Regional scheduling method and system for garbage transport vehicle

Abstract

The invention belongs to the technical field of vehicle scheduling, and discloses a garbage transport vehicle regional scheduling method and system, which can dynamically evaluate the vehicle workload by sensing the states of a vehicle team and a collection point in real time, identify transferable tasks and appropriate receiving vehicles and calculate the task transfer cost when the load imbalance condition is found and processed, and can realize the regional scheduling of garbage transport vehicles. A reasonable task transfer scheme can be determined; therefore, the problem caused by disconnection of the static plan and the actual situation is avoided, and the efficiency and quality of garbage collection operation are improved.

Description

Technical Field

[0001] The present application relates to the field of vehicle dispatching technology, and in particular to a method and system for regional dispatching of garbage trucks. Background Art

[0002] In urban waste collection and transportation systems, transport fleets typically rely on static scheduling plans for task allocation. These plans, based on historical data and predictive models of waste generation patterns, pre-assign a fixed sequence of collection points and routes to each vehicle, aiming to cover all service points. However, the dynamic and complex nature of the urban environment leads to significant deviations between actual waste distribution and planned operations. For example, waste generation rates vary significantly across different area types (e.g., commercial, residential, and industrial areas): commercial areas may quickly overflow during peak hours, residential areas experience a surge in waste volumes during holidays, and industrial areas may experience uneven waste volumes due to production fluctuations.

[0003] The rigidity of static scheduling leads to a serious imbalance in the workload of the fleet. High-load vehicles are overloaded with tasks and find it difficult to complete the collection on time, resulting in garbage accumulation and environmental degradation; low-load vehicles are not able to complete the tasks due to insufficient tasks, resulting in idle vehicles, manpower and fuel resources. This problem is particularly prominent under dynamic interference. For example, traffic control, road construction or vehicle failure may cause the original route to fail, and the existing system lacks real-time adjustment capabilities, making it difficult to reallocate tasks in a timely manner. In addition, the service time window requirements of collection points (such as high-priority points need to be processed within a short time) and the differences in vehicle functions (such as special vehicles can only handle specific types of garbage and have load restrictions) further increase the complexity of scheduling.

[0004] Existing dispatch systems have significant limitations when addressing the above challenges. First, the static models they rely on are unable to perceive vehicle status (such as location, load, task progress) and collection point dynamics (such as overflow level and garbage type) in real time, resulting in a disconnect between plans and reality. Second, rescheduling requires comprehensive consideration of multiple constraints: path accessibility requires the integration of real-time road conditions and traffic light data; task transfers must match vehicle functions and garbage types; and load balancing requires a balance between time window compliance, load restrictions, and global efficiency. However, traditional systems lack the ability to optimize multiple objectives and make rapid decisions. They often use a single indicator (such as the shortest path) or manual intervention, making it difficult to generate feasible solutions that balance efficiency and constraints within a limited time.

[0005] More critically, the system's ability to respond to emergencies is weak. For example, when a collection point overflows and requires emergency treatment, the system must quickly identify available vehicles. However, existing technologies often experience response delays due to inefficient algorithms or delayed data integration. Furthermore, inter-task dependencies (such as the need for clearance tasks to take precedence over transport tasks) are not adequately modeled, and arbitrary task transfers can lead to logical conflicts. These issues make it difficult for the fleet to achieve dynamic load balancing, ultimately impacting waste collection efficiency and urban environmental hygiene.

[0006] In summary, there is an urgent need for a scheduling method that can perceive data in real time, dynamically evaluate loads, and intelligently optimize task allocation in order to break through the limitations of static planning, improve resource utilization and operational adaptability, and ensure the efficient and stable operation of the urban waste collection and transportation system. Summary of the Invention

[0007] The purpose of this application is to provide a method and system for regional scheduling of garbage trucks, which can avoid problems caused by the disconnection between static plans and actual conditions, and improve the efficiency and quality of garbage collection operations.

[0008] In a first aspect, the present application provides a method for regional dispatching of garbage transport vehicles for dispatching garbage transport fleets within a city. The method comprises the following steps:

[0009] A1. Obtain real-time status information for each garbage truck and collection point within the fleet during pre-planned transportation tasks.

[0010] A2. Calculate the real-time workload of each garbage truck based on real-time status information of each garbage truck and collection point, and assess the overall workload balance of the fleet.

[0011] A3. If the overall workload distribution of the fleet is uneven, the system identifies tasks that meet the transfer criteria from the current task lists of garbage trucks whose real-time workload exceeds the load threshold. Based on the real-time status information of each garbage truck and collection point, as well as the task type of the task to be transferred, it identifies candidate vehicles in the fleet that can receive the task.

[0012] A4. Calculate the task transfer cost of each candidate vehicle receiving the task to be transferred;

[0013] A5. Determine the final task transfer plan based on the task transfer cost, generate a new task sequence and driving route for the relevant vehicles, and send it to the on-board terminals of the relevant vehicles.

[0014] Preferably, the real-time status information of each garbage transport vehicle includes the vehicle location, current load and current task sequence; the real-time status information of each garbage collection point includes the collection point location, garbage overflow level and garbage type.

[0015] Preferably, step A2 includes:

[0016] A201. Based on the vehicle location and collection point location, combined with real-time traffic information, estimate the predicted time for the garbage truck to arrive at each collection point for uncompleted tasks;

[0017] A202. Based on the vehicle's mission sequence, predict the remaining time for the garbage truck to complete the current mission sequence. Combined with the garbage truck's current load capacity, calculate the garbage truck's remaining load capacity.

[0018] A203. Garbage collection points are divided into different priorities based on their overflow levels. The service time window for each garbage collection point is determined based on the type of garbage. The higher the priority, the shorter the corresponding service time window.

[0019] A204. Calculate the real-time workload of each garbage truck based on the predicted arrival time of each garbage collection point for its unfinished mission, the remaining load capacity of the garbage truck, the priority of the garbage collection point, and the service time window;

[0020] A205. Based on the real-time workload of each garbage truck, calculate the variance of the overall workload of the fleet and evaluate the balance of the overall workload distribution of the fleet. If the variance is greater than a preset variance threshold, it is determined that the overall workload distribution of the fleet is unbalanced.

