A rolling horizon based online-to-offline ride-hailing routing and rebalancing scheduling method and system

Abstract

The present application belongs to the field of intelligent transportation, and in particular relates to a method and system for online car-hailing routing and rebalancing scheduling based on a rolling time domain. The method comprises: obtaining an online car-hailing operating area and operating period and dividing it into multiple service areas and multiple time domain sequences; dividing the time domain sequences to form a vehicle routing time domain window, a demand prediction time domain window, and a rebalancing scheduling time domain window, with some overlap between adjacent two time domain windows; obtaining vehicle states and passenger trip requests; in the vehicle routing time domain window, based on real-time vehicle states and trip requests, forming a vehicle routing model for the target to optimize vehicle paths; in the rebalancing scheduling time domain window, based on idle state vehicles, forming a vehicle rebalancing model for the target to obtain a cross-area vehicle scheduling plan and an updated regional vehicle distribution; and performing data collection for the next time sequence to form a window rolling closed loop.

Description

Technical Field

[0001] This invention belongs to the field of intelligent transportation, and specifically relates to a method and system for ride-hailing routing and rebalancing scheduling based on rolling time domain. Background Technology

[0002] As a crucial component of modern transportation services, ride-hailing services directly impact user experience and a company's market competitiveness through their operational efficiency and service quality. However, existing ride-hailing dispatching methods suffer from a significant disconnect between vehicle route planning and rebalancing scheduling, resulting in overall low dispatching efficiency. Specifically, vehicle route planning typically only optimizes for immediate demand, failing to establish a collaborative mechanism with rebalancing scheduling; conversely, rebalancing scheduling lacks support for real-time dynamic route adjustments during execution. This fragmented operational model not only increases inefficient vehicle travel but also hinders the maintenance of supply-demand balance across regions, thus restricting the improvement of operational efficiency and service quality. This invention proposes a rolling time-domain-based ride-hailing route and rebalancing scheduling method that combines short-term demand forecasting to achieve a balance between global optimization and real-time response, thereby improving operational efficiency and service quality. Summary of the Invention

[0003] The present invention is proposed based on the above-mentioned needs of the prior art. The technical problem to be solved by the present invention is to provide a method and system for ride-hailing routing and rebalancing scheduling based on rolling time domain to improve operational efficiency.

[0004] To address the above problems, the technical solution provided by this invention includes: A method for ride-hailing routing and rebalancing scheduling based on rolling time domain is provided, including: obtaining the ride-hailing operating area and operating time period; dividing the ride-hailing operating area into multiple non-overlapping service areas and dividing the ride-hailing operating time period into multiple non-overlapping time domain sequences; forming a vehicle routing time domain window, a demand forecasting time domain window, and a rebalancing scheduling time domain window based on the time domain sequences, with partial overlap between adjacent time domain windows; obtaining the vehicle status and passenger travel requests of each ride-hailing vehicle in each service area within the operating time period, wherein the vehicle status includes idle or carrying passengers, and the travel requests include pick-up and drop-off points; within the vehicle routing time domain window, based on the real-time vehicle status and travel requests, ... To develop a vehicle routing model for the target, and optimize vehicle routes, whereby... The parameter represents the vehicle's origin from node [node name missing]. To the node The cost of driving; As decision variables, Indicates service area Inner vehicle From node Drive to the node ,otherwise , For travel requests within service area r, For service area The set of all vehicles in the system; simultaneously satisfying , , , , , , , , ,in, Indicates vehicle Maximum passenger capacity Indicates each service area Inner vehicle Reaching the node Time, Indicates vehicle In the service area Inner slave node To the node Travel time, This represents the maximum allowed travel time; the prediction time window is divided into multiple time intervals, and future travel demand is predicted based on historical vehicle status and travel requests using a short-term prediction model; within the rebalancing scheduling time window, vehicles in idle status are used to... A vehicle rebalancing model is developed to obtain a cross-regional vehicle scheduling plan and an updated regional vehicle distribution, wherein: , For service area Service level difference variables, As decision variables, Indicates vehicle From service area Dispatch to service area ,otherwise ; Indicates vehicle From service area Dispatch to service area Time; To balance service level differences With vehicle dispatch time The weight parameters for importance must simultaneously satisfy: , , , , , , , The next time-series data collection is carried out to form a rolling closed loop of vehicle routing time-domain window, demand forecasting time-domain window and rebalancing scheduling time-domain window.

