and the vehicle information includes information of vehicles in different service states; Model building module: build an empty car dispatch matching model to match demand stations with parking stations according to the demand volume of demand stations; Solution calculation module: construct a solution algorithm for the empty vehicle scheduling problem of demand-responsive public transportation, and solve the empty vehicle scheduling problem of demand-responsive public transportation; Vehicle dispatch module: obtains the best dispatch decision for each vehicle, dispatches empty vehicles according to the decision, and updates new vehicle information, station information, and station status information.

[0007] The beneficial effects of the present invention are as follows: the present invention dispatches vehicles according to the expected demand at the station, so that the number of empty vehicles dispatched to the station matches the station demand, thereby optimizing the operational efficiency of demand-responsive public transportation. The present invention reasonably optimizes the objective function so that the objective function can reflect the marginal diminishing effect of the number of vehicles matching demand and handle the situation where there is no vehicle coverage. The linear nature of the algorithm makes the solution more reasonable. The farthest dispatch distance and maximum response time parameters introduced in the present invention can be adjusted according to the actual situation of the system to achieve response to the dynamic demand environment and flexibly adapt to environmental changes in display operations.

[0008] The present invention can solve the aspects that were ignored by previous algorithms and further improve the system operation efficiency. The present invention considers the priority of site demand and dispatches empty vehicles to cover the site. The objective function of the present invention takes into account the diminishing marginal utility effect. When dealing with vehicle coverage, the contribution of the first vehicle is 1, and the contribution of subsequent vehicles is 0.5, and the situation of no vehicle coverage is considered. At the same time, the linear nature of the model makes the model calculation complexity reasonable.

[0009] The present invention helps to better match the parking sites with high-demand sites, optimize the vehicle dispatching strategy, and improve the overall operating efficiency of the public transportation system. The present invention analyzes the site demand, uses advanced algorithms to reasonably dispatch empty vehicles, dynamically adjusts the vehicle configuration at the parking sites according to demand, and maximizes the coverage of the service scope of the public transportation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a flow chart of an empty vehicle dispatching method for demand-responsive public transportation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0011] It should be noted that, in the absence of conflict, the embodiments in the present application and the features in the embodiments may be combined with each other. The present invention is further described in detail below in conjunction with the drawings and specific embodiments.

[0012] In the embodiments of the present invention, if there are directional indications (such as up, down, left, right, front, back, etc.), they are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0013] In addition, in the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of the features.

[0014] The method for dispatching empty buses of a demand-responsive bus according to the embodiment of the present invention comprises the following steps S1 to S4. Figure 1 .

[0015] S1: Determine the station information and vehicle information of the demand-responsive bus (a bus mode that is different from the traditional bus operating on a fixed line network. In this mode, the bus responds to the needs of passengers and runs between determined stations. The route changes dynamically with the changes in passenger demand. In this mode, relevant algorithms are required to optimize the vehicle's route, and the algorithm plays an important role in operational decisions). The station information includes station information, stop station information and demand station information, and the vehicle information includes information on vehicles in different service states. The present invention inputs the demand quantity of each station, derives the demand priority of each station, and then reasonably allocates empty vehicles to each stop station. Empty vehicle scheduling: In a demand-responsive bus operation system, the process of scheduling empty vehicles to stop stations. The present invention meets the maximum passenger demand by optimizing the empty vehicle scheduling behavior.

[0016] Specific process: Form a station collection : Determine the number of vehicle stops (i.e., vehicle stops, service stops for demand-responsive public transportation, the number of passengers carried by the stops, the stops are connected to form a service line network for demand-responsive public transportation, and each stop corresponds to its nearest vehicle stop) and determine their location information.

[0017] Form a collection of parking sites (i.e. sites where empty vehicles are parked. When vehicles are not in service, they are parked here. The sites have capacity constraints) : Determine the number of parking sites and collect information about each station, including location, number of original vehicles, and maximum capacity.

[0018] Form a collection of demand sites :Determine the number of demand stations, collect the demand for each station, and reflect the number of passengers that the demand station needs to satisfy.

[0019] Forming a vehicle collection , the vehicles are grouped into corresponding sets according to their different states.

[0020] in, Represents the collection of vehicles in operation. Represents the set of vehicles in the process of dispatching empty vehicles to parking sites. To gather vehicles waiting for service at the stop. A collection of vehicles that have completed their service and are waiting to be dispatched.

