Vehicle dispatching method, system, medium and product based on intermodal round trip transport

Abstract

The application provides a vehicle scheduling method and system based on intermodal circulation transportation, a medium and a product, and relates to the field of logistics transportation. The method comprises the following steps: planning an initial intermodal path connecting an existing transportation task node and a new transportation task node according to a transportation cost index; identifying an opportunistic battery swapping node in the initial intermodal path according to battery swapping station network data; calculating an opportunistic battery swapping cost generated by the vehicle deviating from the initial intermodal path to the opportunistic battery swapping node, completing battery swapping and returning to the path, and a necessary battery swapping cost generated by the vehicle abandoning battery swapping at the opportunistic battery swapping node and continuing to travel along the initial intermodal path until the power decreases to a preset threshold, so as to reach the nearest battery swapping station; and when the necessary battery swapping cost is greater than the opportunistic battery swapping cost, the corresponding opportunistic battery swapping node is added to the initial intermodal path to generate an intermodal circulation transportation plan. The method can reduce the comprehensive operation cost of the entire intermodal transportation cycle.

Description

Technical Field

[0001] This application relates to the field of logistics and transportation, and in particular to a vehicle scheduling method, system, medium, and product based on intermodal circular transportation. Background Technology

[0002] In the modern logistics and transportation industry, with the booming development of e-commerce and the refinement of supply chain management, the intermodal circular transportation model has been widely applied. Under this model, transport vehicles no longer perform single point-to-point tasks, but instead continuously complete multiple loading and unloading tasks according to a predetermined plan within a work cycle, forming a closed-loop or chain-like operational process. This model can significantly improve vehicle loading rates and operational efficiency. At the same time, in response to the global call for energy conservation and emission reduction and to lower fuel costs, more and more logistics companies are beginning to deploy pure electric vehicles (especially electric trucks that support rapid battery swapping) as their main transportation force, promoting the development of green logistics.

[0003] In related technologies, electric vehicles have limited driving range, and the time required for energy replenishment (charging or battery swapping) is much longer than that for refueling. A common vehicle scheduling method is for the scheduling system to pre-plan a complete driving route based on a determined sequence of transportation task nodes before the vehicle departs. To handle the energy replenishment problem along the way, the system incorporates an energy management logic based on a fixed threshold: the system simulates the vehicle's remaining battery charge (SOC) along the route, and when the simulated battery charge first falls below a preset safety threshold (e.g., 20%), the system will forcibly insert a replenishment instruction into the current route planning, instructing the vehicle to go to the nearest battery swapping station for battery swapping.

[0004] However, the battery swapping strategies of these technologies rely entirely on the battery level reaching a fixed, low "danger" threshold. When a vehicle is traveling through a complex road network, and the battery level drops to the threshold in an area far from main roads with few nearby battery swapping stations, the system may direct the vehicle to the "nearest" battery swapping station, potentially causing the vehicle to deviate significantly from its predetermined optimal transportation route, resulting in additional mileage and time consumption. This detour cost incurred to perform necessary battery swaps can significantly increase the overall operating cost of the entire transportation cycle. Summary of the Invention

[0005] This application provides a vehicle scheduling method, system, medium, and product based on intermodal circular transportation to address the problem that related technologies, due to their passive and emergency energy replenishment strategies, cause vehicles to make unplanned detours to complete necessary battery swaps, resulting in increased overall operating costs.

[0006] Firstly, this application provides a vehicle dispatching method based on intermodal circular transportation, applied to a vehicle dispatching system, the method comprising:

[0007] Obtain the new transportation task node for the new transportation order, and plan an initial connecting path between the existing transportation task node and the new transportation task node based on the preset transportation cost index.

[0008] Based on real-time acquired battery swapping station network data, opportunistic battery swapping nodes that can be swapped in the initial connecting route are identified, and the shortest travel distance between the opportunistic battery swapping node and the initial connecting route is less than a preset distance threshold.

[0009] Calculate the opportunistic battery swapping cost incurred when a vehicle deviates from the initial connecting route to complete a battery swap at the opportunistic battery swapping node and returns to the route, and the necessary battery swapping cost incurred when a vehicle abandons the opportunistic battery swapping node and continues to travel along the initial connecting route until the battery level drops to a preset threshold, in order to reach the nearest battery swapping station.

[0010] When the necessary battery swap cost is greater than the opportunistic battery swap cost, the opportunistic battery swap node corresponding to the opportunistic battery swap cost is added to the initial connecting route to generate a connecting loop transportation plan.

[0011] By adopting the above technical solution, when planning intermodal transport routes, the system no longer passively waits for vehicle battery levels to drop to a low threshold before searching for the nearest battery swapping station. Instead, it proactively identifies opportunistic battery swapping nodes near the initial route. By estimating and comparing the economic impact of "the cost of opportunistic battery swapping now" versus "the cost of necessary battery swapping when battery power is depleted in the future," the system can make more cost-effective decisions. When the system anticipates that the cost of necessary battery swapping in the future will be higher, it includes lower-cost opportunistic battery swapping nodes in the transport plan in advance. This proactive and predictive energy management approach avoids the high time and mileage costs incurred by forced detours due to battery depletion, transforming emergency, rigid battery swapping needs into planned, flexible battery swapping options. This reduces the overall operating cost of the entire intermodal transport cycle and improves the intelligence level of vehicle scheduling and overall transport efficiency.

[0012] In some embodiments, the step of planning an initial connecting route between the existing transportation task node and the new transportation task node based on a preset transportation cost index specifically includes:

[0013] Construct a spatiotemporal network graph that includes existing transportation task nodes and new transportation task nodes. The node attributes of the spatiotemporal network graph include task type, time window constraints, cargo loading and unloading requirements, and road network connectivity between nodes.

[0014] Based on the vehicle's current location, remaining battery power, cargo status, and assigned task sequence, a multi-dimensional cost function for the vehicle to complete a new transportation task is calculated. The multi-dimensional cost function includes travel distance cost, time cost, battery consumption cost, and task delay penalty cost.

[0015] The dynamic programming algorithm is used to search for a set of feasible paths in the spatiotemporal network graph that satisfy the time window constraint and the vehicle battery power constraint, and the comprehensive cost value of each feasible path is calculated according to the multi-dimensional cost function.

[0016] Select the path with the lowest overall cost from the set of feasible paths as the initial connecting path.

[0017] By adopting the above technical solution, when planning the initial connecting route, the system first constructs a spatiotemporal network graph containing multi-dimensional information such as time, space, and task attributes to comprehensively depict the complex transportation environment. Based on this, the system does not only consider a single travel distance but also establishes a multi-dimensional cost function by combining factors such as time cost, power consumption, and task delay penalties. Using a dynamic programming algorithm, the system can accurately search and select the path with the lowest overall cost value from numerous feasible paths that satisfy multiple constraints such as time windows and power consumption, ensuring that the initial connecting route itself is the optimal choice under multiple real-world constraints.

[0018] In some embodiments, when the vehicle abandons battery swapping at the opportunistic battery swapping node and continues traveling along the initial connecting route until the battery level drops to a preset threshold, the steps for incurring the necessary battery swapping costs to reach the nearest battery swapping station specifically include:

[0019] Based on the vehicle's current battery status and the predicted battery consumption curve, a warning location is determined when the vehicle travels along the initial connecting route to trigger a necessary battery swap. The warning location is the position where the vehicle's remaining battery power drops to the sum of a safety threshold and an emergency reserve battery power.

[0020] Based on the battery swapping station network data, determine the dynamic service capability data of all battery swapping stations within a preset range around the warning location point. The dynamic service capability data includes the number of available batteries at the battery swapping station, queuing time, battery swapping service efficiency, and real-time road condition information for reaching the battery swapping station.

[0021] Calculate the estimated cost of deviation from the warning location to each candidate battery swapping station. The estimated cost includes detour distance cost, queuing time cost, battery swapping service time cost, and cascading cost of subsequent task delays caused by emergency battery swapping.

[0022] The Monte Carlo simulation method is used to evaluate the necessary battery swapping risk cost under different traffic conditions and battery swapping station load conditions, and the risk cost is weighted and summed with the estimated cost to obtain the comprehensive necessary battery swapping cost.