[0005] Preferably, in the k-th time-domain sequence, the start time of the vehicle routing time-domain window Represented as, ,in Let be the start time of the rebalancing scheduling time-domain window in the (k-1)th time-domain sequence. To rebalance the length of the scheduling time-domain window, Representing time-domain sequences Rebalancing scheduling time-domain window and time-domain sequence The overlap length of the vehicle routing time-domain window.

[0006] Preferably, in the k-th time-domain sequence, the start time of the demand forecasting time-domain window is... Represented as, , The length of the vehicle routing time-domain window. Representing time-domain sequences Vehicle routing time-domain window and time-domain sequence The overlap length of the demand forecast time domain window.

[0007] Preferably, in the k-th time-domain sequence, the start time of the rebalancing scheduling time-domain window is... Represented as, , The length of the time-domain window for demand forecasting. Representing time-domain sequences Demand forecasting time-domain window and time-domain series The overlap length of the rebalancing scheduling time-domain window.

[0008] Preferably, the time-domain sequence is represented as .

[0009] A rolling time-domain-based ride-hailing routing and rebalancing scheduling system is also provided, comprising: an acquisition module for acquiring ride-hailing operating areas and operating time periods; an operating area division module for dividing the ride-hailing operating areas into multiple non-overlapping service areas; an operating time period division module for dividing the ride-hailing operating time periods into multiple non-overlapping time-domain sequences; a time-domain sequence division module for dividing the time-domain sequences to form vehicle routing time-domain windows, demand forecasting time-domain windows, and rebalancing scheduling time-domain windows, with partial overlap between adjacent time-domain windows; a data acquisition module for acquiring the vehicle status of each ride-hailing vehicle and passenger travel requests in each service area within the operating time period, wherein the vehicle status includes idle or carrying passengers, and the travel requests include pick-up and drop-off points; and a vehicle routing window processing module for, within the vehicle routing time-domain window, based on real-time vehicle status and travel requests, ... To develop a vehicle routing model for the target, and optimize vehicle routes, whereby... The parameter represents the vehicle's origin from node [node name missing]. To the node The cost of driving; As decision variables, Indicates service area Inner vehicle From node Drive to the node ,otherwise , For travel requests within service area r, For service area The set of all vehicles in the system; simultaneously satisfying , , , , , , , , ,in, Indicates vehicle Maximum passenger capacity Indicates each service area Inner vehicle Reaching the node Time, Indicates vehicle In the service area Inner slave node To the node Travel time, This indicates the maximum allowed travel time; the prediction time-domain window processing module divides the prediction time-domain window into multiple time intervals, and predicts future travel demand based on historical vehicle status and travel requests using a short-term prediction model; the rebalancing scheduling time-domain window processing module, based on idle vehicles, uses... A vehicle rebalancing model is developed to obtain a cross-regional vehicle scheduling plan and an updated regional vehicle distribution, wherein: , For service area Service level difference variables, As decision variables, Indicates vehicle From service area Dispatch to service area ,otherwise ; Indicates vehicle From service area Dispatch to service area Time; To balance service level differences With vehicle dispatch time The weight parameters for importance must simultaneously satisfy: , , , , , , , .

[0010] Compared to existing technologies, this invention divides ride-hailing operations into regional and time-period divisions, further subdividing the resulting time-domain sequence into multiple time-domain windows with partial overlap between adjacent windows. Different models are established within each window to align with the optimal solution for that specific window. This minimizes total travel costs within the vehicle routing time-domain window and minimizes vehicle scheduling costs within the rebalancing scheduling window. This enables global ride-hailing optimization based on demand. Vehicle routing and rebalancing scheduling form a collaborative mechanism for dynamic adjustments, mutually supporting and influencing each other. Combined with short-term demand forecasting, this achieves a balance between global optimization and real-time response, thereby improving operational efficiency. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings.