[0021] Preferably, step A203 includes:

[0022] Based on the location of the collection point, determine the area type of the area where the collection point is located to determine the estimated waste generation rate of the collection point; the area type includes commercial area, residential area and industrial area;

[0023] Based on the estimated waste generation rate and the current waste overflow level, predict the waste overflow level of each waste collection point within a preset time period in the future;

[0024] Combined with the predicted overflow level and garbage type, fuzzy logic algorithm is used to determine the priority of each garbage collection point;

[0025] According to the priority and garbage type of garbage collection points, the service time window of each garbage collection point is determined using a piecewise function.

[0026] Preferably, step A204 includes:

[0027] The time pressure index is calculated based on the predicted time for the garbage truck to arrive at each garbage collection point with unfinished tasks and the service time window of the garbage collection point;

[0028] Calculate the remaining load ratio based on the remaining load capacity and maximum load of the garbage truck;

[0029] According to the priority of the garbage collection point, a preset priority impact factor query table is searched to determine the priority impact factor corresponding to the priority;

[0030] The real-time workload of each garbage truck is calculated based on the time pressure index, remaining load ratio and priority influencing factor.

[0031] Preferably, step A3 includes:

[0032] A301. For garbage trucks whose real-time workload exceeds the load threshold, analyze the dependencies between tasks in their current task sequence and construct a task dependency graph. The task dependency graph uses tasks as nodes and the dependencies between tasks as directed edges, which represent the order in which tasks are executed.

[0033] A302. Based on the task dependency graph, starting from the end of the current task sequence, traverse the task nodes in reverse to identify the task to be transferred. If the current task node does not depend on any subsequent task node and meets the preset task transfer conditions, it is determined to be a task to be transferred;

[0034] A303. According to the task type to be transferred and the processing function of the garbage truck, select the vehicle with the matching processing function from each garbage truck whose real-time workload does not exceed the load threshold;

[0035] A304. According to the vehicle position of the vehicle matching the processing function and the location of the collection point corresponding to the task to be transferred, determine the accessibility of each processing function matching the vehicle to the garbage collection point corresponding to the task to be transferred;

[0036] A305. Based on the accessibility judgment result, identify candidate vehicles that can receive the task to be transferred from vehicles with matching processing functions. Candidate vehicles must meet the following requirements: the accessibility judgment result is reachable, and the vehicle's remaining load capacity is greater than the amount of garbage corresponding to the task to be transferred.

[0037] Preferably, step A304 includes:

[0038] Get real-time traffic information; real-time traffic information includes traffic flow, average speed and congestion conditions of each road section;

[0039] Calculate the minimum estimated travel time for each vehicle matching the processing function to reach the garbage collection point corresponding to the task to be transferred using the Dijkstra algorithm based on the vehicle position of the vehicle matching the processing function, the location of the collection point corresponding to the task to be transferred, and the real-time road condition information;

[0040] If the minimum estimated driving time exceeds the preset time threshold, it is determined that the corresponding vehicle is unreachable to the garbage collection point corresponding to the task to be transferred; otherwise, it is determined that the corresponding vehicle is reachable to the garbage collection point corresponding to the task to be transferred.

[0041] Preferably, step A4 includes:

[0042] A401. Obtain the current task sequence of each candidate vehicle, extract the garbage collection point location and service time window corresponding to each task in the current task sequence;

[0043] A402. For each candidate vehicle, insert the task to be transferred into a different position in its current task sequence to generate multiple alternative task sequences. Based on real-time traffic information and using a path optimization algorithm, calculate the total task duration for each alternative task sequence and the estimated completion time for each task in each alternative task sequence.

[0044] A403. For each candidate task sequence, calculate the time window violation penalty based on the expected completion time of each task, the service time window of the garbage collection point, and the priority of the garbage collection point;

[0045] A404. For each candidate mission sequence, an overload penalty is calculated based on the vehicle's load after completing the mission. If the vehicle does not exceed the preset load occupancy threshold, the overload penalty is zero. If the vehicle exceeds the preset load occupancy threshold, the overload penalty is proportional to the excess.

[0046] A405. For each alternative task sequence, calculate the variance of the fleet's overall workload after task transfer as a measure of the fleet's overall workload imbalance;

[0047] A406. Normalize the total mission duration, the time window violation penalty, the overload penalty, and the overall fleet load imbalance metric;

[0048] A407. For each candidate task sequence of each candidate vehicle, the task transfer cost is calculated using the weighted average method based on the normalized total task duration, the normalized time window violation penalty, the normalized overload penalty, and the normalized overall fleet load imbalance measure.

[0049] Preferably, step A402 includes:

[0050] Obtain traffic light data for urban roads; traffic light data includes the location, timing plan, and cycle duration of traffic lights;

[0051] For each candidate task sequence of each candidate vehicle, based on real-time road conditions and traffic light data, a path optimization algorithm is used to replan the vehicle's driving route in each candidate task sequence to obtain an updated candidate task sequence;

[0052] Based on the updated alternative task sequences, the total task duration of each alternative task sequence and the estimated completion time of each task in each alternative task sequence are calculated.

[0053] In a second aspect, the present application provides a regional dispatching system for garbage trucks, which is used for dispatching garbage truck fleets within a city. The system includes multiple vehicle-mounted terminals and a dispatching server. Each garbage truck is provided with an on-board terminal, and each on-board terminal is in communication with the dispatching server.

[0054] The vehicle-mounted terminal is used to send the real-time status information of the garbage transport vehicle to the dispatch server;

[0055] The scheduling server is provided with a scheduling program, and the scheduling program includes:

[0056] The data acquisition module is used to obtain real-time status information of each garbage truck and each garbage collection point in the garbage transportation fleet during the process of the garbage transportation fleet performing pre-planned transportation tasks;

[0057] The load assessment module is used to calculate the real-time workload of each garbage truck based on the real-time status information of each garbage truck and each garbage collection point, and evaluate the overall workload distribution balance of the fleet;

[0058] an identification module for identifying, if the overall workload distribution of the fleet is uneven, tasks to be transferred that meet the transfer conditions from the current task sequences of garbage trucks whose real-time workloads exceed a load threshold, and identifying candidate vehicles in the fleet that can receive the tasks to be transferred based on the real-time status information of each garbage truck and each garbage collection point and the task type of the tasks to be transferred;

[0059] A cost calculation module, configured to calculate the task transfer cost of each candidate vehicle receiving the task to be transferred;

[0060] The solution generation module is used to determine the final task transfer solution based on the task transfer cost, generate a new task sequence and driving route for the relevant vehicles, and send it to the on-board terminal of the relevant vehicles.