[0023] By adopting the above technical solution, the system does not simply estimate detour distance when calculating the "necessary battery swapping cost." It first accurately locates the points where vehicles might trigger low battery warnings based on power consumption prediction. Then, the system not only considers the distance cost of detouring to the battery swapping station but also integrates dynamic service capacity data of the station, such as queuing time and service efficiency, making the cost calculation more realistic. More importantly, by introducing Monte Carlo simulation to assess the risk costs caused by changes in traffic and battery swapping station load, the system quantifies uncertainties and predicts the cascading costs of subsequent task delays that may result from emergency battery swapping. This makes the assessment of the "necessary battery swapping cost" more comprehensive and accurate, greatly improving the reliability of decisions compared with opportunistic battery swapping costs.

[0024] In some embodiments, after the step of adding the opportunistic battery swapping node corresponding to the opportunistic battery swapping cost to the initial connecting route to generate a connecting loop transportation plan, the method further includes:

[0025] Send a reservation request to the battery swapping station corresponding to the opportunistic battery swapping node in the intermodal transport plan, and obtain battery swapping time window confirmation information;

[0026] Based on the battery swapping time window confirmation information, the dwell time of the vehicle at each transportation task node is adjusted to generate an updated transportation plan that includes battery swapping time constraints.

[0027] The updated transportation plan is pushed to the vehicle terminal, and a battery swapping reminder is sent when the vehicle approaches an opportunistic battery swapping node.

[0028] By adopting the above technical solution, after determining the optimal opportunistic battery swapping node, the system does not simply add the node to the route, but performs a series of closed-loop collaborative operations. First, the system sends a reservation request to the battery swapping station, transforming the uncertain battery swapping behavior into a deterministic event with a guaranteed time window. Then, based on the time confirmed by the battery swapping station, the system intelligently adjusts the dwell time of the transportation task nodes in reverse, ensuring the feasibility of the entire transportation plan. Finally, by pushing the updated plan and reminder information to the driver, seamless connection is achieved from cloud-based decision-making to terminal execution. This series of operations ensures that opportunistic battery swapping can be executed smoothly and efficiently, avoiding the risk of vehicles arriving at the battery swapping station unable to swap batteries in time, and guaranteeing the stability and reliability of the entire intermodal transportation plan.

[0029] In some embodiments, prior to the step of identifying opportunistic battery swapping nodes in the initial connection path based on real-time acquired battery swapping station network data, the method further includes:

[0030] Obtain the actual power consumption records from the vehicle's historical transportation data, and calculate the average power consumption rate under different road sections and different load conditions;

[0031] The vehicle's power consumption in each road segment is estimated based on the average power consumption rate and the road segment attributes of the initial connecting route.

[0032] The power consumption values ​​of each road segment are summed up to determine the power change curve of the vehicle during its journey along the initial connecting route.

[0033] Based on the battery change curve, road sections where the vehicle's remaining battery power is within a preset range are selected as potential battery swapping areas for subsequent opportunistic battery swapping node identification.

[0034] By employing the aforementioned technical solution, the system utilizes historical vehicle transport data to learn and calculate the average power consumption rate under different conditions, thereby enabling more accurate prediction of the vehicle's power consumption curve along future routes. Based on this prediction curve, the system can pre-select "potential battery swapping areas" where the vehicle's power consumption is within the optimal swapping range. This approach avoids blindly searching along the entire route, concentrating computational resources on the most promising segments. This not only improves the computational efficiency of identifying opportunistic battery swapping nodes but also ensures that the selected nodes are at the appropriate power consumption timing, enhancing the overall performance and response speed of the scheduling algorithm.

[0035] In some embodiments, after the step of calculating the opportunistic battery swap cost incurred by the vehicle deviating from the initial connecting route to the opportunistic battery swap node to complete the battery swap and return route, the method further includes:

[0036] Obtain the opportunistic battery swapping cost of multiple opportunistic battery swapping nodes within the same potential battery swapping area;

[0037] The multiple opportunistic battery swapping nodes are sorted according to the opportunistic battery swapping cost to generate a priority list of battery swapping nodes;

[0038] Select the N nodes with the lowest cost from the battery swapping node priority list as candidate battery swapping nodes;

[0039] Calculate the success probability of battery swapping at the candidate battery swapping nodes, use the product of the success probability and the opportunistic battery swapping cost as a comprehensive evaluation index, and select the node with the best comprehensive evaluation index as the recommended battery swapping node.

[0040] By adopting the above technical solution, the system first sorts all candidate nodes according to the opportunistic battery swapping cost, initially screening out the nodes with the lowest costs. Furthermore, the system introduces the indicator of "probability of successful battery swapping," combining cost and success probability to form a comprehensive evaluation index. This method comprehensively considers both economic and reliability dimensions, avoiding the selection of a node with extremely low cost but a high risk of battery swapping failure. Finally, the node with the optimal comprehensive evaluation index is selected, ensuring that the recommended battery swapping solution for drivers is both economical and reliable.

[0041] In some embodiments, after the step of weighted summing of the risk cost and the estimated cost to obtain the comprehensive necessary battery swapping cost, the method further includes:

[0042] Extract actual necessary battery swapping cases for similar road sections and time periods from historical data, and calculate the average detour distance and average waiting time for historical cases;

[0043] The average detour distance and average waiting time are compared with the currently calculated necessary battery swapping cost to generate a cost deviation rate;

[0044] When the cost deviation rate exceeds a preset threshold, the main factors causing the deviation are identified. These main factors include changes in traffic congestion, changes in the service capacity of battery swapping stations, or errors in electricity prediction.

[0045] The calculation parameters for the necessary battery swapping cost are adjusted based on the aforementioned key factors, the adjusted necessary battery swapping cost is recalculated, and the adjusted necessary battery swapping cost is used for comparison with the opportunistic battery swapping cost.

[0046] By adopting the above technical solution, the system can identify and quantify cost deviations in real time by comparing the currently calculated cost with the actual cost under similar historical scenarios. Once the deviation exceeds a threshold, the system can trace back to the root causes, such as traffic congestion or changes in the service capacity of battery swapping stations, and use these factors to correct the parameters in the cost calculation model. This allows the system to obtain a more accurate and dynamically adaptable cost prediction value through risk assessment and self-learning mechanisms.

[0047] Secondly, this application provides a vehicle dispatching system, the system comprising: one or more processors and a memory;

[0048] The memory is coupled to the one or more processors. The memory is used to store computer program code, which includes computer instructions. The one or more processors call the computer instructions so that the system can implement the vehicle scheduling method based on inter-trip cyclic transportation provided in the above embodiments, which will not be described in detail here.

[0049] Thirdly, this application provides a computer-readable storage medium including instructions, characterized in that, when the instructions are executed on a vehicle dispatching system, the system can implement a vehicle dispatching method based on inter-trip cyclic transportation provided in the above embodiments, which will not be elaborated here.

[0050] Fourthly, this application provides a computer program product that, when run on a vehicle dispatching system, enables the system to implement a vehicle dispatching method based on intermodal cyclic transportation provided in the above embodiments, which will not be elaborated further here.

[0051] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0052] 1. This transforms the energy management of electric vehicles from a passive, threshold-based emergency response model to a proactive, economically-driven dynamic decision-making model. Instead of waiting until the battery is depleted to search for the nearest battery swapping station, it proactively calculates and weighs the low cost of immediate opportunistic battery swapping against the high-risk cost of future necessary battery swapping. This design philosophy turns battery swapping from a chore into a strategic tool for optimizing overall transportation costs, enhancing the intelligence and economic efficiency of vehicle dispatching.

[0053] 2. This application does not employ the static estimation techniques of traditional methods. Instead, it constructs a complex predictive model that integrates real-time dynamic data, probabilistic risk assessment, and self-learning correction capabilities. By introducing Monte Carlo simulation to assess the risks posed by uncertainties such as traffic and queuing at battery swapping stations, and by using historical data for comparative analysis and correction of the predictive model, the system can dynamically adapt cost predictions to environmental changes through a self-learning correction mechanism, thereby continuously improving the accuracy of decision-making.

[0054] 3. After determining the optimal battery swapping solution, the system does not simply issue a simple instruction, but proactively sends a reservation request to the battery swapping station to obtain a confirmed service time window. Based on this, it adjusts the dwell time of other nodes in the entire transportation chain. This mechanism, which locks and synchronizes optimization decisions with physical world resources (battery swapping station services), ensures that the theoretically optimal solution can be reliably executed. Attached Figure Description

[0055] Figure 1 This is a flowchart illustrating a vehicle scheduling method based on intermodal circular transportation in an embodiment of this application.