[0012] Figure 1 This is a flowchart illustrating the steps of a ride-hailing routing and rebalancing scheduling method based on rolling time domain according to the present invention. Figure 2 This is a schematic diagram of the time-domain sequence partitioning window according to an embodiment of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] In the description of the embodiments of the present invention, it should be noted that, unless otherwise explicitly specified and limited, the term "connected" should be interpreted broadly. For example, it can refer to a fixed connection, a detachable connection, or an integral connection; it can refer to a mechanical connection or an electrical connection; it can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in the present invention according to the specific circumstances.

[0015] Throughout the text, the terms “top,” “bottom,” “above,” “below,” and “on top” refer to the relative positions of components of the device, such as the relative positions of the top and bottom substrates within the device. It is understood that the device is multifunctional and independent of its spatial orientation.

[0016] To facilitate understanding of the embodiments of the present invention, further explanations and descriptions will be provided below with reference to the accompanying drawings and specific embodiments. These embodiments do not constitute a limitation on the embodiments of the present invention.

[0017] Example 1 This embodiment provides a method for ride-hailing routing and rebalancing scheduling based on rolling time domain.

[0018] like Figure 1 As shown, the ride-hailing routing and rebalancing scheduling method based on rolling time domain includes: Obtain the operating area and operating hours of ride-hailing services.

[0019] The ride-hailing operation area is divided into multiple non-overlapping service areas, and the ride-hailing operation time period is divided into multiple non-overlapping time domain sequences.

[0020] All service areas constitute a set , This indicates the serial number of a specific service area.

[0021] All time-domain sequences constitute a set , It represents the sequence number of a certain time-domain sequence.

[0022] The time-domain sequence is divided into three types of time-domain windows, and adjacent time-domain windows partially overlap.

[0023] In chronological order, the three time-domain windows are the vehicle routing time-domain window, the demand forecasting time-domain window, and the rebalancing scheduling time-domain window. The vehicle routing time-domain window is used for real-time route planning, the demand forecasting time-domain window is used to predict future demand, and the rebalancing scheduling time-domain window is used for cross-regional vehicle scheduling.

[0024] Specifically, such as Figure 2 As shown, the start time of the vehicle routing time domain window is Window length is The start time of the demand forecasting time domain window is... Window length is The start time of the rebalancing scheduling time-domain window is... Window length is .

[0025] The start times of the above three time-domain windows satisfy the following:

[0026]

[0027]

[0028] in, Representing time-domain sequences Rebalancing scheduling time-domain window and time-domain sequence The overlap length of the vehicle routing time-domain window; Representing time-domain sequences Vehicle routing time-domain window and time-domain sequence The overlap length of the demand forecast time domain window; Representing time-domain sequences Demand forecasting time-domain window and time-domain series The overlap length of the rebalancing scheduling time-domain window.

[0029] The window length of each time domain window satisfies: .

[0030] Execute the time-domain sequence in chronological order. For time-domain sequences Follow the specific steps below.

[0031] The system obtains the vehicle status of each ride-hailing vehicle and the travel requests of passengers in each service area during the operating hours. The vehicle status includes whether the vehicle is idle or carrying passengers, and the travel requests include the pick-up point and the drop-off point.

[0032] in, For service area The collection of all vehicles in the system; For service area The collection of all passengers.

[0033] The vehicle status is recorded as follows: .in: This indicates that the vehicle is idle. This indicates that the vehicle is carrying passengers.

[0034] The travel request includes the passenger pick-up point. Drop-off point , For service area All passengers should gather at their designated boarding and alighting points.

[0035] Within the vehicle routing time-domain window, based on real-time vehicle status and travel requests, A vehicle routing model is developed for the target, and vehicle paths are optimized.