[0061] Beneficial effects: The present application provides a method and system for regional scheduling of garbage trucks. By sensing the status of the fleet and collection points in real time, the system dynamically evaluates the vehicle workload. When load imbalance is discovered and handled, the system identifies transferable tasks and suitable receiving vehicles and calculates the task transfer cost to determine a reasonable task transfer plan. This avoids problems caused by the disconnection between static plans and actual conditions, and improves the efficiency and quality of garbage collection operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1Flowchart of the regional scheduling method for garbage trucks provided in an embodiment of the present application.

[0063] Figure 2 A schematic diagram of the structure of the regional dispatching system for garbage trucks provided in an embodiment of the present application.

[0064] Explanation of reference numerals: 1. On-vehicle terminal; 2. Dispatching server; 201. Data acquisition module; 202. Load assessment module; 203. Identification module; 204. Cost calculation module; 205. Solution generation module. DETAILED DESCRIPTION

[0065] The technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. The components of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for protection, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work fall within the scope of protection of this application.

[0066] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0067] refer to Figure 1 This application proposes a method for regional dispatching of garbage trucks for dispatching garbage truck fleets within a city. The method comprises the following steps:

[0068] A1. Obtain real-time status information for each garbage truck and collection point within the fleet during pre-planned transportation tasks.

[0069] A2. Calculate the real-time workload of each garbage truck based on real-time status information of each garbage truck and collection point, and assess the overall workload balance of the fleet.

[0070] A3. If the overall workload distribution of the fleet is uneven, the system identifies tasks that meet the transfer criteria from the current task lists of garbage trucks whose real-time workload exceeds the load threshold. Based on the real-time status information of each garbage truck and collection point, as well as the task type of the task to be transferred, it identifies candidate vehicles in the fleet that can receive the task.

[0071] A4. Calculate the task transfer cost of each candidate vehicle receiving the task to be transferred;

[0072] A5. Determine the final task transfer plan based on the task transfer cost, generate a new task sequence and driving route for the relevant vehicles, and send it to the on-board terminals of the relevant vehicles.

[0073] Step A1, the foundation of the method, emphasizes continuous acquisition of real-time status information for each vehicle and collection point within the fleet during the mission. This enables the dispatch system to maintain up-to-date information on vehicle location, payload, mission progress, and the amount of waste and overflow at collection points. This overcomes the inherent lag in static dispatch information and provides accurate data support for subsequent dynamic decision-making.

[0074] Step A2, by assessing overall balance, can promptly identify situations where some vehicles are overloaded while others are underloaded, thereby determining whether rescheduling is necessary. For example, workload can be assessed by factors such as the number of remaining service points, the estimated time required to complete remaining tasks, and the ratio of current load to maximum load. The balance of the fleet's overall workload distribution can be measured by calculating the variance or standard deviation of each vehicle's workload. If this statistic exceeds a preset threshold, the fleet is deemed unbalanced.

[0075] In step A3, transfer conditions can include whether the task has started, whether the task has a time window constraint, and whether the task has dependencies on other tasks. Candidate vehicles must be capable of handling the type of waste corresponding to the task to be transferred, and their current workload must not exceed the load threshold. Furthermore, the vehicle's current location must be accessible to the location of the task to be transferred. This step is the core of dynamic adjustment, determining the tasks to be adjusted and the range of vehicles that can handle them, providing the basis for subsequent solution generation.

[0076] Calculating the transfer cost in step A4 involves quantitatively evaluating different transfer options to select the optimal one. For example, the task transfer cost can comprehensively consider factors such as the increase in total mileage and total operating time after the candidate vehicle completes the original task sequence plus the task to be transferred, whether the time window constraints of other tasks will be violated, whether overloading will occur, and the degree of improvement in the overall load distribution of the fleet after the transfer. These costs can be calculated using a path planning algorithm and a task sequence evaluation model.

[0077] In step A5, the lowest-cost solution is selected as the final task transfer plan. Then, based on the finalized task transfer plan, a new task sequence and driving route (routes can be generated based on existing path optimization algorithms, which are not limited here) are generated for the vehicles involved in the task transfer (the vehicle transferring the task to be transferred and the vehicle receiving the task to be transferred). These updated instructions are then sent to the vehicle's onboard terminal. This enables the vehicles to execute tasks according to the new scheduling plan, thereby achieving dynamic and balanced adjustments to the fleet's workload, improving overall operational efficiency and emergency response capabilities.

[0078] Specifically, this method aims to solve the problem of unbalanced workload of urban garbage transportation fleets in a dynamic environment due to the deviation between static scheduling plans and actual conditions. Through real-time perception, dynamic evaluation, task transfer and solution optimization, the rational allocation of fleet resources is achieved, and the overall operational efficiency and responsiveness are improved. As a result, vehicles can perform tasks according to the latest scheduling plan, thereby dynamically adjusting the workload distribution of the fleet, alleviating the pressure on high-load vehicles, making full use of the resources of low-load vehicles, avoiding problems caused by the disconnection between static plans and actual conditions, and improving the efficiency and quality of garbage collection operations. Compared with solutions that rely solely on static plans, this method can respond to on-site changes in real time, dynamically optimize resource allocation, and significantly improve the fleet's ability to cope with complex urban environments.

[0079] Through the above technical solution, this application solves the problem of unbalanced vehicle workloads in urban garbage transport fleets during the execution of tasks, and performs dynamic scheduling. This method can perceive the status of the fleet and collection points in real time, dynamically evaluate the vehicle workload, and promptly detect and handle load imbalances. By identifying transferable tasks and suitable receiving vehicles, and calculating the cost of task transfer, a reasonable task transfer plan can be determined. In this way, the dynamic optimization of fleet resources is achieved, the pressure on high-load vehicles is alleviated, the utilization rate of low-load vehicles is improved, the long-term backlog of garbage in some areas is avoided, and the overall efficiency and responsiveness of garbage collection operations are improved.