[0056] Figure 2 This is another flowchart illustrating a vehicle scheduling method based on intermodal circular transportation in an embodiment of this application;

[0057] Figure 3 This is a schematic diagram of the physical device structure of a vehicle dispatching system in an embodiment of this application. Detailed Implementation

[0058] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0059] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0060] Although this embodiment mainly uses battery-swapping electric vehicles as an example for illustration, those skilled in the art will understand that the energy replenishment decision model based on cost-benefit prediction proposed in this application is also applicable to optimizing the selection of charging stations for electric vehicles, only requiring the replacement of battery-swapping related parameters with charging related parameters.

[0061] For ease of understanding, the method provided in this implementation is described in process below. Please refer to [link / reference]. Figure 1 This is a flowchart illustrating a vehicle scheduling method based on intermodal circular transportation in an embodiment of this application.

[0062] S101. Obtain the new transportation task node for the new transportation order, and plan an initial connecting path between the existing transportation task node and the new transportation task node based on the preset transportation cost index.

[0063] Among them, a new transportation order refers to the latest transportation demand instruction received by the vehicle dispatching system that has not yet been included in the existing transportation plan, including core information such as cargo type, transportation origin and destination, and time requirements; a new transportation task node represents the cargo loading node (origin) and unloading node (end) specified in the new transportation order, which is the basic unit constituting the transportation route; the preset transportation cost index refers to the core parameters that the system pre-set to measure the economic efficiency of the transportation plan, which usually covers dimensions such as driving distance, time consumption, electricity cost, and task delay penalty; the initial connecting route refers to the continuous transportation route that connects existing task nodes and new task nodes, planned only based on the transportation cost index without considering the need for battery swapping.

[0064] Specifically, the dispatching system first obtains complete information about new transportation orders through the order receiving module, extracting key new transportation task nodes (such as the origin and destination in "Beijing Fangshan Warehouse → Tianjin Port Free Trade Zone"). Then, the system retrieves existing transportation task nodes stored in the database (such as the "Shijiazhuang Zhengding Warehouse → Beijing Fangshan Warehouse" task node currently pending execution) to determine the sequence of nodes that need to be connected. Next, based on preset transportation cost indicators, the system comprehensively evaluates the economics of different routes—for example, considering factors such as the mileage of different routes (affecting fuel / electricity costs), the traffic efficiency of the road segments (affecting time costs, such as the choice between highways and national roads), and whether they meet the task time window (such as goods needing to be delivered within 24 hours, with penalties for delays). Finally, it selects the route with the lowest overall cost as the initial connecting route (such as "Shijiazhuang Zhengding Warehouse → Beijing Fangshan Warehouse → Tianjin Port Free Trade Zone").

[0065] Optionally, the system can construct a "node-road network" mapping table containing existing task nodes and new task nodes, marking the road network connectivity between each node (such as whether there is a direct road or whether a detour is required); based on preset cost indicators, set weights for each indicator (such as a 40% weight for travel distance, a 30% weight for time cost, and a 30% weight for delay penalty), and calculate the "cost coefficient" between any two nodes; use Dijkstra's algorithm to traverse all possible node connection routes, calculate the total weighted cost of each route, and select the route with the lowest total cost as the initial connecting path.

[0066] Optionally, the system can also call the Geographic Information System (GIS) interface to obtain real-time road network data (such as road congestion and construction sections) around existing and new task nodes; based on the vehicle's current location, remaining battery power, and other status, routes that are beyond the vehicle's range or cannot be reached within the time window are excluded; for the remaining feasible routes, the travel distance cost, time cost, and delay risk cost are calculated separately, and the comprehensive cost of each route is accumulated to obtain the comprehensive cost of each route, and the route with the lowest cost is selected as the initial connecting route.

[0067] It is understandable that other methods can be used to achieve initial path planning, such as introducing intelligent optimization algorithms like genetic algorithms to search for paths in large-scale node networks more efficiently; this is not a limitation here.

[0068] S102. Based on the real-time acquired network data of the battery swapping stations, identify opportunistic battery swapping nodes in the initial connection path where battery swapping can be performed.

[0069] Among them, the network data of the battery swapping station refers to the data obtained by the dispatch system through real-time interaction with the battery swapping station management platform, including the location coordinates of the battery swapping station, the number of available batteries, the number of vehicles currently in the queue, the battery swapping service duration, and the operating status (normal / fault). Opportunistic battery swapping nodes refer to battery swapping stations around the initial connecting route where vehicles can temporarily deviate from the route to swap batteries without incurring excessive additional costs. The preset distance threshold refers to the distance standard (such as 5 kilometers) pre-set by the system to determine whether a battery swapping station is "opportunistic". Battery swapping stations exceeding this threshold are not considered opportunistic battery swapping nodes because the detour cost is too high.

[0070] Specifically, the dispatch system first establishes a real-time connection with the battery swapping station network platform via an API interface to obtain data on all battery swapping stations covering the area of ​​the initial connecting route, and then cleans the data (excluding faulty stations and stations without available batteries). Next, the system imports the coordinate sequence of the initial connecting route and the location coordinates of the battery swapping stations into the geographic information module to calculate the shortest travel distance between each station and the initial connecting route—for example, using the "path distance analysis" function of a GIS tool to locate the nearest perpendicular point from the station to the initial route and calculate the mileage from that perpendicular point to the station. Finally, the system compares the calculated shortest travel distance with a preset distance threshold (e.g., 3 kilometers), filtering out battery swapping stations with distances less than the threshold and marking them as opportunistic battery swapping nodes (e.g., if the initial route is the "Beijing → Tianjin" expressway section, and a battery swapping station is located 2 kilometers from the expressway exit and meets the distance threshold, it is identified as an opportunistic battery swapping node).

[0071] Optionally, the system divides the initial connecting route into multiple "segment units" (e.g., each unit is 10 kilometers long), records the start and end coordinates of each unit; for each segment unit, it searches for all battery swapping stations within a 5-kilometer radius (based on a preset distance threshold), obtains the real-time operational data of these battery swapping stations (e.g., the number of available batteries ≥ 2); excludes battery swapping stations with abnormal operating status or insufficient available resources, and the remaining battery swapping stations are the opportunistic battery swapping nodes corresponding to that segment unit, and summarizes the nodes of all segment units to obtain a complete list.

[0072] S103. Calculate the opportunistic battery swapping cost incurred when the vehicle deviates from the initial connecting route to an opportunistic battery swapping node to complete the battery swap and return to the route, and the necessary battery swapping cost incurred when the vehicle abandons the opportunistic battery swapping node and continues to travel along the initial connecting route until the battery level drops to a preset threshold, in order to reach the nearest battery swapping station.

[0073] Among them, the necessary battery swapping cost refers to the cost incurred by a vehicle in order to reach the nearest battery swapping station when it abandons the opportunity to swap batteries and continues to travel along the initial path until the battery level drops to a preset threshold. This cost usually includes emergency detour costs, battery swapping station queuing costs, and penalty costs for delaying subsequent tasks due to battery swapping. The nearest battery swapping station refers to the nearest battery swapping station with service capabilities in the vicinity when the vehicle's battery level drops to the preset threshold.

[0074] Specifically, when the system calculates the opportunistic battery swap cost, it first determines the detour distance the vehicle travels from the initial path to the battery swap node (e.g., 10 km from the deviation point on the initial path to the battery swap station, 10 km return path, total detour of 20 km). Combined with the unit mileage energy consumption under the vehicle's current load (e.g., 1.5 kWh / km), the detour energy cost is calculated (20 × 1.5 × electricity price). Then, the service fee of the battery swap station (e.g., 80 yuan per battery swap) and the time increase cost caused by the detour (e.g., 0.5 hours delay × unit time operating cost) are added. The sum is the opportunistic battery swap cost.

[0075] When calculating the necessary battery swap cost, the system first determines the location where the battery level drops to a preset threshold based on the vehicle's current battery level and the predicted power consumption of each segment of the initial route (e.g., high power consumption on uphill sections and low power consumption on flat sections). Then, it searches for the nearest battery swap station around that location and calculates the emergency detour mileage to that station and the corresponding power cost. At the same time, it obtains the real-time queuing data of the battery swap station (e.g., a 2-hour wait) and calculates the waiting time cost. If the battery swap causes delays in subsequent tasks, a delay penalty cost (e.g., a fine of 200 yuan for every hour of delay) is also added. The final sum is the necessary battery swap cost.