[0021] For vehicles in different service states, determine the number of vehicles in each state and collect basic information of each vehicle, such as vehicle ID, current location, etc.

[0022] S2: Construct an empty vehicle dispatch matching model to match the demand station with the parking station according to the demand of the demand station. The present invention dispatches empty vehicles from the parking station to the corresponding demand station for service by identifying the station demand. With the goal of maximizing the vehicle coverage station demand, an objective function considering the diminishing marginal effect and the lack of vehicle coverage is established, and its feasibility constraints are analyzed, and finally an empty vehicle dispatch optimization model is established.

[0023] (1) The set definition is shown in Table 1:

[0024] (2) The definitions of time-space related parameters are shown in Table 2:

[0025] (3) System parameter definitions are shown in Table 3:

[0026] (4) The decision variables are shown in Table 4:

[0027] (5) The intermediate calculation variables are shown in Table 5:

[0028] (6) Objective function: Maximize the weighted service effect of the entire system, taking into account demand distribution and diminishing marginal utility: (1) The objective function is for all time periods and all stations Perform a double summation.

[0029] in Indicates the station In time period The expected demand is used as the weight coefficient.

[0030] The term reflects the marginal contribution of the number of vehicles, The term reflects the basic contribution of whether there is car coverage, and the two are added together to calculate the coverage effect.

[0031] (7) Constraints: (2) Formula (2) controls and Whether the dispatchable vehicles in the set are allowed to be dispatched, where is an external input 0-1 parameter. When the vehicle Can be dispatched, otherwise cannot be dispatched; this constraint allows the system to interface with external systems to control the dispatch authority of the vehicle.

[0032] (3) Formula (3) indicates that each dispatchable vehicle can be assigned to at most one parking station; (4) Formula (4) represents the , in each period The number of vehicles in a parking station does not exceed its maximum capacity; (5) Formula (5) indicates that the vehicle driving distance generated by the vehicle scheduling decision cannot exceed the maximum parking distance threshold set by the system to ensure scheduling efficiency; (6) Formula (6) is the coverage determination constraint of the parking site; (7) Formula (7) calculates the total number of available vehicles. The first term reflects the vehicles on the way that are about to arrive, the second term reflects the vehicles currently at the parking station, and the third term reflects the vehicles brought by the new scheduling decision; (8) Formula (8) controls the coverage representation variable , the contribution value is 0 when achieving car-free coverage; (9) Formula (9) controls the variable domain, Indicates scheduling decision, coverage relationship and coverage status, which is a 0-1 variable. Represents the number of vehicles, a non-negative integer.

[0033] S3: Construct an algorithm to solve the empty vehicle dispatching problem of demand-responsive public transportation. The specific process is as follows:

[0034] 1. Input: .

[0035] 2. Output: Return a list of vehicle dispatch information and a list of the status of each stop.

[0036] 3. Main loop: S31 initializes the network structure; calculates and assigns the nearest stop for each station; traverses each empty car that needs to be dispatched ; Build a coverage matrix from site to stop .

[0037] 4. S32 defines decision variables: constructs a 0-1 variable to determine whether vehicle v goes to stop i ; Construct a 0-1 variable to determine whether a vehicle covers the parking site i at time t ; Construct a 0-1 variable to determine whether stop i can cover station j within the maximum response time ; Construct the total number of available vehicles at stop i in time period t ; Construct the total number of covered vehicles variable for station j in time period t .

[0038] 5. S33 builds optimization objectives: establishes an objective function to maximize site demand coverage.

[0039] 6. S34 adds constraints: Constraint 1: Scheduling feasibility constraint; Constraint 2: Vehicle single assignment constraint; Constraint 3: Stop capacity constraint; Constraint 4: Maximum parking distance constraint.

[0040] 7. S35 solves the optimization problem: calls the solver to solve the mixed integer programming problem; Case 1: Find the optimal solution; extract the values ​​of the decision variables and construct a plan for allocating vehicles to parking stations; Case 2: There is no optimal solution, and an empty allocation plan is returned.

[0041] 8. End: Return a list of vehicle dispatch information and a list of the status of each stop.