[0080] Preferably, the real-time status information of each garbage transport vehicle includes the vehicle location, current load and current task sequence; the real-time status information of each garbage collection point includes the collection point location, garbage overflow level and garbage type.

[0081] Vehicle location refers to the real-time coordinates of the garbage truck in geographic space, which can be obtained through positioning modules such as the vehicle's GPS module or Beidou module. Current load refers to the actual weight of garbage currently loaded by the garbage truck, which can be obtained through load sensors. Current task sequence refers to the list of garbage collection tasks currently being executed and those to be executed later, which is obtained by pre-planning transportation tasks.

[0082] The collection point location refers to the fixed coordinates of each collection point in geographic space and can be pre-recorded. The overflow level refers to the level of garbage filling within the collection point and can be captured using ultrasonic sensors, infrared sensors, or image recognition technology. The garbage type refers to the type of garbage collected at the collection point, such as food waste, recyclables, hazardous waste, or other waste, and can be pre-set and recorded.

[0083] By clarifying and obtaining the above-mentioned specific real-time status information, the necessary data foundation is provided for the subsequent steps of calculating the real-time workload of each garbage truck, evaluating the overall workload distribution balance of the fleet, identifying tasks to be transferred that meet the transfer conditions, identifying candidate vehicles that can receive tasks to be transferred, and calculating the cost of task transfer.

[0084] In some embodiments, step A2 comprises:

[0085] A201. Based on the vehicle location and collection point location, combined with real-time traffic information, estimate the predicted time for the garbage truck to arrive at each collection point for uncompleted tasks;

[0086] A202. Based on the vehicle's mission sequence, predict the remaining time for the garbage truck to complete the current mission sequence. Combined with the garbage truck's current load capacity, calculate the garbage truck's remaining load capacity.

[0087] A203. Garbage collection points are divided into different priorities based on their overflow levels. The service time window for each garbage collection point is determined based on the type of garbage. The higher the priority, the shorter the corresponding service time window.

[0088] A204. Calculate the real-time workload of each garbage truck based on the predicted arrival time of each garbage collection point for its unfinished mission, the remaining load capacity of the garbage truck, the priority of the garbage collection point, and the service time window;

[0089] A205. Based on the real-time workload of each garbage truck, calculate the variance of the overall workload of the fleet and evaluate the balance of the overall workload distribution of the fleet. If the variance is greater than a preset variance threshold, it is determined that the overall workload distribution of the fleet is unbalanced.

[0090] In step A201, the estimated time required for the vehicle to reach these collection points is calculated based on the current location of the garbage truck and the locations of the collection points to be completed in its task sequence, taking into account real-time traffic conditions. This can be achieved using a path planning algorithm, such as the Dijkstra algorithm or the A* algorithm, combined with real-time traffic data (such as road congestion and average vehicle speed) to refine the path calculation results, resulting in a more realistic predicted arrival time.

[0091] In step A202, the total time required to complete the vehicle's currently assigned task sequence is predicted, and the remaining garbage load is calculated based on the vehicle's current load capacity. The estimated remaining time to complete the current task sequence can be calculated by adding the estimated service time for each task in the task sequence (for example, the time spent at a collection point to load garbage can be calculated by taking the historical average service time for each collection point as the estimated service time based on historical data) and the estimated travel time between tasks. The remaining load capacity of the garbage truck can be calculated by subtracting the vehicle's current actual load capacity from its maximum load capacity.

[0092] In step A203, collection points are classified according to their importance and urgency based on their fill level and type of waste, and a timeframe is set for collection completion. Collection points with high overflow levels or those containing waste types requiring urgent disposal (e.g., kitchen waste) are assigned higher priority and have shorter service windows, indicating that these points require prompt service. This can be achieved using pre-set rules or functions. For example, priority can be determined by looking up a table based on overflow level and waste type, or by using a fuzzy logic algorithm. The service time window is then determined using a piecewise function based on the priority and waste type.

[0093] In step A204, the information calculated or determined in the previous steps is comprehensively utilized to calculate a comprehensive indicator reflecting the current workload of each garbage truck. Specifically, this can be achieved by weighting or fusing information from multiple dimensions, including time pressure (e.g., the degree of alignment between the estimated arrival time and the service time window), load pressure (e.g., the ratio of remaining load capacity to total load), and task importance (e.g., the priority of the collection point to be served). A higher value indicates a higher workload for the vehicle.

[0094] In step A205, the dispersion of the workload values ​​for all vehicles in the fleet is calculated, and based on this dispersion, the uniformity of the fleet's overall load distribution is determined. This can be achieved by calculating the variance of the real-time workload values ​​for all vehicles in the fleet. Variance is a commonly used statistic for measuring data dispersion. A larger variance indicates greater differences in load between vehicles, and a more uneven distribution. By comparing the calculated variance with a pre-set threshold, it is possible to quantitatively determine whether the fleet is in a state requiring dynamic adjustment.

[0095] Specifically, this solution proposes a method for calculating the real-time workload of garbage trucks by comprehensively considering multiple factors, including time, load, task priority, and service time windows. It also uses variance to quantitatively assess the distribution and balance of the fleet's overall workload, thereby more accurately reflecting the fleet's actual operating status and providing a more reliable basis for subsequent dynamic scheduling decisions. This method overcomes the inaccurate and incomplete workload calculations found in existing technologies, making the assessment of the fleet's operating status more objective and reliable. This enables more timely and effective dynamic adjustments, preventing some collection points from overflowing for extended periods or idle vehicle resources, and improving the overall efficiency and responsiveness of garbage collection operations.

[0096] Preferably, step A203 may include:

[0097] Based on the location of the collection point, determine the area type of the area where the collection point is located to determine the estimated waste generation rate of the collection point; the area type includes commercial area, residential area and industrial area;

[0098] Based on the estimated waste generation rate and the current waste overflow level, predict the waste overflow level of each waste collection point within a preset time period in the future;

[0099] Combined with the predicted overflow level and garbage type, fuzzy logic algorithm is used to determine the priority of each garbage collection point;

[0100] According to the priority and garbage type of garbage collection points, the service time window of each garbage collection point is determined using a piecewise function.