[0076] Optionally, in one specific embodiment, the opportunistic battery swapping cost is calculated as follows: The system obtains the round-trip detour mileage of the vehicle from the initial path deviation point to the battery swapping station (calculated using GIS tools); based on the vehicle's current cargo load (e.g., 10 tons), it queries historical power consumption data to determine the power consumption per unit mileage (e.g., 1.8 kWh / km), and calculates the detour power cost (detour mileage × unit power consumption × electricity price); it obtains the real-time service price of the battery swapping station (e.g., 90 yuan per trip), and adds the time cost caused by the detour (detour time × driver's hourly wage), the sum of which is the opportunistic battery swapping cost.

[0077] Calculate the necessary battery swapping cost: Based on the slope and length of each segment of the initial path, the system predicts the location where the vehicle's battery level will drop to a preset threshold (15%); it searches for battery swapping stations within 50 kilometers of this location, selects the closest one (e.g., 30 kilometers away), and calculates the detour cost (30 × 1.8 × electricity price per unit); it obtains the real-time queuing time of the battery swapping station (e.g., 3 hours) and calculates the waiting cost (3 × operating cost per unit time). If subsequent tasks are delayed, a delay penalty cost is added (delay of 1 hour × 200 yuan). The sum is the necessary battery swapping cost.

[0078] It is understandable that other methods can be used to calculate battery swapping costs, such as introducing machine learning models to predict future cost fluctuations based on historical battery swapping data; this is not a limitation here.

[0079] S104. When the necessary battery swapping cost is greater than the opportunistic battery swapping cost, add the opportunistic battery swapping node corresponding to the opportunistic battery swapping cost to the initial connecting route to generate a connecting loop transportation plan.

[0080] The intermodal transport plan refers to a complete transport scheme that includes transport task nodes and battery swapping nodes. It covers vehicle routes, arrival / departure times of each node, battery swapping reservation information, etc., to ensure that vehicles can make reasonable energy replenishment while completing multiple continuous transport tasks.

[0081] Specifically, the system compares the necessary battery swap cost with the opportunistic battery swap cost—for example, if the necessary battery swap cost is 500 yuan and the opportunistic battery swap cost is 350 yuan, then the necessary battery swap cost is higher. Subsequently, the system determines the insertion position of the opportunistic battery swap node in the initial connecting route:

[0082] Prioritize locations that do not impact the time window of the main tasks (such as between two transportation task nodes, where there is sufficient time in that interval). For example, if the initial path is "Node A → Node B → Node C", insert the battery swapping node between "Node A → Node B" to form a route of "Node A → Battery Swapping Node → Node B → Node C". Finally, the system integrates the arrival and departure times of each node (e.g., loading goods at Node A takes 1 hour, and swapping batteries at the battery swapping node takes 0.5 hours) to generate a connecting loop transportation plan that includes the battery swapping schedule, ensuring that the plan meets the time requirements of all tasks.

[0083] Furthermore, in other embodiments, after the system adds the opportunistic battery swapping nodes corresponding to the opportunistic battery swapping costs to the initial connecting route and generates a connecting loop transportation plan, the system can also send reservation requests to the battery swapping stations corresponding to the opportunistic battery swapping nodes in the connecting loop transportation plan to obtain battery swapping time window confirmation information; adjust the dwell time of the vehicle at each transportation task node according to the battery swapping time window confirmation information to generate an updated transportation plan containing battery swapping time constraints; push the updated transportation plan to the vehicle terminal, and send battery swapping reminder information when the vehicle approaches the opportunistic battery swapping node.

[0084] In the above embodiments, when planning intermodal transport routes, the system no longer passively waits for vehicle battery levels to drop to a low threshold before searching for the nearest battery swapping station. Instead, it proactively identifies opportunistic battery swapping nodes near the initial route. By estimating and comparing the economic impact of two strategies—"the cost of opportunistic battery swapping now" and "the cost of necessary battery swapping when the battery is depleted in the future"—the system can make more cost-effective decisions. When the system anticipates that the cost of necessary battery swapping in the future will be higher, it includes lower-cost opportunistic battery swapping nodes in the transport plan in advance. This proactive and predictive energy management approach effectively avoids the high time and mileage costs incurred by forced detours due to battery depletion, transforming emergency rigid battery swapping needs into planned flexible battery swapping options. This reduces the overall operating cost of the entire intermodal transport cycle and improves the intelligence level of vehicle scheduling and overall transport efficiency.

[0085] The following provides a more detailed description of the process of the method provided in this implementation. Please refer to [link / reference]. Figure 2 This is another flowchart illustrating a vehicle scheduling method based on intermodal circular transportation in an embodiment of this application.

[0086] S201. Construct a spatiotemporal network graph that includes existing transportation task nodes and new transportation task nodes.

[0087] Specifically, the vehicle dispatching system retrieves the core information of all existing transportation task nodes from the database, including the geographical coordinates of each node (such as latitude and longitude), task type (loading / unloading), the determined time window (e.g., the time window for the loading node at the Suzhou factory is "8:00-10:00 the next day"), and cargo loading and unloading requirements (e.g., the Suzhou factory needs to use a 3-ton forklift for loading and unloading). At the same time, it extracts the information of new transportation task nodes from new transportation orders, and supplements the location of the node, task type (e.g., the Shanghai warehouse is the loading node and the Guangzhou distribution center is the unloading node), time window (e.g., loading at the Shanghai warehouse needs to be completed "12:00-14:00 the next day"), and loading and unloading requirements (e.g., electronic products need to be handled with care and small loading and unloading equipment should be used).

[0088] Next, the system constructs the basic framework of the spatiotemporal network graph with nodes as the core:

[0089] Each transportation task node is transformed into a node element in the graph. Information such as task type, time window constraints, and cargo loading and unloading requirements are entered into the node attributes. Then, the road network connectivity between any two nodes is analyzed. For example, it is determined whether there is a direct freight highway from the Suzhou factory node to the Shanghai warehouse node, whether the road allows vehicles to pass at the current time, and whether there are temporary construction sections affecting traffic efficiency. If they are connected, an "edge" is established between the two nodes, and attributes such as road length, estimated travel time, and road restrictions (such as speed limit of 60km / h) are marked on the edge.

[0090] Optionally, the system can call the Geographic Information System (GIS) interface to obtain the geographic location data (latitude and longitude, detailed address) of existing and new transportation task nodes, and automatically match the road network information around each node (such as nearby highway entrances and freight channels); extract the task type, time window constraints (such as "unloading time window of Guangzhou distribution center is 10:00-16:00 within 3 days" from the order management module and historical task database), and cargo loading and unloading requirements (such as "the commonly used loading and unloading equipment in Shanghai warehouse is a 2-ton electric forklift, and the single loading and unloading takes about 30 minutes" from historical data) of each node, and enter this information into the node attribute library; use road network analysis tools (such as the path planning interface of Baidu Maps Open Platform) to verify the road network connectivity between any two nodes one by one. For example, when verifying the connectivity from Suzhou factory to Shanghai warehouse, input the coordinates of two nodes, obtain the recommended freight route, determine whether the route is feasible (if there is no road closure information, it is determined to be connected), and associate information such as route length and estimated travel time with the "edges" between nodes, and finally generate a spatiotemporal network graph.

[0091] S202. Based on the vehicle's current location, remaining battery power, cargo status, and assigned task sequence, calculate the multi-dimensional cost function for the vehicle to complete a new transportation task.

[0092] Specifically, the vehicle dispatching system collects real-time vehicle status data: it obtains the vehicle's current location and remaining battery power through the vehicle terminal; it retrieves the cargo status from the vehicle management module; and it obtains the assigned task sequence (such as "loading at Suzhou factory (May 20, 8:00-10:00) → unloading at Shanghai warehouse (May 20, 12:00-14:00)") from the task allocation module, thus clarifying the connection order between the new transportation task ("loading at Shanghai warehouse → unloading at Guangzhou distribution center") and the existing tasks.

