Multi-vehicle-type electric shared vehicle scheduling method under predetermined order

Abstract

The invention discloses a multi-vehicle-type electric shared vehicle scheduling method under a predetermined order, and the method comprises the following steps: constructing a mixed integer linear programming model according to the information of the predetermined order on the current day, and constructing a joint decision of vehicle delivery, vehicle type selection, order selection, charging planning and personnel scheduling by the model by taking system revenue maximization as an objective function; solving the mixed integer linear programming model by using a solver to generate a vehicle scheduling scheme and a scheduling personnel path planning scheme; according to the invention, the benefit of the shared travel platform in the whole operation process is obviously improved, and the operation cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of resource placement and scheduling, and in particular to a method for scheduling multiple types of electric shared vehicles under a predetermined order. Background Art

[0002] Due to the significant differences in maintenance, repair, relocation, and holding costs between electric shared cars and shared bicycles, electric shared cars are a scarce resource and cannot be deployed in large quantities in real life. In order to ensure that operators can make stable profits and maintain customer satisfaction at an acceptable level, reservation services are widely used in the operation of shared travel platforms. In the field of shared travel, users generally need to make reservations on the platform one or a few days in advance for the time and location of the car, so that operators have sufficient time to respond to and assign orders. For customers, this way of providing relatively stable services is very attractive to travel users who often plan ahead. Even during peak travel times and when resources are limited, it can minimize user waiting time and improve travel experience. For enterprises, the reservation service reserves sufficient time for the dispatch of enterprise vehicles. Enterprises can use this time to inspect, maintain, and charge vehicles to maximize user experience and improve user stickiness. On the other hand, the reservation-based operation model gives enterprises more time to calculate costs and revenues, providing them with insights into market changes and analytical ideas for designing corresponding marketing promotion activities, avoiding losses caused by blindly expanding order volumes, and allowing enterprises to use more limited resources to meet key orders that can maximize corporate profits. Therefore, the operation of electric shared cars based on the reservation model is a type of shared service model that can better resist market risks and maintain corporate profits.

[0003] The common process of booking electric shared cars is as follows: users initiate car needs on the platform in advance, including the pick-up time, rental duration, and pick-up and return locations. The system will estimate the distance and cost of the trip based on the starting point and the time of use. This series of information will be uniformly fed back to the operator. The operator will dispatch and plan the vehicle within the specified time, and feedback the order response and assignment results to the user. In actual situations, some orders are often not successfully responded to due to vehicle and personnel restrictions.

[0004] In the B2C (Business to Customer) short-term shared car rental model, the operator owns the vehicle and is the developer and executor of the vehicle scheduling decision plan. Therefore, the operator needs to consider various constraints to develop an optimal vehicle scheduling plan in order to maximize profits. A complete scheduling plan needs to solve four problems: (1) order selection and assignment; (2) the number of vehicles put into use and their placement at the beginning of the operation period; (3) the charging time planning of electric vehicles; and (4) the routes of the dispatchers responsible for moving electric vehicles during the operation period. Summary of the invention

[0005] Purpose of the invention: The purpose of the present invention is to provide a method for dispatching multiple types of electric shared cars under a pre-order, which is based on the initial resource allocation and dispatching problems faced by the electric shared car system based on the pre-order, and solves the vehicle dispatching problem in the pre-order sharing business.

[0006] Technical solution: The method for dispatching multiple types of electric shared cars under a pre-order described in the present invention comprises the following steps: constructing a mixed integer linear programming model based on the pre-order information of the day, the model takes maximizing system revenue as the objective function, and constructs a joint decision based on vehicle deployment, model selection, order selection, charging planning and personnel scheduling; using a solver to solve the mixed integer linear programming model to generate a vehicle scheduling plan and a scheduling personnel path planning plan.

[0007] Furthermore, the mixed integer linear programming model includes: a vehicle mobile network graph (Car Network, CN) and a dispatcher mobile network graph (Relocation-staff Network, RN), the two networks share the same set of node sets and are composed of different arcs; wherein the node set consists of an order starting point, an order end point, a charging station node, and a warehouse node; the objective function includes system revenue and system total expenditure, wherein the system revenue is the service revenue after a successful order response, and the system total expenditure includes vehicle relocation cost, dispatcher salary cost, vehicle charging cost, and vehicle daily average depreciation and amortization cost.