[0036] in, The parameter represents the vehicle's origin from node [node name missing]. To the node The cost of driving; As decision variables, Indicates service area Inner vehicle From node Drive to the node ,otherwise .

[0037] The above objectives are set to minimize total driving costs.

[0038] In addition, the vehicle routing model must also satisfy:

[0039]

[0040] The above settings ensure that each passenger's pick-up and drop-off points must be accessed.

[0041]

[0042]

[0043] in, Indicates vehicle The maximum passenger capacity. The above settings are to ensure the continuity of vehicle routes and that the number of passengers carried by each vehicle does not exceed its capacity limit.

[0044]

[0045] The above settings ensure that only available vehicles can be assigned to pick up passengers.

[0046]

[0047] in, Indicates each service area Inner vehicle Reaching the node Time, Indicates vehicle In the service area Inner slave node To the node Travel time, This indicates the maximum allowed driving time. The above settings ensure that the total driving time does not exceed the maximum allowed value.

[0048] The prediction time window is divided into multiple time intervals, and future travel demand is predicted based on the historical vehicle status and travel requests using a short-term prediction model.

[0049] by For an interval, the prediction time-domain window Divided into A time interval of equal length In each service area Within this framework, passenger travel demand is statistically analyzed in chronological order for each time interval. Based on the short-term traffic demand forecasting model, future passenger travel demand is generated, namely:

[0050] in, It can be represented as a set of various functions, such as historical average models, multi-source linear regression models, random forest models, and graph convolutional neural networks.

[0051] Within the rebalancing scheduling time window, vehicles in idle state are... To achieve the goal, a vehicle rebalancing model is developed to obtain cross-regional vehicle scheduling plans and updated regional vehicle distribution.

[0052] in: , For service area The service level difference variable reflects the degree of difference between the actual vehicle dispatch situation and the ideal uniform dispatch state. The smaller the area The closer the service level is to the ideal situation; As decision variables, Indicates vehicle From service area Dispatch to service area ,otherwise ; Indicates vehicle From service area Dispatch to service area Time; To balance service level differences With vehicle dispatch time Weighting parameters for importance.

[0053] The above settings are used to perform cross-service area dispatching of idle ride-hailing vehicles.

[0054] In addition, the vehicle rebalancing model also needs to meet the following requirements:

[0055] The above restrictions ensure that each vehicle can be dispatched at most once.

[0056]

[0057] The above limitations are used to ensure that dispatching is directed to the service area. The number of vehicles does not exceed the service area Forecast demand .

[0058]

[0059]

[0060] The above limitations are designed to ensure that, under ideal conditions, vehicles are evenly distributed to each area.

[0061]

[0062] The above limitations aim to minimize the differences in service levels between service areas when the demand and the total number of available vehicles in each service area are not perfectly matched. Perform linearization.

[0063] Data collection is carried out for the next time series, forming a rolling closed loop of vehicle routing time domain window, demand forecasting time domain window, and rebalancing scheduling time domain window.

[0064] Example 2 This embodiment provides a ride-hailing routing and rebalancing scheduling system based on rolling time domain.

[0065] The ride-hailing routing and rebalancing scheduling system based on rolling time domain includes: The acquisition module is used to obtain the operating area and operating time of ride-hailing services.

[0066] The operation area division module is used to divide the ride-hailing operation area into multiple non-overlapping service areas.

[0067] The operation period segmentation module is used to divide the ride-hailing operation period into multiple non-overlapping time domain sequences.

[0068] The time-domain sequence partitioning module is used to partition the time-domain sequence to form a vehicle routing time-domain window, a demand forecasting time-domain window, and a rebalancing scheduling time-domain window. There is some overlap between adjacent time-domain windows.

[0069] The data acquisition module obtains the vehicle status of each ride-hailing vehicle and the travel requests of passengers in each service area during the operating period. The vehicle status includes whether it is idle or carrying passengers, and the travel requests include the pick-up point and the drop-off point.