[0093] Next, the system defines the cost calculation rules for each item in the multi-dimensional cost function, including but not limited to travel distance cost, time cost, electricity consumption cost, and task delay penalty cost. Optionally, the system extracts the total mileage of the vehicle's travel path from the spatiotemporal network graph (e.g., the total mileage of the Suzhou→Shanghai→Guangzhou path is 1400km), combines it with the company's preset unit mileage cost (e.g., the unit mileage cost for freight vehicles is 2.5 yuan / km), and calculates it according to "total mileage × unit mileage cost" (1400×2.5=3500 yuan); the time cost is divided into travel time cost and loading / unloading time cost. The travel time is obtained from the "edge" attribute of the spatiotemporal network graph (e.g., the estimated travel time from Suzhou→Shanghai is 2 hours, the estimated travel time from Shanghai→Guangzhou is 12 hours, and the total travel time is 14 hours), and is calculated according to the unit time operating cost. (For example, a driver's hourly wage of 40 yuan + vehicle depreciation of 20 yuan / hour, totaling 60 yuan / hour) calculates the driving time cost (14 × 60 = 840 yuan); loading and unloading time is determined based on the cargo loading and unloading requirements of each node (e.g., loading at the Suzhou factory takes 1 hour, unloading at the Shanghai warehouse takes 0.5 hours, loading takes 1 hour, unloading at the Guangzhou distribution center takes 0.5 hours, totaling 3 hours). The loading and unloading time cost is calculated based on the unit loading and unloading time cost (e.g., a loading and unloading worker's hourly wage of 30 yuan) (3 × 30 = 90 yuan). The total time cost is 840 + 90 = 930 yuan; for electricity consumption costs, it can be calculated based on the vehicle... The power consumption characteristics of each route segment are calculated based on the loading / unloading status (empty vehicle → loaded vehicle → unloading → loaded vehicle → unloading) and the load conditions (e.g., 0.8 kWh / km for empty vehicles and 1.2 kWh / km for loaded vehicles on highways, and 0.7 kWh / km for empty vehicles and 1.1 kWh / km for loaded vehicles on national highways). For example, the power consumption for each segment is calculated as follows: Suzhou to Shanghai is a 100km highway empty vehicle route with a power consumption of 100 × 0.8 = 80 kWh; Shanghai to Guangzhou is a 1300km highway loaded vehicle route with a power consumption of 1300 × 1.2 = 1560 kWh, for a total power consumption of 1640 kWh. The unit electricity cost (e.g., 1.2 yuan / kWh for industrial electricity) is used to calculate the electricity consumption cost (1640 × 1.2 = 1968 yuan). Alternatively, based on the time window constraints of the nodes in the spatiotemporal network graph, the arrival time of the vehicle at each node can be predicted (e.g., it is expected to arrive at the Suzhou factory for loading at 8:30 on May 20, within the time window; it is expected to arrive at the Shanghai warehouse for unloading at 12:30 on May 20, exceeding the time window (12:00-14:00) by 0.5 hours), and the delay penalty cost is calculated according to the preset unit delay penalty amount (e.g., 100 yuan / hour) (0.5 × 100 = 50 yuan).

[0094] Finally, the system adds up all costs to construct a multi-dimensional cost function, namely "Comprehensive Cost = Travel Distance Cost + Time Cost + Electricity Consumption Cost + Task Delay Penalty Cost". Substituting the above data, we can get the comprehensive cost = 3500 + 930 + 1968 + 50 = 6448 yuan. The comprehensive cost of different paths can be calculated through this function in the future.

[0095] S203. Use dynamic programming algorithm to search for a set of feasible paths in the spatiotemporal network graph that satisfy the time window constraint and vehicle power constraint, and calculate the comprehensive cost value of each feasible path according to the multi-dimensional cost function.

[0096] Specifically, the system first determines the core parameters and boundary conditions of the dynamic programming algorithm: the transportation task nodes are the stages of the algorithm (e.g., the first stage is the loading node at the Suzhou factory, the second stage is the unloading / loading node at the Shanghai warehouse, and the third stage is the unloading node at the Guangzhou distribution center), the vehicle's arrival time and remaining battery power are the states, and the cost and state changes from the current node to the next node are the decisions. Next, starting from the initial node (e.g., the Suzhou factory where the vehicle is currently located), the system traverses all nodes in the spatiotemporal network graph stage by stage, determining whether the route between each node satisfies the constraints: for time window constraints, the system calculates the estimated arrival time of the vehicle from the current node to the next node. If this time falls within the time window of the next node (e.g., expected arrival at the Shanghai warehouse at 13:00, within the 12:00-14:00 window), the time constraint is satisfied; for vehicle battery power constraints, the system calculates the battery consumption during the journey based on the route mileage and the vehicle's cargo status. If the remaining battery power is still higher than the safety threshold when reaching the next node (e.g., 20% remaining after consuming 100kWh, higher than the 15% threshold), the battery power constraint is satisfied.

[0097] Routes that meet the constraints will be included in the set of feasible paths. Simultaneously, the system calls a multi-dimensional cost function to calculate the route's distance cost (e.g., route mileage × unit mileage cost), time cost (travel time × unit operating cost + loading / unloading time × unit loading / unloading cost), electricity consumption cost (driving electricity consumption × unit electricity cost), and task delay penalty cost (calculated if a time window expires, otherwise zero). These costs are then summed to obtain the comprehensive cost value of the feasible path. For example, a feasible path "Suzhou Factory → Shanghai Warehouse → Guangzhou Distribution Center" has a comprehensive cost value of 6448 yuan calculated using the function. The system will record this path and its corresponding cost value, ultimately forming a set containing multiple feasible paths and their respective comprehensive costs.

[0098] S204. Select the path with the lowest overall cost from the set of feasible paths as the initial connecting path.

[0099] Specifically, the system organizes all paths in the feasible path set and their corresponding comprehensive cost values ​​to form a "path-comprehensive cost" lookup table. For example, the feasible path set contains three paths: Path 1 "Suzhou Factory → Shanghai Warehouse → Guangzhou Distribution Center" (comprehensive cost 6448 yuan), Path 2 "Suzhou Factory → Hangzhou Transit Warehouse → Guangzhou Distribution Center" (comprehensive cost 6800 yuan), and Path 3 "Suzhou Factory → Nanjing Logistics Park → Guangzhou Distribution Center" (comprehensive cost 7200 yuan). Next, the system sorts the comprehensive cost values ​​in the lookup table numerically, from largest to smallest or smallest to largest, and identifies the path with the lowest cost value. In the example above, Path 1's cost of 6448 yuan is the lowest. The system further verifies the completeness and feasibility of this path: confirming that the path includes all necessary transportation task nodes (existing and new nodes), that the node connection order conforms to transportation logic (e.g., loading before unloading, reasonable connection according to geographical order), and that there are no omitted constraints (e.g., whether the time window and power constraints are still met). If the verification passes, the path will be determined as the initial connecting path. The system will record detailed information such as the path's node sequence, the estimated arrival / departure time of each node, and the mileage and power consumption prediction of each segment, providing basic data for subsequent steps.

[0100] S205. Obtain the actual power consumption record from the vehicle's historical transportation data, and calculate the average power consumption rate under different road sections and different load conditions.

[0101] Specifically, the system first filters historical transportation data of the target vehicle from the database: based on the vehicle's unique identifier (such as license plate number, VIN code), it extracts all transportation task records of the vehicle within the past 6 months. Each record must include the coordinates of the start / end point of the route, the route type (such as highway, national highway), the mileage, the weight of the transported goods (used to determine load conditions), the battery level at departure, and the battery level at arrival (the difference between the two is used to calculate the actual power consumption). Next, the system classifies the historical data: it divides the data into categories such as "highway", "national highway", and "mountain road" according to the route type, and into three load conditions according to the weight of the goods: "empty vehicle" (0 tons), "lightly loaded" (1-5 tons, assuming a rated load of 10 tons), and "heavily loaded" (6-10 tons), forming a two-dimensional classification matrix of "route type-load condition".

[0102] Then, the system calculates the average power consumption rate for each category: For the "Highway - Heavy Load" category, it summarizes the actual power consumption and corresponding mileage of all historical records under this category, calculates the ratio of total power consumption to total mileage, and obtains the average power consumption rate for this category. For example, if there are 3 records under this category: Record 1 (mileage 100km, power consumption 120kWh), Record 2 (mileage 150km, power consumption 180kWh), and Record 3 (mileage 200km, power consumption 240kWh), the total power consumption is 540kWh, the total mileage is 450km, and the average power consumption rate is 540÷450=1.2kWh / km. The system repeats this calculation process for all "Road Segment Type - Load Condition" categories to form an average power consumption rate comparison table, while removing abnormal data (such as abnormally high power consumption records caused by vehicle malfunctions) to ensure the accuracy of the calculation results.