[0008] Furthermore, the arcs in the vehicle mobile network graph CN include: warehouse arc: connecting the warehouse node with the order starting point or the charging station node, for traffic balancing; order arc: directly connecting the order starting point i and the end point i+N; relocation arc: including arcs from the order end point to other order starting points, the order end point to the charging station, and the charging station to other order starting points, for vehicle relocation operations.

[0009] Furthermore, the arcs in the dispatcher mobile network graph RN include: the warehouse arc, which is the path for dispatchers to depart from or return to the warehouse node; the vehicle delivery arc, which is the path for dispatchers to perform vehicle relocation, corresponding to the relocation arc in CN; and the vehicle pickup arc: the path for dispatchers to go to the vehicle pickup point, including the path from the order starting point or charging station to other nodes.

[0010] Furthermore, the constraints of the model include: vehicle network constraints, including network node constraints, node time constraints and vehicle power constraints; relocation network constraints, including network node constraints and node time constraints; network connection constraints, which are used to connect the node inflow and outflow and time constraints of the vehicle network and the relocation network.

[0011] Furthermore, the vehicle network constraints specifically include: each order is responded to by at most one vehicle; each vehicle departs from the warehouse and eventually returns to the charging station; the battery power of the vehicle during operation decreases linearly with the distance traveled and increases linearly with the charging time; the charging time of the vehicle at the charging station is not fixed, but the last charge before returning to the warehouse must be a full charge.

[0012] Furthermore, the relocation network constraints specifically include: each dispatcher visits each node at most once; each dispatcher starts from the warehouse and eventually returns to the warehouse; the dispatcher's relocation path is consistent with the vehicle relocation path.

[0013] Furthermore, the network connection constraints specifically include: the vehicle and the dispatcher pass through the relocation arc at the same time and have the same departure time; the paths of the vehicle and the dispatcher are connected through the common inflow and outflow of the nodes and the time constraints.

[0014] Furthermore, the model assumptions include the following: the operator has sufficient vehicles of various models and basic supporting facilities for charging piles; each electric vehicle is fully charged at the beginning of the operation period and returns to the charging station for charging at the end of the operation; vehicle relocation is only carried out during the operation period; the differences between different models are reflected in the cruising range and average daily depreciation and amortization; vehicles only exist in four states: service, relocation, charging and waiting.

[0015] Furthermore, the model is solved using a commercial solver, and the vehicle placement, vehicle model selection, charging plan, and personnel scheduling path are determined based on the model solution results.

[0016] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: the present invention constructs a joint decision-making model that takes the operation of one-way free-floating electric shared cars under the B2C (Business to Customer) mode as the research object, and considers vehicle deployment, model selection, charging planning, order selection and personnel scheduling, expanding and enriching the resource deployment and scheduling theory in the car reservation sharing business; the proposed vehicle and employee joint scheduling decision model introduces multiple model alternative assumptions and joint optimization of multiple operation decision stages, avoiding the short-sightedness of decisions in a single operation stage, thereby ensuring the maximization of platform revenue during the entire operation process; the introduction of multiple models can deeply explore the model configuration strategy of the shared car operation platform under different order scale levels, and provide model decision support for the platform to make efficient profits; by comprehensively considering the electric vehicle reservation sharing resource scheduling problem of multiple models and multiple operation stages, using advanced mathematical optimization technology and efficient business solvers, not only the revenue of the shared travel platform in the entire operation process is significantly improved, the operating cost is reduced, but also the platform's recognition and response to high-value orders is improved, and the ability to quickly respond to market changes is realized. Compared with the prior art, it shows obvious advantages in both revenue and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2 is a vehicle path planning diagram of the present invention; Figure 3 It is the scheduling staff path planning diagram of the present invention. DETAILED DESCRIPTION

[0018] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0019] like Figure 1 As shown, the embodiment of the present invention provides a method for dispatching multiple models of electric shared cars under a scheduled order. According to the scheduled order information of the day, a mixed integer linear programming model is constructed. The model takes the maximization of system revenue as the objective function, and constructs a joint decision-making of vehicle deployment, model selection, order selection, charging planning and personnel scheduling; the mixed integer linear programming model is solved by a solver to generate a vehicle scheduling plan and a scheduling personnel path planning plan; specifically, the following steps are included: S1 Model Assumptions: The following assumptions are made about the key characteristics of the electric shared car operation process: Facility resources: The operator has sufficient number of vehicles of various models and has built enough basic supporting facilities of charging piles at the charging stations.