[0101] Among them, the estimated garbage generation rate of the garbage collection point is determined according to the area type. This can be achieved by querying the preset area type-generation rate comparison table, which stores the average garbage generation rate of different area types per unit time (the rate can be obtained in advance through historical data statistics).

[0102] The predicted amount of garbage can be calculated first, and then converted into a predicted overflow level (i.e., the overflow level of garbage collection points within a preset future time period). Specifically, the current overflow level can be converted into the current amount of garbage based on the capacity of each garbage collection point. The estimated waste increment can be calculated by multiplying the estimated generation rate by the length of the preset time period. The predicted amount of garbage can be calculated by adding the estimated waste increment to the current amount of garbage. The predicted amount of garbage can then be converted into the overflow level of each garbage collection point within a preset future time period based on the capacity of each garbage collection point. The preset time period can be set, for example, for the next hour, three hours, or longer.

[0103] Among them, the priority of each garbage collection point is determined by combining the predicted garbage overflow level and garbage type, using a fuzzy logic algorithm. Specifically, this can be achieved by building a fuzzy inference system. The system takes the predicted garbage overflow level (for example, defining low, medium, and high fuzzy sets) and garbage type (for example, defining kitchen waste, recyclable, and other fuzzy sets) as input variables, performs fuzzy inference through a fuzzy rule base (for example, if the predicted overflow level is high and the garbage type is kitchen waste, then the priority is very high), and defuzzifies the inference result to obtain a clear priority value or level.

[0104] Among them, according to the priority and garbage type of the garbage collection point, a piecewise function is used to determine the service time window of each garbage collection point. Specifically, this can be achieved by defining a multi-segment mapping function, which outputs a specific time window range or duration based on the input priority value or level and garbage type. For example, for a point with a priority higher than a certain threshold and a garbage type of kitchen waste, its service time window is set to a shorter time period from the current moment; for a point with a lower priority, its service time window can be set to a flexible interval within a longer time period in the future. The specific piecewise function can be set according to actual needs and is not limited here.

[0105] Therefore, by introducing regional types and estimated garbage generation rates, the solution can predict future state changes at garbage points, overcoming the limitations of relying solely on current status for judgment. The predicted future overflow level and garbage type combined with the fuzzy logic algorithm can more finely assess the actual importance and urgency of garbage points, improving the accuracy of priority determination. Based on more accurate priorities and garbage types, the service time window is determined using a piecewise function, so that the time window setting is more in line with the actual needs of different points, providing more reliable input for subsequent real-time workload calculations. This predictive evaluation and refined priority / time window determination method, combined with the overall framework for calculating vehicle workload based on real-time status, can more effectively respond to the dynamics of urban garbage generation rates, improve the accuracy of workload assessment, and thus support more effective dynamic scheduling and load balancing.

[0106] Preferably, step A204 may include:

[0107] The time pressure index is calculated based on the predicted time for the garbage truck to arrive at each garbage collection point with unfinished tasks and the service time window of the garbage collection point;

[0108] Calculate the remaining load ratio based on the remaining load capacity and maximum load of the garbage truck;

[0109] According to the priority of the garbage collection point, a preset priority impact factor query table is searched to determine the priority impact factor corresponding to the priority;

[0110] The real-time workload of each garbage truck is calculated based on the time pressure index, remaining load ratio and priority influencing factor.

[0111] Among them, the time pressure index refers to the degree of time urgency faced by the garbage truck when completing the unfinished task. It can be calculated by the relative position relationship between the predicted time and the service time window. For example, the time pressure index can be calculated by the following formula: TP=max(0,(prediction time-service time window deadline) / service time window duration)+max(0,(service time window start time-prediction time) / service time window duration), TP is the time pressure index.

[0112] Among them, the remaining load ratio refers to the ratio of the amount of garbage that the garbage truck can currently load (i.e. the remaining load capacity) to its maximum load. It can be calculated by dividing the remaining load capacity by the maximum load. The lower the remaining load ratio, the greater the load pressure of the vehicle.

[0113] Among them, the priority impact factor refers to the weight of the importance or urgency of the garbage collection point task on the vehicle workload calculation. It can be determined by querying the preset priority impact factor query table. The table stores the impact factor values ​​corresponding to different priorities. The higher the priority, the larger the corresponding impact factor value.

[0114] The real-time workload is a comprehensive reflection of the pressure on garbage trucks across multiple dimensions, including time, load, and mission importance. Specifically, this can be calculated using a weighted combination of the time pressure index, remaining load ratio, and priority impact factor. For example, a weighted sum or weighted average can be used to combine the time pressure index, the inverse of the remaining load ratio, and the priority impact factor according to preset weights. This calculation method combines the time urgency of the vehicle's arrival at the mission point, the vehicle's current load status, and the importance of the mission itself, overcoming the limitations of considering only a single factor.

[0115] This solution combines the predicted arrival time of vehicles at uncompleted tasks (dependent on vehicle location, collection point location, and real-time road conditions), the vehicle's remaining load capacity (dependent on the vehicle's task sequence and current load), and the priority and service time window of waste collection points (dependent on collection point overflow and waste type). This approach, based on existing technologies, provides a more comprehensive and refined workload quantification method. This approach considers the dynamic state of vehicles and collection points, making the calculated workload more reflective of the actual conditions faced by vehicles. This provides a more accurate basis for subsequent assessment of the overall workload balance of the fleet, thereby supporting more effective rescheduling decisions.

[0116] In some embodiments, step A3 comprises:

[0117] A301. For garbage trucks whose real-time workload exceeds the load threshold, analyze the dependencies between tasks in their current task sequence and construct a task dependency graph. The task dependency graph uses tasks as nodes and the dependencies between tasks as directed edges, which represent the order in which tasks are executed.