[0070] The vehicle routing window processing module, within the vehicle routing time domain window, based on real-time vehicle status and travel requests, ... To develop a vehicle routing model for the target, and optimize vehicle routes, whereby... The parameter represents the vehicle's origin from node [node name missing]. To the node The cost of driving; As decision variables, Indicates service area Inner vehicle From node Drive to the node ,otherwise , For travel requests within service area r, For service area The set of all vehicles in the system; simultaneously satisfying , , , , , , , , ,in, Indicates vehicle Maximum passenger capacity Indicates each service area Inner vehicle Reaching the node Time, Indicates vehicle In the service area Inner slave node To the node Travel time, This indicates the maximum permitted driving time.

[0071] The prediction time-domain window processing module divides the prediction time-domain window into multiple time intervals and predicts future travel demand based on historical vehicle status and travel requests using a short-term prediction model.

[0072] The rebalancing scheduling time-domain window processing module, based on vehicles in idle state, uses... A vehicle rebalancing model is developed to obtain a cross-regional vehicle scheduling plan and an updated regional vehicle distribution, wherein: , For service area Service level difference variables, As decision variables, Indicates vehicle From service area Dispatch to service area ,otherwise ; Indicates vehicle From service area Dispatch to service area Time; To balance service level differences With vehicle dispatch time The weight parameters for importance must simultaneously satisfy: , , , , , , , .

[0073] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for ride-hailing routing and rebalancing scheduling based on rolling time domain, characterized in that, include: Obtain information on the operating areas and hours of ride-hailing services; The ride-hailing operation area is divided into multiple non-overlapping service areas, and the ride-hailing operation time period is divided into multiple non-overlapping time domain sequences; The time-domain sequence is divided into vehicle routing time-domain windows, demand forecasting time-domain windows, and rebalancing scheduling time-domain windows, with partial overlap between adjacent time-domain windows; in the k-th time-domain sequence, the start time of the vehicle routing time-domain window is... Represented as, ,in Let be the start time of the rebalancing scheduling time-domain window in the (k-1)th time-domain sequence. To rebalance the length of the scheduling time-domain window, Representing time-domain sequences Rebalancing scheduling time-domain window and time-domain sequence The overlap length of the vehicle routing time-domain window; the start time of the demand prediction time-domain window. Represented as, , The length of the vehicle routing time-domain window. Representing time-domain sequences Vehicle routing time-domain window and time-domain sequence The overlap length of the demand forecasting time-domain window; the start time of the rebalancing scheduling time-domain window. Represented as, , The length of the time-domain window for demand forecasting. Representing time-domain sequences Demand forecasting time-domain window and time-domain series The overlap length of the rebalancing scheduling time-domain window; the time-domain sequence is represented as ; Obtain the vehicle status and passenger travel requests of each ride-hailing vehicle in each service area during the operating period. The vehicle status includes whether it is idle or carrying passengers, and the travel requests include pick-up and drop-off points. Within the vehicle routing time-domain window, based on real-time vehicle status and travel requests, To develop a vehicle routing model for the target, and optimize vehicle routes, whereby... The parameter represents the vehicle's origin from node [node name missing]. To the node The cost of driving; As decision variables, Indicates service area Inner vehicle From node Drive to the node ,otherwise , For travel requests within service area r, For service area The set of all vehicles in the system; simultaneously satisfying , , , , , , , , ,in, Indicates vehicle Maximum passenger capacity Indicates each service area Inner vehicle Reaching the node Time, Indicates vehicle In the service area Inner slave node To the node Travel time, Indicates the maximum permitted driving time. Indicates the service area Inside, vehicle Does it depart from its boarding point to the node? binary decision variables, =1 indicates that in the service area Inside, vehicle Depart from its boarding point to the node ,otherwise =0; Indicates the service area Inside, vehicle From node The binary decision variable for driving to its drop-off point, =1 indicates that in the service area Inside, vehicle From node Drive to their drop-off point, otherwise =0; Indicates the service area Inside, vehicle From node Drive to the node binary decision variables, =1 indicates that in the service area Inside, vehicle From node Travel to node i, otherwise =0; Represented as in the time domain window At the beginning, the vehicle The initial operating state, when the vehicle is not idle. ; The set of all passengers in the service area r; The prediction time window is divided into multiple time intervals, and future travel demand is predicted based on the historical vehicle status and travel requests using a short-term prediction model. Within the rebalancing scheduling time window, vehicles in idle state are... A vehicle rebalancing model is developed to obtain a cross-regional vehicle scheduling plan and an updated regional vehicle distribution, wherein: , For service area Service level difference variables, As decision variables, Indicates vehicle From service area Dispatch to service area ,otherwise ; Indicates vehicle From service area Dispatch to service area Time; To balance service level differences With vehicle dispatch time The weight parameters for importance must simultaneously satisfy: , , , , , , , ,in, This indicates the target service area within the current rebalancing scheduling service area. Passenger travel demand forecasts; Data collection is carried out for the next time series, forming a rolling closed loop of vehicle routing time domain window, demand forecasting time domain window, and rebalancing scheduling time domain window.