[0020] Vehicle operation: At the beginning of the operation period, each electric vehicle (EV) is deployed at the origin of its first user in a fully charged state. During the operation period, the EV can be charged at any charging station at any time. At the end of the day, the EV will be transferred to the charging station for charging.

[0021] Vehicle relocation: Vehicle relocation is carried out during the operation period. EV relocation outside the operation period, such as reallocating EVs to the first batch of customers designated for the next operation period, belongs to another research issue and is not within the scope of this invention.

[0022] Charging and replenishment: All EVs are fully charged at the beginning and end of the operation cycle. The battery power decreases linearly with the increase of mileage and increases linearly with the charging time. During the operation period, the charging and replenishment of EVs adopts a partial charging strategy, that is, the charging time of EVs at charging stations is not fixed, and the last charge of EVs before returning to the depot must adopt a full charging strategy.

[0023] Differences in vehicle models: The differences between different vehicle models are reflected in the cruising range and average daily depreciation and amortization, while the performance loss and other differences of the vehicles during operation are not taken into account in this invention.

[0024] Vehicle status: There are only four vehicle statuses: service, relocation, charging, and waiting. The situation where the vehicle is interrupted due to a malfunction is not considered.

[0025] S2 Mathematical Modeling Consider that there are two types of paths in this problem, namely, vehicle movement paths and dispatcher movement paths, such as Figure 1 As shown. Using the modeling method of the dynamic dispatching problem of electric motorcycles, the vehicle movement network diagram (Car Network, CN) and the dispatching staff movement network diagram (Relocation Staff Network, RN) are used to characterize the movement paths of vehicles and dispatching staff respectively. These two directed graphs share the same set of nodes (V), but are composed of different arcs (Arc). According to this modeling method, the problem can be expressed as a vehicle routing problem with time windows (Vehicle Routing Problem with Simultaneous Pickup and Delivery and Time Windows, VRPSPDTW). Unlike the traditional VRPSPDTW problem, the present invention first deals with the order response problem, determines the nodes that need to be picked up and delivered, and then plans and solves the dispatching personnel path. The entire planning scheme needs to be solved under the condition of maximizing the total revenue of the system, rather than solving the two sub-problems separately. The components and meanings of the network nodes are explained in detail below.

[0026] The S21 node set V consists of four types of nodes: order starting point , order end point , repeated charging station nodes , warehouse node ; Where N represents N OD pairs, that is, N orders, and the OD pair of the order is represented by, which represents the number of charging stations. In order to simplify the construction of the model, each order has a backup of this charging station. The node replication method has the following advantages in model construction: it ensures that the vehicle is charged at most once in a single rental, but during the entire operation period, the vehicle can go to the corresponding station for charging multiple times; it prevents the sub-circuit or problem of vehicle or relocation personnel from being infeasible due to multiple visits to a node.

[0027] Take 10 orders and 2 charging stations as an example , 10 orders OD The pairs are represented as (1, 11), (2, 12), … (10, 20) respectively. There are 20 charging station nodes in total ( ), where (21, 22) corresponds to the charging station node set of the first order, and (39, 40) corresponds to the charging station node set of the tenth order.

[0028] S22 Vehicle Mobility Network Diagram: The arcs on CN represent the actual movement path of electric vehicles (EVs). CN is a directed graph. G C =(V,A C ） , V is the set of all nodes, A C It consists of three types of arcs in CN: warehouse arc, order arc, and relocation arc.

[0029] a warehouse arc :The warehouse arc is the outflow of each vehicle from the warehouse to the order origin and the flow of each vehicle from the last charging station to the warehouse The warehouse arc is set to balance the flow in the network and avoid vehicle sub-loops. Because each vehicle is already at the first user starting point of the day's service at the beginning of the operation and at the end of the operation at the charging station, it is necessary to set a virtual start node and end node for each vehicle. , indicating that the vehicle has not been put into use. The specific expression is as follows: ; ; ; It is worth noting that on the warehouse arc, there is no actual vehicle movement and the travel time is also 0, because the present invention is about the path planning problem of decision-making vehicles and dispatchers during the operation period, and outside the operation period, such as relocating the vehicle to the starting point of the next day's service order every night, is not within the scope of the present invention.