[0118] A302. Based on the task dependency graph, starting from the end of the current task sequence, traverse the task nodes in reverse to identify the task to be transferred. If the current task node does not depend on any subsequent task node and meets the preset task transfer conditions, it is determined to be a task to be transferred;

[0119] A303. According to the task type to be transferred and the processing function of the garbage truck, select the vehicle with the matching processing function from each garbage truck whose real-time workload does not exceed the load threshold;

[0120] A304. According to the vehicle position of the vehicle matching the processing function and the location of the collection point corresponding to the task to be transferred, determine the accessibility of each processing function matching the vehicle to the garbage collection point corresponding to the task to be transferred;

[0121] A305. Based on the accessibility judgment result, identify candidate vehicles that can receive the task to be transferred from vehicles with matching processing functions. Candidate vehicles must meet the following requirements: the accessibility judgment result is reachable, and the vehicle's remaining load capacity is greater than the amount of garbage corresponding to the task to be transferred.

[0122] Constructing a task dependency graph involves abstracting each task in the garbage truck's current task sequence into a node, and representing the execution order or prerequisite relationships between tasks as directed edges. For example, if Task B must be executed after Task A is completed, there will be a directed edge from Task A to Task B in the task dependency graph. Specifically, data structures such as adjacency matrices or adjacency lists can be used to store the task dependency graph. Dependencies between tasks can be determined based on task type, geographic location, or a pre-set logical order.

[0123] The pre-set task transfer conditions can include whether the task has not yet started, whether the task type allows transfer, and whether the task is time-sensitive. This dependency-based reverse traversal method ensures that only tasks that will not disrupt the execution logic of subsequent tasks after being removed from the task sequence are considered for transfer.

[0124] The task type can be characterized by the type of waste it handles. Selecting vehicles with matching processing capabilities involves filtering out vehicles with corresponding waste handling capabilities from within the fleet's current workload threshold, based on the type of waste to be handled by the task (e.g., household waste, kitchen waste, recyclables, etc.). For example, a kitchen waste collection task requires a vehicle equipped with specialized collection and compression equipment. This information can be obtained from the vehicle's real-time status information or pre-set in the system.

[0125] Determining reachability involves assessing whether a vehicle with the matching processing function can actually reach the collection point corresponding to the waste transfer task. This typically involves calculating the path from the vehicle's current location to the collection point, taking into account factors such as real-time road conditions, traffic control, and vehicle type restrictions. Specifically, a path planning algorithm (such as the Dijkstra algorithm or the A* algorithm) combined with real-time traffic data can be used to estimate the estimated travel time or distance, which is then compared to a preset reachability threshold.

[0126] Identifying candidate vehicles involves further screening out vehicles with sufficient load capacity from among the available processing function-matched vehicles. The amount of garbage to be transferred can be calculated based on the garbage overflow level within a preset future time period and the capacity of the garbage collection point, as calculated in step A203.

[0127] Therefore, by analyzing task dependencies, we can accurately identify tasks that can be safely transferred, avoiding task sequence interruptions or logical errors caused by arbitrarily transferring tasks. Through functional matching, reachability assessment, and remaining load capacity checks, we can systematically screen vehicles capable of accepting and executing the tasks to be transferred. These interrelated steps work together to ensure the feasibility and effectiveness of task transfers, improving the success rate of dynamic scheduling. For example, if one task is to empty all trash bins at a collection point, and another is to transport the trash to a processing plant, the transport task is dependent on the emptying task. During the reverse traversal, the transport task is checked first. If it does not depend on subsequent tasks (for example, no subsequent emptying task requires this batch of trash) and meets the transfer criteria, it may be transferred. However, the emptying task depends on its predecessor (full trash) and may be a prerequisite for a subsequent transport task. Therefore, its transfer criteria are more complex or it may not be identified as a task to be transferred. Furthermore, only vehicles that can handle the type of trash, can reach the collection point, and have sufficient space are selected as candidates. This systematic screening process makes the identified tasks to be transferred and candidate vehicles more accurate and reliable.

[0128] Through the above steps, this solution can specifically and effectively identify tasks truly suitable for transfer from high-load vehicles, as well as those vehicles in the fleet that have the actual receiving capacity (including functions, location, and load capacity). Compared to simply assigning tasks based on distance or current load, this solution considers the inherent logical relationships between tasks and the multiple capacity constraints of vehicles, thereby improving the success rate and efficiency of task transfers, avoiding the transfer of tasks to inappropriate vehicles or the transfer of tasks that should not be transferred. Ultimately, it helps to more effectively balance the overall workload of the fleet and improve the overall efficiency and reliability of waste collection services.

[0129] Preferably, step A304 may include:

[0130] Get real-time traffic information; real-time traffic information includes traffic flow, average speed and congestion conditions of each road section;

[0131] Calculate the minimum estimated travel time for each vehicle matching the processing function to reach the garbage collection point corresponding to the task to be transferred using the Dijkstra algorithm based on the vehicle position of the vehicle matching the processing function, the location of the collection point corresponding to the task to be transferred, and the real-time road condition information;

[0132] If the minimum estimated driving time exceeds the preset time threshold, it is determined that the corresponding vehicle is unreachable to the garbage collection point corresponding to the task to be transferred; otherwise, it is determined that the corresponding vehicle is reachable to the garbage collection point corresponding to the task to be transferred.

[0133] Traffic flow on a road section reflects the number of vehicles passing through it per unit time; average speed reflects the average speed of vehicles traveling on that road section; and congestion levels, expressed for example as a congestion level or delay index, reflect the efficiency of that road section. Obtaining this information is the basis for dynamically assessing vehicle travel times.

[0134] Real-time traffic information can be used to construct or update a graph model of the road network, where nodes represent intersections or key locations, and edges represent road segments. Edge weights are no longer fixed distances or theoretical travel times, but rather estimated travel times calculated based on real-time traffic flow, average speed, and congestion. For example, travel time can be estimated by dividing the length of the segment by the real-time average speed, or adjusted by a delay factor multiplied by the congestion level. The Dijkstra algorithm is a classic single-source shortest path algorithm used to find the shortest path from a source node to all other nodes in a weighted directed graph. In this solution, the Dijkstra algorithm is applied to a road network graph with the vehicle's current location as the source node and the location of the collection point corresponding to the task to be transferred as the target node. The algorithm calculates the shortest estimated travel time required for the vehicle to reach the collection point, taking into account real-time road conditions. This calculation method more accurately reflects the actual vehicle capacity and travel time under current traffic conditions. The Dijkstra algorithm is a state-of-the-art technique, and its calculation process will not be described in detail here.