[0042] The present invention establishes a network structure, defines decision variables, constructs optimization objectives, adds constraints, reasonably sets model parameters, and uses a mixed integer programming solver to solve the model to obtain the optimal scheduling plan and stop information, thereby optimizing the operation decision of demand-responsive public transportation.

[0043] S4: Obtain the optimal scheduling decision for each vehicle, schedule empty vehicles according to the decision, and update new vehicle information, station information, and parking station status information. The present invention calls a solver to solve the problem, obtains the optimal scheduling decision for each vehicle, and updates new vehicles, stations, and parking station status.

[0044] 1. Output scheduling results: Output the dispatch information of each vehicle and display the vehicle Which stop to go to .

[0045] 2. Output the status of the parking station: Count and output the current number of vehicles at each stop, whether there are any vehicles parked there, and the required stations it covers.

[0046] 3. Evaluate optimization results: Calculate and output the final optimization target value to check the optimization effect and system performance.

[0047] The demand-responsive public transport empty vehicle dispatching system of the embodiment of the present invention includes: Information determination module: determining the station information and vehicle information of the demand-responsive bus, wherein the station information includes station information, stop station information and demand station information, and the vehicle information includes information of vehicles in different service states; Model building module: build an empty car dispatch matching model to match demand stations with parking stations according to the demand volume of demand stations; Solution calculation module: construct a solution algorithm for the empty vehicle scheduling problem of demand-responsive public transportation, and solve the empty vehicle scheduling problem of demand-responsive public transportation; Vehicle dispatch module: obtains the best dispatch decision for each vehicle, dispatches empty vehicles according to the decision, and updates new vehicle information, station information, and station status information.

[0048] As an implementation, the model uses the following objective function for all time periods and all stations Perform a double sum: ; in, is the set of stations (i.e., vehicle stops), T is the set of time intervals, Indicates the station In time period The expected demand, For the station In time period Total number of covered vehicles; It is a 0-1 variable, indicating the station In time period Is there vehicle coverage? If so, ,otherwise .

[0049] As an implementation method, the constraints of the model include: (1) Scheduling feasibility constraints: ; in, is an external input 0-1 parameter. When the vehicle can be scheduled, otherwise cannot be scheduled; I is the set of parking sites; A 0-1 variable representing a vehicle Whether to dispatch to the parking site , if the vehicle is heading to the station ,but ,otherwise in ; Gather vehicles waiting for service at the stop; A collection of vehicles that have completed their service and are waiting to be dispatched; (2) Vehicle single allocation constraint: ; (3) Capacity constraints of parking sites: ; in, For parking sites In time period Total number of available vehicles; For parking sites Total berth capacity; (4) Maximum parking distance constraint: ; in, For vehicles To the parking site The driving time, is the preset maximum parking distance threshold; (5) Constraints for determining the coverage of parking sites: ; in, Indicates a stop Whether the station can be covered within the maximum response time , is the maximum response time threshold, From the parking site To the station The driving time; (6) Calculate the total number of available vehicles: ; in, The collection of vehicles in the process of dispatching empty vehicles to parking sites, Indicates empty car on the way Arrival at the stop time period, Indicates vehicle Is it at the parking site? , Indicates vehicle Arrival at the stop the time period; (7) Control coverage representation variables , the contribution value is 0 when no car coverage is achieved: ; Where M is a sufficiently large number (in the big M method, M is a constant much larger than other variables to ensure that the constraint is still valid when a certain condition is not met); (8) Control variable domain: ; in, Indicates scheduling decision, coverage relationship, and coverage status, which is a 0-1 variable; Represents the number of vehicles, a non-negative integer.

[0050] As an implementation method, the solution calculation module solves the empty vehicle scheduling problem of the demand-responsive public transportation according to the following steps: Initialize the network structure: calculate and assign the nearest stop to each station, and traverse each empty car that needs to be dispatched , build a coverage matrix from site to stop site ; Define decision variables: Construct a 0-1 variable to determine whether vehicle v goes to stop i , construct a 0-1 variable to indicate whether a vehicle covers the stop i at time t , construct a 0-1 variable to determine whether stop i can cover station j within the maximum response time , construct the variable of the total number of available vehicles at the parking station i in time period t , construct the total number of covered vehicles variable for station j in time period t ; Construct optimization objectives: Establish objective functions to maximize site demand coverage; Add constraints: scheduling feasibility constraints, vehicle single allocation constraints, parking site capacity constraints, and maximum parking distance constraints; Solve the empty vehicle dispatch problem of demand-responsive public transportation. If the optimal solution is found, extract the values ​​of the decision variables and construct a vehicle-to-station allocation plan; if there is no optimal solution, return an empty allocation plan. After the end, return a vehicle dispatch information list and a status list of each station.