[0030] b Order Arc :The order arc represents the actual moving path of the vehicle for each order, expressed as The travel time of a vehicle on each arc is only related to the length of the arc, and it is assumed here that the vehicle maintains a constant speed throughout the entire movement process.

[0031] c relocation arc : The relocation arc is composed of all arcs that may be relocated, expressed as: ; ; ; in, represents the end point corresponding to order i, is the set of backup charging station nodes corresponding to order i, is the starting point of all orders except order i. The movement of the vehicle on these arcs indicates that the vehicle has been relocated by the operator.

[0032] S23 dispatcher mobile network diagram: d warehouse arc : Indicates that the dispatcher departs from or returns to the warehouse node, which is specifically composed of the following arcs: ; Unlike the warehouse arc in the vehicle network diagram, employees have actual travel time when passing through the warehouse arc. , it indicates that the employee did not perform the vehicle relocation task on that day.

[0033] e delivery arc :The delivery arc indicates that the dispatcher needs to relocate an EV to the next order starting point or the corresponding charging station. Here, the delivery arc has the same meaning as the relocation arc in the vehicle movement network diagram.

[0034] f Pick up arc : The pick-up arc represents that the relocation personnel go through this arc to the end of the arc to pick up the car. After the pick-up is completed, the car moving operation starts immediately. On the pick-up arc, only personnel move, but no vehicle moves. It does not include the arc starting from the warehouse node to pick up the car. It can be specifically represented as the following set: ; in 、 Indicates order i The corresponding starting point, end point and charging station, is the endpoint of all orders other than the order, A collection of backup charging stations for orders other than the order.

[0035] S24 builds the objective function The modeling in the present invention considers maximizing the operator's profit. By examining the income and expenditure items in the entire system, it can be found that the system income only comes from the service income after the order is successfully responded to, as shown in formula (1): (1); The total system expenditure consists of four parts: the relocation cost of the vehicle repositioning on the day, the basic salary cost of the dispatcher on the day, the charging cost during the vehicle operation period, and the average daily depreciation and amortization cost of the vehicle. The vehicle rebalancing cost is the sum of the relocation costs of all vehicles, as shown in formula (2); the charging cost is calculated by calculating the total charging amount of all vehicles during the operation period, as shown in formula (3).

[0036] (2); (3); The fixed cost of a vehicle is expressed as the sum of the daily depreciation and amortization costs of all vehicles dispatched on that day. Since each vehicle is already at the starting point of the first order served on that day at the beginning of operation, a depot node is set as the starting and ending node of each vehicle for the convenience of modeling. If each vehicle is put into use on that day, it must go from the depot node to the order starting point, so it is only necessary to sum the depot→O node (i.e. The fixed cost of each vehicle can be calculated as shown in formula (4): (4); The fixed salary of employees on that day is expressed as the sum of the basic salaries of all employees dispatched on that day to perform relocation operations, as shown in formula (5): (5); The overall objective function can be expressed as all revenue items minus cost items, as shown in formula (6): (6); S25 establishes constraints: S251 Vehicle Network Constraints g Network node constraints For each order, at most one vehicle responds to the service, as shown in formula (7): (7); If a vehicle is going out to perform an order task on the same day, then the vehicle must start from the warehouse 0 node and go to a certain order starting point O Otherwise, the car byte passes through the warehouse arc, from 0 to At the same time, each vehicle that executes an order must go to the charging station for charging before returning to the warehouse on the same day, which can be achieved through constraints (8) and (9).

[0037] (8); (9); In the entire network graph, since the nodes are replicated, in order to prevent the nodes from being visited multiple times, each node is unique and can be visited by at most one vehicle, unless the node is a depot node. At the same time, for each vehicle, once it enters any node in the network (except the depot), it must exit the node to balance the traffic of the entire network. The above two restrictions can be achieved by constraints (10) and (11).