[0135] The preset time threshold is a pre-defined upper limit, determined, for example, by the urgency of the task, the service window requirements of the collection point, or the response time requirements of the dispatch system. Comparing the calculated minimum estimated travel time with this threshold provides a quantitative, time-based reachability criterion. Only vehicles that are expected to arrive within the specified time are considered reachable. This judgment result is directly used to screen candidate vehicles that can accept the task to be transferred.

[0136] This solution works in conjunction with the step of identifying candidate vehicles capable of receiving the task to be transferred. Before identifying candidate vehicles, it is necessary to determine which vehicles with matching processing capabilities can arrive at the task location in a timely manner. By introducing real-time traffic information and time-based Dijkstra algorithm calculations, this solution provides a more accurate accessibility assessment than based solely on location judgment. This precise accessibility judgment serves as a screening condition, ensuring that the candidate vehicles subsequently identified not only have the function to handle the task, but also have the ability to respond in a timely manner under current traffic conditions. This improves the accuracy and reliability of candidate vehicle identification, avoids assigning tasks to vehicles that cannot arrive in time due to traffic reasons, and thus improves the effectiveness of the subsequent task transfer scheme.

[0137] In some embodiments, step A4 comprises:

[0138] A401. Obtain the current task sequence of each candidate vehicle, extract the garbage collection point location and service time window corresponding to each task in the current task sequence;

[0139] A402. For each candidate vehicle, insert the task to be transferred into a different position in its current task sequence to generate multiple alternative task sequences. Based on real-time traffic information and using a path optimization algorithm, calculate the total task duration for each alternative task sequence and the estimated completion time for each task in each alternative task sequence.

[0140] A403. For each candidate task sequence, calculate the time window violation penalty based on the expected completion time of each task, the service time window of the garbage collection point, and the priority of the garbage collection point;

[0141] A404. For each candidate mission sequence, an overload penalty is calculated based on the vehicle's load after completing the mission. If the vehicle does not exceed the preset load occupancy threshold, the overload penalty is zero. If the vehicle exceeds the preset load occupancy threshold, the overload penalty is proportional to the excess.

[0142] A405. For each alternative task sequence, calculate the variance of the fleet's overall workload after task transfer as a measure of the fleet's overall workload imbalance;

[0143] A406. Normalize the total mission duration, the time window violation penalty, the overload penalty, and the overall fleet load imbalance metric;

[0144] A407. For each candidate task sequence of each candidate vehicle, the task transfer cost is calculated using the weighted average method based on the normalized total task duration, the normalized time window violation penalty, the normalized overload penalty, and the normalized overall fleet load imbalance measure.

[0145] Step A401 obtains the candidate vehicle's current task sequence and the garbage collection point location and service time window corresponding to each task in the sequence. This information is the basic data for evaluating the impact of task transfer.

[0146] In step A402 , the path optimization algorithm may be Dijkstra algorithm, A* algorithm or genetic algorithm, etc. By simulating the execution path and time after task transfer, data is provided for subsequent cost calculation.

[0147] Among them, in step A403, if the expected completion time is within the service time window, the penalty is zero. If the expected completion time is earlier than the start of the time window or later than the end of the time window, a penalty is generated. The size of the penalty can be related to the degree of violation of the time window and can be adjusted according to the priority of the garbage collection point. The higher the priority point, the greater the penalty for violating the time window. For example, the time window violation penalty can be calculated according to the following method: if the expected completion time is earlier than the start moment of the time window, then the time window violation penalty = time advance amount × advance penalty coefficient × priority; if the expected completion time is later than the end moment of the time window, then the time window violation penalty = delay time amount × delay penalty coefficient × priority; wherein, the time advance amount is the difference between the start moment of the time window and the expected completion time, the delay time amount is the difference between the expected completion time and the end moment of the time window, and the total time window violation penalty is equal to the sum of the time window violation penalties of each subsequent task.

[0148] In step A404, the vehicle's load increases as it performs a task. After completing a task, if the vehicle's load exceeds a preset load occupancy threshold (e.g., 80% of the vehicle's maximum load), an overload penalty is incurred. The penalty is proportional to the load exceeding the threshold. This quantifies the risk of vehicle overload caused by task shifting. The overload penalty is calculated by multiplying the amount by which the vehicle's load exceeds the preset load occupancy threshold (as a percentage) by a preset proportional coefficient. The total overload penalty is equal to the sum of the overload penalties for each subsequent task.

[0149] In step A405, for each candidate task sequence, the variance of the fleet's overall workload after task transfer is calculated (see step A2 for the calculation process). The variance of the fleet's overall workload measures the degree of dispersion among the workloads of individual vehicles within the fleet. A larger variance indicates a more uneven load. Calculating the variance after task transfer allows us to assess the impact of the transfer solution on the overall load balance of the fleet.

[0150] Step A406 normalizes the total mission duration, time window violation penalty, overload penalty, and fleet load imbalance metric. These metrics have different dimensions and numerical ranges. Normalization converts them to a unified scale, such as the [0, 1] interval, to facilitate weighted calculations. Normalization methods can include min-max normalization or Z-score normalization.

[0151] In step A407, the normalized total mission duration, time window violation penalty, overload penalty, and overall fleet load imbalance metric are multiplied by their corresponding weights, and the sum is calculated to obtain the final mission transfer cost. Weights can be set based on the importance the scheduling strategy places on different factors. Through weighted averaging, the impact of multiple dimensions is combined into a single cost value.

[0152] Through the above technical solution, the present application can comprehensively and accurately calculate the task transfer cost of each candidate vehicle receiving the task to be transferred. By obtaining the current task information of the vehicle and simulating various execution plans after the task transfer, and optimizing the path in combination with the real-time road conditions, the total task duration and task completion time close to reality can be obtained. On this basis, the impact of task transfer on time window constraints, vehicle load and overall load balance of the fleet is quantitatively evaluated and converted into penalties or metrics. After normalizing these multi-dimensional indicators, the weighted average method is used to comprehensively calculate the task transfer cost. This cost value comprehensively reflects the advantages and disadvantages of the task transfer plan in many aspects, and provides a quantitative decision-making basis for the subsequent determination of the optimal task transfer plan. As a result, the scheduling system can select the task transfer plan with the least impact on individual vehicles and the fleet as a whole and the best benefits, thereby effectively solving the problem of incomplete and inaccurate calculation of task transfer costs and improving the scheduling effect of the garbage transportation fleet.