[0051] The present invention can significantly improve the operational efficiency of the demand-responsive public transportation system, allowing the system to better respond to site demands. The present invention can help optimize the empty vehicle dispatching strategy, so that empty vehicles are reasonably matched with site demands.

[0052] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for dispatching empty vehicles of demand-responsive public transportation, characterized in that: The following steps are involved: S1: Determine the site information and vehicle information of the demand-responsive bus, wherein the site information includes station information, stop site information and demand site information, and the vehicle information includes information of vehicles in different service states; S2: Build an empty car dispatch matching model to match demand stations with parking stations according to the demand volume of demand stations; S3: Construct an algorithm to solve the empty vehicle dispatching problem of demand-responsive public transportation and solve the empty vehicle dispatching problem of demand-responsive public transportation; S4: Obtain the best dispatch decision for each vehicle, dispatch empty vehicles according to the decision, and update new vehicle information, station information, and parking station status information.

2. The method for dispatching empty vehicles of demand-responsive public transportation as claimed in claim 1, characterized in that: The model uses the following objective function for all time periods and all stations Perform a double summation: ； in, is the set of stations, T is the set of time intervals, Indicates the station In time period The expected demand, For the station In time period Total number of covered vehicles; It is a 0-1 variable, indicating the station In time period Is there vehicle coverage? If so, ,otherwise .

3. The method for dispatching empty vehicles of demand-responsive public transportation as claimed in claim 2, characterized in that: The constraints of the model include: (1) Scheduling feasibility constraints: ； in, is an external input 0-1 parameter. When the vehicle can be scheduled, otherwise cannot be scheduled; I is the set of parking sites; A 0-1 variable representing a vehicle Whether to dispatch to the parking site , if the vehicle is heading to the station ,but ,otherwise ,in ; Gather vehicles waiting for service at the stop; A collection of vehicles that have completed their service and are waiting to be dispatched; (2) Vehicle single allocation constraint: ； Where, I is the set of stop sites; A 0-1 variable representing a vehicle Whether to dispatch to the parking site , ; Gather vehicles waiting for service at the stop; A collection of vehicles that have completed their service and are waiting to be dispatched; (3) Capacity constraints of parking sites: ； in, For parking sites In time period Total number of available vehicles; For parking sites Total berth capacity; (4) Maximum parking distance constraint: ； in, For vehicles To the parking site The driving time, is the preset maximum parking distance threshold; (5) Constraints for determining the coverage of parking sites: ； in, Indicates a stop Whether the station can be covered within the maximum response time , is the maximum response time threshold, From the parking site To the station The driving time; (6) Calculate the total number of available vehicles: ； in, The collection of vehicles in the process of dispatching empty vehicles to parking sites. Indicates empty car on the way Arrival at the stop time period, Indicates vehicle Is it at the parking site? , Indicates vehicle Arrival at the stop the time period; (7) Control coverage representation variables , the contribution value is 0 when no car coverage is achieved: ； Where M is a sufficiently large number; (8) Control variable domain: ； in, , , Indicates scheduling decision, coverage relationship, and coverage status, which is a 0-1 variable; , Represents the number of vehicles, a non-negative integer.

4. The method for dispatching empty vehicles of demand-responsive public transportation as claimed in claim 3, characterized in that: Step S3 includes the following sub-steps: S31, initialize the network structure: calculate and assign the nearest stop to each station, and traverse each empty car that needs to be dispatched , build a coverage matrix from site to stop site ; S32, define decision variables: construct a 0-1 variable to determine whether vehicle v goes to stop i , construct a 0-1 variable to indicate whether a vehicle covers the stop i at time t , construct a 0-1 variable to determine whether stop i can cover station j within the maximum response time , construct the variable of the total number of available vehicles at the parking station i in time period t , construct the total number of covered vehicles variable for station j in time period t ; S33, construct optimization objectives: establish an objective function to maximize site demand coverage; S34, adding constraints: scheduling feasibility constraint, vehicle single allocation constraint, parking site capacity constraint, and maximum parking distance constraint; S35, solve the empty vehicle scheduling problem of demand-responsive public transportation. If an optimal solution is found, the values ​​of the decision variables are extracted to construct an allocation plan for vehicles to the parking sites; if there is no optimal solution, an empty allocation plan is returned.