[0038] (10); (11); After each order is completed, the EV is located at its corresponding charging station to perform a charging operation at most once, as shown in constraint (12): (12); h-node time constraint The time when each vehicle departs from the depot node is time 0. If the order i By vehicle k Services, then vehicles k Orders must be met i The time window constraint is as shown in equations (13) and (14): (13); (14); If the vehicle k When a vehicle passes through an arc in the vehicle network, the two nodes before and after it must satisfy the corresponding time constraints, as shown in equations (15)-(17): (15); (16); (17); iVehicle power limit All vehicles are fully charged when they depart from the depot. If they do not pass through the arc on the charging station, the remaining range of the vehicle cannot jump. When passing through the arc on the charging station, the vehicle's power can jump, but it cannot exceed the vehicle's k The maximum range of the vehicle is that it must be fully charged at the charging station before returning to the depot after completing the order for the day. The above constraints can be satisfied by equations (18)-(21): (18); (19); (20); (twenty one); S252 Relocation Network Constraints J Network Node Constraints It is consistent with the node constraints in the vehicle network. O or D , is visited at most once, because each person goes O or D It must be a pick-up or delivery operation. Once accessed, it means that the node task has been completed by an employee. Subsequent accesses will only increase the relocation cost.

[0039] For all nodes except depot, each dispatcher r Maximum visit: 1 time, staff r A node cannot be visited repeatedly. Each node must satisfy the equal flow of people in and out constraint. Every employee who needs to perform relocation operations on the day must start from the depot and return to the depot after completing the day's tasks. The above conditions can be achieved by constraints (22)-(26): (twenty two); (twenty three); (twenty four); (25); (26); k-node time constraint In the relocation network, the time when each dispatcher departs from the depot is 0, and if the dispatcher r Through the arc , then the departure time of the two nodes must satisfy the relevant constraints. As shown in constraints (27) and (28): (27); (28); S253 Network connection constraints The two networks are connected mainly through the common inflow and outflow of nodes and the common departure time of nodes. k and the employees who will be relocated r must pass through the arc at the same time, and employees r and vehicles k The time of leaving the starting point of the relocation arc should be consistent. The above constraints can be achieved by equations (29) and (30): (29); (30); S3 uses the commercial solver GUROBI for simulation to verify the effectiveness of the model and generate vehicle dispatch plans and dispatcher path planning plans.

[0040] The following is an explanation of the symbols used in the present invention.

[0041] Table 1 Description of the sets in the VRPSPDTW mathematical model ; Table 2 Parameters in VRPSPDTW mathematical model ; Table 3 Description of decision variables in VRPSPDTW mathematical model .

[0042] like Figure 2-Figure 3 As shown, the present invention will verify the effectiveness of the above model in solving the scheduling problem of multi-model electric shared cars under scheduled orders through a numerical experiment of 10 orders. Assume that there are 10 scheduled orders and 1 charging station in the current system. The OD pair location information, rental duration, departure time window, and mileage of each order are shown in Table 4.

[0043] Table 4 Order reservation information ; The distance between different nodes is expressed by the Euclidean distance between the node centers. The location coordinates of each node, charging station, and depot node are shown in Table 5. There are currently three models available in the fleet, namely A, B, and C. The range of the three models is 100km, 180km, and 320km, respectively, and their daily average fixed depreciation and amortization costs are 100 CNY, 150 CNY, and 180 CNY, respectively. The unit mileage relocation cost is 1.5 CNY / km, the unit electricity cost is 1.2 CNY / km, the unit mileage order service charge is 15 CNY / km, and the daily basic salary of each dispatcher is 90 CNY.

[0044] Table 5 Node coordinates

[0045] By using the solution software GUROBI to optimize the vehicle dispatching paths and dispatching personnel movement paths for the above 10 orders, the results of the multi-model vehicle dispatching plan under the scheduled orders are shown in Table 6 below. Vehicle dispatching plan It indicates that a k-th vehicle of model h starts from the depot node, passes through the nodes along the way, leaves this node, and has the remaining range of the vehicle.

[0046] Dispatching employee movement routes It shows the route that employee r takes from the depot node to perform the pick-up and delivery task, as well as the time to leave each node. Through this itinerary chain, we can get the scale of fleet operations, initial placement location and order response under the existing orders, as well as the charging time planning for each vehicle, to help operators better manage vehicles. In this example, GUROBI obtained the solution to the problem, gap=0%, the calculation time was 2527.3s, 3 Model A vehicles were used to complete the response to 8 orders, and 2 employees were used to complete the relocation, and the net profit that could be obtained was 753.01CNY.