[0103] S206. Estimate the vehicle's power consumption on each road segment based on the average power consumption rate and the road segment attributes of the initial connecting route, and generate a power consumption change curve of the vehicle during its journey along the initial connecting route.

[0104] Specifically, the system first divides the initial connecting route into multiple continuous segments: based on the geographical characteristics of the route (such as road type change points, city boundary points) or fixed mileage intervals (such as every 50 kilometers), the initial connecting route is split into several segments, such as the "Suzhou factory → Shanghai warehouse" segment (highway, 100km) and the "Shanghai warehouse → Guangzhou distribution center" segment (highway, 1300km), and records the attributes of each segment (segment type: highway, mileage: 100km / 1300km, load conditions: Suzhou → Shanghai is empty, Shanghai → Guangzhou is heavily loaded). Next, the system queries the average power consumption rate lookup table generated in step S205 to match the average power consumption rate corresponding to "segment type - load conditions" for each segment. For example, "highway - empty" corresponds to an average power consumption rate of 0.8kWh / km, and "highway - heavily loaded" corresponds to 1.2kWh / km.

[0105] Then, the system calculates the estimated power consumption for each road segment: Estimated power consumption = segment mileage × corresponding average power consumption rate, such as the Suzhou → Shanghai segment (100km × 0.8kWh / km = 80kWh), and the Shanghai → Guangzhou segment (1300km × 1.2kWh / km = 1560kWh). Subsequently, starting from the vehicle's initial power level (e.g., a full charge of 400kWh), the system calculates the remaining power at the end of each segment according to the travel sequence: Remaining power at the end of the Suzhou → Shanghai segment = 400kWh - 80kWh = 320kWh. After loading goods at the Shanghai warehouse (assuming no power consumption), the remaining power decreases according to the power consumption rate for each mile traveled on the Shanghai → Guangzhou segment (e.g., if 120kWh is consumed while traveling 100km, the remaining power decreases from 320kWh to 200kWh). Finally, the system uses "mileage" as the horizontal axis (from 0km to 1400km) and "remaining battery power" as the vertical axis (from 400kWh to 0kWh) to convert the power consumption data of each road segment into a continuous curve, namely the power change curve. Each point on the curve represents the vehicle's remaining battery power at the corresponding mileage.

[0106] S207. Based on the battery change curve, road sections where the vehicle's remaining battery power is within a preset range are selected as potential battery swapping areas for subsequent opportunistic battery swapping node identification.

[0107] Specifically, the system first retrieves the generated battery level change curve and clarifies the correspondence between the horizontal axis (mileage) and the vertical axis (remaining battery power). For example, the curve shows that when the vehicle travels 0-50km, the remaining battery power decreases from 400kWh to 360kWh; when traveling 50-150km, the remaining battery power decreases from 360kWh to 240kWh; and when traveling 150-300km, the remaining battery power decreases from 240kWh to 120kWh, and so on. Next, the system marks the upper and lower limits of a preset range on the vertical axis of the battery level change curve (e.g., 120kWh-280kWh corresponding to 30%-70%). By the intersection of the curve and the boundary of the range, the system determines the mileage range within which the remaining battery power falls within the preset range: for example, the mileage at which the remaining battery power first enters 280kWh (upper limit) is 80km, and the mileage at which it first falls below 120kWh (lower limit) is 300km, then within the corresponding mileage range of 80-300km, the vehicle's remaining battery power always remains between 120kWh and 280kWh. Finally, the system marks the initial connecting route segments corresponding to this mileage range as potential battery swapping areas, such as the "Shanghai suburban expressway section - Hangzhou ring expressway section" corresponding to 80-300km. Subsequent searches will only be conducted within this area to avoid inefficient waste caused by indiscriminate searching of the entire route.

[0108] S208. Obtain the opportunistic battery swapping costs of multiple opportunistic battery swapping nodes within the same potential battery swapping area and sort them to generate a priority list of battery swapping nodes.

[0109] Specifically, the system first retrieves information on all identified opportunistic battery swapping nodes within the potential battery swapping area from the battery swapping station network database, including the location coordinates of each node, the battery swapping service price, and the number of vehicles currently in the queue. Next, it calculates the opportunistic battery swapping cost for each node: the cost consists of detour mileage cost (total detour distance from the initial path deviation point to the battery swapping station and back × unit mileage operating cost, e.g., 10km detour × 2.5 yuan / km = 25 yuan), battery swapping service cost (single service price at the battery swapping station, e.g., 80 yuan), and time cost (detour time + battery swapping time + queuing time × unit time cost, e.g., detour time 0.5 hours + battery swapping time 0.3 hours + queuing time 0.2 hours = 1 hour × 60 yuan / hour = 60 yuan). These three costs are added together to obtain the opportunistic battery swapping cost for each node (e.g., 25 + 80 + 60 = 165 yuan).

[0110] After the calculation is completed, the system sorts all nodes from low to high according to the opportunistic battery swapping cost, for example, node A (150 yuan), node B (165 yuan), node C (180 yuan), and node D (200 yuan). Each node is marked with a priority (node ​​A is level 1, node B is level 2, node C is level 3, and node D is level 4). Finally, a priority list of battery swapping nodes is generated. The list will also supplement other key information of the nodes (such as the number of available batteries and the operating status of the battery swapping station) to ensure that subsequent screening not only takes cost into account, but also node availability.

[0111] S209. Select the N nodes with the lowest cost from the priority list of battery swapping nodes as candidate battery swapping nodes, and calculate the success probability of battery swapping execution for the candidate battery swapping nodes.

[0112] Specifically, the system first reads the preset N value (e.g., N=3), extracts the three nodes with the lowest cost from the priority list of battery swapping nodes, such as node A (150 yuan, level 1), node B (165 yuan, level 2), and node C (180 yuan, level 3), and determines these three nodes as candidate battery swapping nodes. Next, the system calculates the success probability of battery swapping for each candidate node: real-time dynamic data of each node is obtained through the battery swapping station network platform, including the number of available batteries (e.g., node A has 5 available batteries), the number of vehicles currently in the queue (e.g., 2 vehicles), the historical failure rate of the equipment (e.g., a failure rate of 2% in the last 30 days), and the estimated arrival time of the vehicles (e.g., 1 hour later). Based on this data, a probability calculation model is constructed. For example, the probability that the number of available batteries is greater than or equal to the number of vehicles in the queue + 1 (ensuring that there are batteries when the vehicles arrive) is 80%, and the probability that the equipment is fault-free is 98%. The product of the two is the preliminary success probability (80% × 98% = 78.4%). Combined with the traffic flow prediction within the estimated arrival time (e.g., the number of vehicles in the queue may increase by 1 after 1 hour, and the probability of adjusting the available batteries to meet this requirement is 70%), the final success probability of battery swapping for node A is 68.6%. Similarly, the success probabilities of node B (60%) and node C (75%) are calculated to provide reliability data for subsequent comprehensive evaluation.

[0113] S210. The product of the success probability and the opportunistic battery swapping cost is used as a comprehensive evaluation index, and the node with the best comprehensive evaluation index is selected as the recommended battery swapping node.

[0114] Specifically, the system first retrieves the opportunistic battery swapping cost and success probability data for all candidate battery swapping nodes, such as node A (cost 150 yuan, success rate 68.6%), node B (cost 165 yuan, success rate 60%), and node C (cost 180 yuan, success rate 75%). Then, it calculates the index value for each node using the formula "comprehensive evaluation index = success probability × opportunistic battery swapping cost": node A = 0.686 × 150 ≈ 102.9, node B = 0.6 × 165 = 99, and node C = 0.75 × 180 = 135. The system then compares the comprehensive evaluation index values ​​of the three nodes and finds that node B has the smallest index value (99), meaning it has the lowest cost per unit success probability (only 1.65 yuan per 1% success rate, lower than node A's 2.19 yuan and node C's 2.4 yuan). Therefore, node B is determined to be the node with the optimal comprehensive evaluation index. Finally, the system marks node B as a recommended battery swapping node and records its detailed information (location, estimated swapping time, service price) to provide a basis for subsequently including it in the transportation plan and scheduling battery swapping services.

[0115] S211. Based on the vehicle's current battery status and the predicted battery consumption curve, determine the warning location point that triggers the necessary battery swap when the vehicle travels along the initial connecting route.