5. A demand-responsive public transport empty vehicle dispatching system, characterized in that: include: Information determination module: determining the station information and vehicle information of the demand-responsive bus, wherein the station information includes station information, stop station information and demand station information, and the vehicle information includes information of vehicles in different service states; Model building module: build an empty car dispatch matching model to match demand stations with parking stations according to the demand volume of demand stations; Solution calculation module: construct a solution algorithm for the empty vehicle scheduling problem of demand-responsive public transportation, and solve the empty vehicle scheduling problem of demand-responsive public transportation; Vehicle dispatch module: obtains the best dispatch decision for each vehicle, dispatches empty vehicles according to the decision, and updates new vehicle information, station information, and station status information.

6. The demand-responsive public transport empty vehicle dispatching system as claimed in claim 5, characterized in that: The model uses the following objective function for all time periods and all stations Perform a double summation: ； in, is the set of stations, T is the set of time intervals, Indicates the station In time period The expected demand, For the station In time period Total number of covered vehicles; It is a 0-1 variable, indicating the station In time period Is there vehicle coverage? If so, ,otherwise .

7. The demand-responsive bus empty vehicle dispatching system as claimed in claim 6, characterized in that: The constraints of the model include: (1) Scheduling feasibility constraints: ； in, is an external input 0-1 parameter. When the vehicle can be scheduled, otherwise cannot be scheduled; I is the set of parking sites; A 0-1 variable representing a vehicle Whether to dispatch to the parking site , if the vehicle is heading to the station ,but ,otherwise in ; Gather vehicles waiting for service at the stop; A collection of vehicles that have completed their service and are waiting to be dispatched; (2) Vehicle single allocation constraint: ； Where, I is the set of stop sites; A 0-1 variable representing a vehicle Whether to dispatch to the parking site , ; Gather vehicles waiting for service at the stop; A collection of vehicles that have completed their service and are waiting to be dispatched; (3) Capacity constraints of parking sites: ； in, For parking sites In time period Total number of available vehicles; For parking sites Total berth capacity; (4) Maximum parking distance constraint: ； in, For vehicles To the parking site The driving time, is the preset maximum parking distance threshold; (5) Constraints for determining the coverage of parking sites: ； in, Indicates a stop Whether the station can be covered within the maximum response time , is the maximum response time threshold, From the parking site To the station The driving time; (6) Calculate the total number of available vehicles: ； in, The collection of vehicles in the process of dispatching empty vehicles to parking sites. Indicates empty car on the way Arrival at the stop time period, Indicates vehicle Is it at the parking site? , Indicates vehicle Arrival at the stop the time period; (7) Control coverage representation variables , the contribution value is 0 when no car coverage is achieved: ； Where M is a sufficiently large number; (8) Control variable domain: ； in, Indicates scheduling decision, coverage relationship, and coverage status, which is a 0-1 variable; Represents the number of vehicles, a non-negative integer.

8. The demand-responsive public transport empty vehicle dispatching system as claimed in claim 7, characterized in that: The solution calculation module solves the empty vehicle dispatching problem of demand-responsive public transportation according to the following steps: Initialize the network structure: calculate and assign the nearest stop to each station, and traverse each empty car that needs to be dispatched , build a coverage matrix from site to stop site ; Define decision variables: Construct a 0-1 variable to determine whether vehicle v goes to stop i , construct a 0-1 variable to indicate whether a vehicle covers the stop i at time t , construct a 0-1 variable to determine whether stop i can cover station j within the maximum response time , construct the variable of the total number of available vehicles at the parking station i in time period t , construct the total number of covered vehicles variable for station j in time period t ; Construct optimization objectives: Establish objective functions to maximize site demand coverage; Add constraints: scheduling feasibility constraints, vehicle single allocation constraints, parking site capacity constraints, and maximum parking distance constraints; Solve the empty vehicle dispatching problem of demand-responsive public transportation. If the optimal solution is found, extract the values ​​of the decision variables and construct an allocation plan for vehicles to the stop sites; if there is no optimal solution, return an empty allocation plan.

Images