2. A ride-hailing routing and rebalancing scheduling system based on rolling time domain, characterized in that, Includes a module for executing the ride-hailing routing and rebalancing scheduling method based on the rolling time domain as described in claim 1: The acquisition module is used to obtain the operating area and operating hours of ride-hailing services; The operation area division module is used to divide the ride-hailing operation area into multiple non-overlapping service areas; The operation period segmentation module is used to divide the ride-hailing operation period into multiple non-overlapping time-domain sequences; The time-domain sequence partitioning module is used to partition the time-domain sequence to form a vehicle routing time-domain window, a demand forecasting time-domain window, and a rebalancing scheduling time-domain window, with at least partial overlap between adjacent time-domain windows; The data acquisition module obtains the vehicle status of each ride-hailing vehicle and the travel requests of passengers in each service area during the operating period. The vehicle status includes whether the vehicle is idle or carrying passengers, and the travel requests include the pick-up point and the drop-off point. The vehicle routing window processing module, within the vehicle routing time domain window, based on real-time vehicle status and travel requests, ... To develop a vehicle routing model for the target, and optimize vehicle routes, whereby... The parameter represents the vehicle's origin from node [node name missing]. To the node The cost of driving; As decision variables, Indicates service area Inner vehicle From node Drive to the node ,otherwise , For travel requests within service area r, For service area The set of all vehicles in the system; simultaneously satisfying , , , , , , , , ,in, Indicates vehicle Maximum passenger capacity Indicates each service area Inner vehicle Reaching the node Time, Indicates vehicle In the service area Inner slave node To the node Travel time, Indicates the maximum permitted driving time. Indicates the service area Inside, vehicle Does it depart from its boarding point to the node? binary decision variables, =1 indicates that in the service area Inside, vehicle Depart from its boarding point to the node ,otherwise =0; Indicates the service area Inside, vehicle From node The binary decision variable for driving to its drop-off point, =1 indicates that in the service area Inside, vehicle From node Drive to their drop-off point, otherwise =0; Indicates the service area Inside, vehicle From node Drive to the node binary decision variables, =1 indicates that in the service area Inside, vehicle From node Travel to node i, otherwise =0, Represented as in the time domain window At the beginning, the vehicle The initial operating state, when the vehicle is not idle. ; The set of all passengers in the service area r; The prediction time-domain window processing module divides the prediction time-domain window into multiple time intervals and predicts future travel demand based on historical vehicle status and travel requests using a short-term prediction model. The rebalancing scheduling time-domain window processing module, based on vehicles in idle state, uses... A vehicle rebalancing model is developed to obtain a cross-regional vehicle scheduling plan and an updated regional vehicle distribution, wherein: , For service area Service level difference variables, As decision variables, Indicates vehicle From service area Dispatch to service area ,otherwise ; Indicates vehicle From service area Dispatch to service area Time; To balance service level differences With vehicle dispatch time The weight parameters for importance must simultaneously satisfy: , , , , , , , ,in, This indicates the target service area within the current rebalancing scheduling service area. Passenger travel demand forecasts.

Images