[0116] Specifically, the system obtains the vehicle's current battery status, for example, a real-time remaining battery of 380 kWh (out of a total of 400 kWh, representing 95%), and retrieves the predicted battery consumption curve generated in step S206. This curve marks the change in remaining battery power at each mileage segment along the initial connecting route (e.g., "Suzhou factory → Shanghai warehouse → Guangzhou distribution center"), such as 300 kWh remaining after traveling 100 km, or 200 kWh remaining after traveling 300 km. Next, the system determines the specific value of the "sum of the safety threshold and the emergency reserve battery power": assuming the safety threshold is 15% of the total battery power (60 kWh, ensuring the vehicle can handle unexpected road conditions), and the emergency reserve battery power is 5% of the total battery power (20 kWh, for emergency driving in extreme situations), the sum of which is 80 kWh.

[0117] Subsequently, the system searches the power consumption curve for the mileage corresponding to the first drop in remaining battery power to 80kWh. For example, if the curve shows that the remaining battery power is exactly 80kWh after traveling 600km, the system then uses the mileage-location mapping relationship of the initial connecting route (e.g., the 600km mark corresponds to "near the Jinhua service area on the Hangzhou-Guangzhou expressway") to determine the specific geographical coordinates corresponding to that mileage. This coordinate point is the warning location that triggers the necessary battery swap. The system records detailed information about this location (such as latitude and longitude, and surrounding road network conditions).

[0118] S212. Determine the dynamic service capacity data of all battery swapping stations within a preset range around the warning location point based on the battery swapping station network data.

[0119] Specifically, the system determines the coordinates of the warning location and a preset search range (e.g., 50km). It then sends a query request to the battery swapping station management platform via the station's network data interface. The request includes a list of all battery swapping stations in normal operation within a preset radius centered on the warning location, along with their dynamic service capacity data. Upon receiving the request, the management platform filters eligible stations and returns dynamic service capacity data for each station, including but not limited to the number of available batteries, queuing time, battery swapping service efficiency, and real-time road conditions for reaching the station.

[0120] S213. Calculate the estimated cost of deviation from the warning location to each candidate battery swapping station.

[0121] Specifically, the system first filters candidate battery swapping stations from those surrounding the warning location: stations that are faulty or have no usable batteries are excluded, while those that are operating normally and have at least one usable battery are retained. Next, for each candidate station, the system calculates the four components of the estimated cost:

[0122] Detour distance cost: Calculate the vehicle's travel distance from the warning location to the battery swapping station (e.g., 8km at Jinhua battery swapping station) and the distance from the battery swapping station back to the shortest point on the initial path (8km) using GIS tools. The total detour distance is 16km. Calculate the detour distance cost at a unit mileage operating cost of 2.5 yuan / km. The detour distance cost = 16 × 2.5 = 40 yuan.

[0123] Queuing time cost: Based on the queuing time of the candidate battery swapping station (e.g., 40 minutes at Jinhua battery swapping station), and calculated at a unit time cost of 1 yuan / minute (including driver working hours and vehicle idle cost), the queuing time cost = 40 × 1 = 40 yuan;

[0124] Battery swapping service time cost: Based on the battery swapping service efficiency of the battery swapping station (e.g., 40 minutes for battery swapping at the Jinhua battery swapping station), and calculated using the same unit time cost, the battery swapping service time cost = 40 × 1 = 40 yuan;

[0125] Cascading costs of subsequent task delays: Calculate the total time required for necessary battery swapping (detour time + queuing time + battery swapping time). For example, the total time for the Jinhua battery swapping station = (16km ÷ 60km / h × 60 minutes) + 40 + 40 = 16 + 40 + 40 = 96 minutes (1.6 hours). If this delay causes subsequent transportation tasks (such as unloading at the Guangzhou distribution center) to exceed the time limit by 2 hours (because there is still 3 hours of travel time on the subsequent route, the total delay of 1.6 hours plus road condition fluctuations will eventually exceed the time limit by 2 hours), calculated according to the preset task overtime penalty standard of 200 yuan / hour, the cascading cost of subsequent task delays = 2 × 200 = 400 yuan.

[0126] Finally, the four costs are added together to obtain the estimated cost of each candidate battery swapping station, forming a "Candidate Battery Swapping Station - Estimated Cost" comparison table.

[0127] S214. The Monte Carlo simulation method is used to evaluate the necessary battery swapping risk cost under different traffic conditions and battery swapping station load conditions, and the risk cost is weighted and summed with the estimated cost to obtain the comprehensive necessary battery swapping cost.

[0128] Specifically, the system determines the input parameters and probability distributions for the Monte Carlo simulation: For traffic conditions, based on historical data, probability distributions are set for "smooth traffic" (40% probability, travel time reduced by 10% from the estimated time), "slow traffic" (50% probability, travel time unchanged), and "congestion" (10% probability, travel time increased by 50%). For the battery swapping station load, probability distributions are set for "reduced load" (20% probability, queuing time reduced by 30%), "normal load" (60% probability, queuing time unchanged), and "increased load" (20% probability, queuing time increased by 50%). Next, the system generates 10,000 random scenarios through Monte Carlo simulation: in each scenario, the traffic conditions and the battery swapping station load are randomly selected, and the difference between the actual cost and the estimated cost (i.e., risk cost) under that scenario is calculated. For example, in one scenario, "congestion" (travel time increases by 50% → 22.5 minutes, an extra 7.5 minutes → an increase in time cost of 7.5 yuan) and "increased load" (queue time increases by 50% → 0 minutes, still 0, no additional cost) are selected, and the risk cost is 7.5 yuan; in another scenario, "smooth traffic" (travel time decreases by 10% → 13.5 minutes, saving 1.5 minutes → risk cost -1.5 yuan) and "reduced load" (no additional cost) are selected, and the risk cost is -1.5 yuan.

[0129] After the simulation is completed, the system calculates the average risk cost of 10,000 scenarios (e.g., an average of 15 yuan represents an average additional risk cost of 15 yuan), and sets weights (estimated cost weight 0.7, risk cost weight 0.3). The comprehensive necessary battery swap cost is calculated as: Estimated cost × 0.7 + Average risk cost × 0.3 = 307.5 × 0.7 + 15 × 0.3 = 215.25 + 4.5 = 219.75 yuan. If there are multiple candidate battery swap stations, the system will calculate the comprehensive necessary battery swap cost for each station separately and select the minimum value (e.g., 219.75 yuan for Yongkang battery swap station is lower than 280 yuan for Jinhua battery swap station and 490 yuan for Yiwu battery swap station) as the final comprehensive necessary battery swap cost for subsequent comparison with opportunistic battery swap costs.

[0130] Furthermore, after weighting and summing the risk cost with the estimated cost to obtain the comprehensive necessary battery swapping cost, the system can also extract actual necessary battery swapping cases for similar road sections and similar time periods from historical data, and calculate the average detour distance and average waiting time of historical cases; compare the average detour distance and average waiting time with the currently calculated necessary battery swapping cost to generate a cost deviation rate; when the cost deviation rate exceeds a preset threshold, identify the main factors causing the deviation, including but not limited to changes in traffic congestion, changes in the service capacity of battery swapping stations, or errors in power prediction; then correct the calculation parameters of the necessary battery swapping cost based on the main factors, recalculate the corrected necessary battery swapping cost, and use the corrected necessary battery swapping cost for comparison with the opportunistic battery swapping cost.

[0131] In one specific embodiment, taking an electric heavy-duty truck with 85% remaining battery power performing a connecting transport mission from Suzhou to Hangzhou to Ningbo as an example, when traveling to the Jiaxing section, the method of this invention actively identifies an opportunistic battery swapping station near the route. The system calculates that if the vehicle goes to this station for opportunistic battery swapping at this time, the total opportunistic battery swapping cost, including detour, time, and service fees, is 126.52 yuan. At the same time, the system evaluates through a prediction model (steps S211-S214) that if this opportunity is not taken, the vehicle will trigger a low battery warning when approaching Ningbo. At that time, in order to reach the nearest battery swapping station, the comprehensive necessary battery swapping cost, including significant detour, long queues, mission delay penalties, and uncertainty risks, will be incurred, estimated to be as high as 542.5 yuan.

[0132] Since the predicted necessary battery swapping cost is far higher than the current opportunistic battery swapping cost, the system immediately makes a decision to add the opportunistic battery swapping station to the transportation route and automatically completes the service reservation and information distribution. Through this proactive battery swapping decision based on cost-benefit prediction, this method successfully transforms a high-risk, high-cost emergency energy replenishment into a planned flexible energy replenishment, not only avoiding the risk of mission delays but also directly saving 415.98 yuan in comprehensive operating costs for this transportation mission, significantly demonstrating the technical effectiveness of this invention in improving the economic efficiency and intelligence level of transportation.