[0047] Table 6 Vehicle and personnel dispatch routes .

Claims

1. A method for dispatching multiple types of electric shared vehicles under a pre-order, characterized in that: The following steps are involved: Based on the order information of the day, a mixed integer linear programming model is constructed. The model takes the maximization of system revenue as the objective function and constructs a joint decision-making based on vehicle deployment, model selection, order selection, charging planning and personnel scheduling; The solver is used to solve the mixed integer linear programming model to generate vehicle scheduling solutions and dispatcher path planning solutions.

2. The method for dispatching multiple types of electric shared vehicles under a predetermined order according to claim 1, characterized in that: The mixed integer linear programming model includes: a vehicle mobile network graph CN and a dispatcher mobile network graph RN. The two networks share the same set of nodes and are composed of different arcs. The node set consists of the order starting point, the order end point, the charging station node and the warehouse node. The objective function includes system revenue and system total expenditure. The system revenue is the service revenue after the order is successfully responded to, and the system total expenditure includes vehicle relocation cost, dispatcher salary cost, vehicle charging cost and vehicle daily average depreciation and amortization cost.

3. The method for dispatching multiple types of electric shared vehicles under a predetermined order according to claim 2, characterized in that: The arcs in the vehicle mobile network graph CN include: warehouse arc: connecting the warehouse node with the order starting point or charging station node, used for traffic balancing; order arc: directly connecting the order starting point i and the end point i+N; relocation arc: including arcs from the order end point to other order starting points, the order end point to the charging station, and the charging station to other order starting points, used for vehicle relocation operations.

4. The method for dispatching multiple types of electric shared vehicles under a predetermined order according to claim 2, characterized in that: The arcs in the dispatcher mobile network graph RN include: the warehouse arc, which is the path for dispatchers to depart from or return to the warehouse node; the vehicle delivery arc, which is the path for dispatchers to relocate vehicles, corresponding to the relocation arc in CN; the vehicle pickup arc: the path for dispatchers to go to the vehicle pickup point, including the path from the order starting point or charging station to other nodes.

5. The method for dispatching multiple types of electric shared vehicles under a predetermined order according to claim 2, characterized in that: The constraints of the model include: vehicle network constraints, including network node constraints, node time constraints and vehicle power constraints; relocation network constraints, including network node constraints and node time constraints; network connection constraints, which are used to connect the node inflow and outflow and time constraints of the vehicle network and the relocation network.

6. The method for dispatching multiple types of electric shared vehicles under a predetermined order according to claim 5, characterized in that: The vehicle network constraints include: each order is served by at most one vehicle; each vehicle starts from the warehouse and eventually returns to the charging station; the battery power of the vehicle during operation decreases linearly with the distance traveled and increases linearly with the charging time; The charging time of the vehicle at the charging station is not fixed, but the last charge before returning to the depot must be a full charge.

7. The method for dispatching multiple types of electric shared vehicles under a predetermined order according to claim 5, characterized in that: The relocation network constraints specifically include: each dispatcher visits each node at most once; each dispatcher starts from the warehouse and eventually returns to the warehouse; the dispatcher's relocation path is consistent with the vehicle relocation path.

8. The method for dispatching multiple types of electric shared vehicles under a predetermined order according to claim 5, characterized in that: The network connection constraints specifically include: vehicles and dispatchers pass through the relocation arc at the same time and leave at the same time; the paths of vehicles and dispatchers are connected through the common inflow and outflow of nodes and time constraints.

9. The method for dispatching multiple types of electric shared vehicles under a predetermined order according to claim 1, characterized in that: Model assumptions include the following: The operator has sufficient vehicles of various models and basic supporting facilities of charging piles; each electric car is fully charged at the beginning of the operation period and returns to the charging station for charging at the end of the operation; vehicle relocation is only carried out during the operation period; the differences between different models are reflected in the cruising range and average daily depreciation and amortization; vehicles only exist in four states: service, relocation, charging and waiting.

10. The method for dispatching multiple types of electric shared vehicles under a predetermined order according to claim 1, characterized in that: The model is solved using a commercial solver, and based on the model solution results, the vehicle deployment location, vehicle model selection, charging plan, and personnel scheduling path are determined.

Images