[0133] It should be noted that the method for selecting battery swapping stations described above in this application is also applicable to selecting charging stations, and will not be elaborated here.

[0134] The vehicle dispatching system of this invention is applied to electronic devices. Figure 3 A schematic diagram of the architecture of an electronic device suitable for implementing embodiments of the present invention is shown.

[0135] It should be noted that, Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0136] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by instructions (computer programs), or by instructions (computer programs) controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor. The electronic device of this embodiment includes a storage medium and a processor, wherein the storage medium stores multiple instructions that can be loaded by the processor to execute any step of the method provided in the embodiments of the present invention.

[0137] Specifically, the storage medium and the processor are electrically connected directly or indirectly to enable data transmission or interaction. For example, these components can be electrically connected to each other via one or more signal lines. The storage medium stores computer-executable instructions that implement data access control methods, including at least one software functional module that can be stored in the storage medium in the form of software or firmware. The processor executes various functional applications and data processing by running the software program and module stored in the storage medium. The storage medium can be, but is not limited to, Random Access Memory (RAM), Read-Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The storage medium stores the program, and the processor executes the program after receiving the execution instructions.

[0138] Furthermore, the software programs and modules within the aforementioned storage medium may also include an operating system, which may include various software components and / or drivers for managing system tasks (e.g., memory management, storage device control, power management, etc.) and can communicate with various hardware or software components to provide an operating environment for other software components. The processor may be an integrated circuit chip with signal processing capabilities. The aforementioned processor may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc., which can implement or execute the methods, steps, and logic flowcharts disclosed in this embodiment. The general-purpose processor may be a microprocessor or any conventional processor.

[0139] Since the instructions stored in the storage medium can execute the steps in any of the methods provided in the embodiments of the present invention, the beneficial effects of any of the methods provided in the embodiments of the present invention can be achieved, as detailed in the preceding embodiments, and will not be repeated here.

[0140] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A vehicle scheduling method based on intermodal circular transportation, applied to a vehicle scheduling system, characterized in that, The method includes: Obtain the new transportation task node for the new transportation order, and plan an initial connecting path between the existing transportation task node and the new transportation task node based on the preset transportation cost index. Obtain the actual power consumption records from the vehicle's historical transportation data, and calculate the average power consumption rate under different road sections and different load conditions; The vehicle's power consumption in each road segment is estimated based on the average power consumption rate and the road segment attributes of the initial connecting route. The power consumption values ​​of each road segment are summed up to determine the power change curve of the vehicle during its journey along the initial connecting route. Based on the battery change curve, road sections where the vehicle's remaining battery power is within a preset range are selected as potential battery swapping areas for subsequent opportunistic battery swapping node identification. Based on real-time acquired battery swapping station network data, opportunistic battery swapping nodes that can be swapped in the initial connecting route are identified, and the shortest travel distance between the opportunistic battery swapping node and the initial connecting route is less than a preset distance threshold. The calculation of the opportunistic battery swapping cost incurred when a vehicle deviates from the initial connecting route to complete a battery swap at the opportunistic battery swapping node and returns to the original route, and the necessary battery swapping cost incurred when a vehicle abandons the opportunistic battery swapping node and continues traveling along the initial connecting route until the battery level drops to a preset threshold, to reach the nearest battery swapping station; the step of calculating the necessary battery swapping cost incurred when a vehicle abandons the opportunistic battery swapping node and continues traveling along the initial connecting route until the battery level drops to a preset threshold to reach the nearest battery swapping station includes: Based on the vehicle's current battery status and the predicted battery consumption curve, a warning location is determined when the vehicle travels along the initial connecting route to trigger a necessary battery swap. The warning location is the position where the vehicle's remaining battery power drops to the sum of a safety threshold and an emergency reserve battery power. Based on the battery swapping station network data, determine the dynamic service capability data of all battery swapping stations within a preset range around the warning location point. The dynamic service capability data includes the number of available batteries at the battery swapping station, queuing time, battery swapping service efficiency, and real-time road condition information for reaching the battery swapping station. Calculate the estimated cost of deviation from the warning location to each candidate battery swapping station. The estimated cost includes detour distance cost, queuing time cost, battery swapping service time cost, and cascading cost of subsequent task delays caused by emergency battery swapping. The Monte Carlo simulation method is used to evaluate the necessary battery swapping risk cost under different traffic conditions and battery swapping station load conditions, and the necessary battery swapping risk cost is weighted and summed with the estimated cost to obtain the comprehensive necessary battery swapping cost; When the necessary battery swap cost is greater than the opportunistic battery swap cost, the opportunistic battery swap node corresponding to the opportunistic battery swap cost is added to the initial connecting route to generate a connecting loop transportation plan.

2. The method according to claim 1, characterized in that, The step of planning an initial connecting route between existing transportation task nodes and the new transportation task node based on a preset transportation cost index specifically includes: Construct a spatiotemporal network graph that includes existing transportation task nodes and new transportation task nodes. The node attributes of the spatiotemporal network graph include task type, time window constraints, cargo loading and unloading requirements, and road network connectivity between nodes. Based on the vehicle's current location, remaining battery power, cargo status, and assigned task sequence, a multi-dimensional cost function for the vehicle to complete a new transportation task is calculated. The multi-dimensional cost function includes travel distance cost, time cost, battery consumption cost, and task delay penalty cost. The dynamic programming algorithm is used to search for a set of feasible paths in the spatiotemporal network graph that satisfy the time window constraint and the vehicle battery power constraint, and the comprehensive cost value of each feasible path is calculated according to the multi-dimensional cost function. Select the path with the lowest overall cost from the set of feasible paths as the initial connecting path.

3. The method according to claim 1, characterized in that, After the step of adding the opportunistic battery swapping node corresponding to the opportunistic battery swapping cost to the initial connecting route to generate a connecting loop transportation plan, the method further includes: Send a reservation request to the battery swapping station corresponding to the opportunistic battery swapping node in the intermodal transport plan, and obtain battery swapping time window confirmation information; Based on the battery swapping time window confirmation information, the dwell time of the vehicle at each transportation task node is adjusted to generate an updated transportation plan that includes battery swapping time constraints. The updated transportation plan is pushed to the vehicle terminal, and a battery swapping reminder is sent when the vehicle approaches an opportunistic battery swapping node.

4. The method according to claim 1, characterized in that, After the step of calculating the opportunistic battery swapping cost incurred by the vehicle deviating from the initial connecting route to the opportunistic battery swapping node to complete the battery swap and return to the original route, the method further includes: Obtain the opportunistic battery swapping cost of multiple opportunistic battery swapping nodes within the same potential battery swapping area; The multiple opportunistic battery swapping nodes are sorted according to the opportunistic battery swapping cost to generate a priority list of battery swapping nodes; Select the N nodes with the lowest cost from the battery swapping node priority list as candidate battery swapping nodes; Calculate the success probability of battery swapping at the candidate battery swapping nodes, use the product of the success probability and the opportunistic battery swapping cost as a comprehensive evaluation index, and select the node with the best comprehensive evaluation index as the recommended battery swapping node.

5. The method according to claim 1, characterized in that, After the step of weighted summing the necessary battery swapping risk cost and the estimated cost to obtain the comprehensive necessary battery swapping cost, the method further includes: Extract actual necessary battery swapping cases for similar road sections and time periods from historical data, and calculate the average detour distance and average waiting time for historical cases; The average detour distance and average waiting time are compared with the currently calculated necessary battery swapping cost to generate a cost deviation rate; When the cost deviation rate exceeds a preset threshold, the main factors causing the deviation are identified. These main factors include changes in traffic congestion, changes in the service capacity of battery swapping stations, or errors in electricity prediction. The calculation parameters for the necessary battery swapping cost are adjusted based on the aforementioned key factors, the adjusted necessary battery swapping cost is recalculated, and the adjusted necessary battery swapping cost is used for comparison with the opportunistic battery swapping cost.

6. A vehicle dispatching system, characterized in that, The system includes: one or more processors and memory; The memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the system to perform the method as described in any one of claims 1-5.

7. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the vehicle dispatching system, it causes the system to perform the method as described in any one of claims 1-5.

8. A computer program product, characterized in that, When the computer program product is run on the vehicle dispatching system, it causes the system to perform the method as described in any one of claims 1-5.

Images