[0153] Preferably, step A402 may include:

[0154] Obtain traffic light data for urban roads; traffic light data includes the location, timing plan, and cycle duration of traffic lights;

[0155] For each candidate task sequence of each candidate vehicle, based on real-time road conditions and traffic light data, a path optimization algorithm is used to replan the vehicle's driving route in each candidate task sequence to obtain an updated candidate task sequence;

[0156] Based on the updated alternative task sequences, the total task duration of each alternative task sequence and the estimated completion time of each task in each alternative task sequence are calculated.

[0157] Among them, based on the sequence of task points that the vehicle needs to visit in sequence, combined with the current traffic congestion situation and the signal status of traffic lights along the way, a driving route with the shortest estimated travel time or the highest efficiency is calculated. Specifically, this can be achieved by using a dynamic path planning algorithm that integrates time and space factors, such as an algorithm based on a time-dependent graph or an improved Dijkstra algorithm that considers traffic light waiting time. The updated alternative task sequence output by this process contains a precisely calculated driving route that takes into account actual travel delays.

[0158] Then, based on the re-planned driving route and its estimated travel time, the driving time between each task point, the working time at the task point, and the delay time caused by waiting for traffic lights are accumulated to obtain the total time required to complete the entire task sequence and determine the estimated start and end time points of each task in the sequence. Specifically, the sequential accumulation method can be used to achieve this. Starting from the first task in the sequence, the arrival time, working time (the estimated service time mentioned above can be used), and departure time are calculated in turn, and used as the start time of the next task until the end of the sequence.

[0159] When calculating the total task duration and estimated completion time of each task in the alternative task sequence, this solution not only considers the impact of real-time traffic information on driving speed, but also introduces the additional impact of traffic light data on travel time. By incorporating the location, timing scheme, and cycle duration of traffic lights into the calculation model of the path optimization algorithm, the algorithm can more accurately predict the waiting time of vehicles when passing through signalized intersections, thereby planning a driving route that is closer to actual conditions. The total task duration and estimated task completion time calculated based on this more accurate route can more accurately reflect the actual time required for vehicles to complete the task sequence compared to the calculation results that only consider real-time road conditions. This more accurate time prediction provides a more reliable data basis for the subsequent calculation of time window violation penalties, evaluation of solution feasibility, and ultimate determination of task transfer costs, helping to select a task transfer solution that is more effective for overall fleet load balancing and efficiency improvement from multiple alternative solutions.

[0160] refer to Figure 2 The present application provides a regional dispatching system for garbage trucks, which is used for dispatching garbage truck fleets within urban areas. The system includes multiple vehicle-mounted terminals 1 and a dispatching server 2. Each garbage truck is provided with a vehicle-mounted terminal 1, and each vehicle-mounted terminal 1 is in communication with the dispatching server 2.

[0161] The vehicle-mounted terminal 1 is used to send the real-time status information of the garbage transport vehicle to the dispatch server 2;

[0162] The scheduling server 2 is provided with a scheduling program, and the scheduling program includes:

[0163] The data acquisition module 201 is used to obtain real-time status information of each garbage truck and each garbage collection point in the garbage transportation fleet during the process of the garbage transportation fleet performing the pre-planned transportation task (for details, refer to step A1 above);

[0164] Load assessment module 202, which is used to calculate the real-time workload of each garbage truck based on the real-time status information of each garbage truck and each garbage collection point, and assess the overall workload distribution balance of the fleet (for details, refer to step A2 above);

[0165] Identification module 203 is configured to, if the overall workload distribution of the fleet is uneven, identify tasks to be transferred that meet the transfer conditions from the current task sequences of garbage trucks whose real-time workloads exceed the load threshold, and identify candidate vehicles in the fleet that can receive the tasks to be transferred based on the real-time status information of each garbage truck and each garbage collection point and the task type of the tasks to be transferred (see step A3 above for details).

[0166] Cost calculation module 204, used to calculate the task transfer cost of each candidate vehicle receiving the task to be transferred (for details, refer to step A4 above);

[0167] The plan generation module 205 is used to determine the final task transfer plan based on the task transfer cost, generate a new task sequence and driving route for the relevant vehicles, and send it to the vehicle terminal 1 of the relevant vehicle (for the specific process, refer to step A5 above).

[0168] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for regional dispatching of garbage trucks, used for dispatching garbage truck fleets within a city, characterized in that: The steps of the method include: A1. Obtain real-time status information for each garbage truck and collection point within the fleet during pre-planned transportation tasks. A2. Calculate the real-time workload of each garbage truck based on real-time status information of each garbage truck and collection point, and assess the overall workload balance of the fleet. A3. If the overall workload distribution of the fleet is uneven, the system identifies tasks that meet the transfer criteria from the current task lists of garbage trucks whose real-time workload exceeds the load threshold. Based on the real-time status information of each garbage truck and collection point, as well as the task type of the task to be transferred, it identifies candidate vehicles in the fleet that can receive the task. A4. Calculate the task transfer cost of each candidate vehicle receiving the task to be transferred; A5. Determine the final task transfer plan based on the task transfer cost, generate a new task sequence and driving route for the relevant vehicles, and send it to the on-board terminals of the relevant vehicles.

2. A method for regional dispatching of garbage trucks according to claim 1, characterized in that: The real-time status information of each garbage truck includes the vehicle location, current load and current task sequence; the real-time status information of each garbage collection point includes the collection point location, garbage overflow level and garbage type.

3. A method for regional dispatching of garbage trucks according to claim 2, characterized in that: Step A2 includes: A201. Based on the vehicle location and collection point location, combined with real-time traffic information, estimate the predicted time for the garbage truck to arrive at each collection point for uncompleted tasks; A202. Based on the vehicle's mission sequence, predict the remaining time for the garbage truck to complete the current mission sequence. Combined with the garbage truck's current load capacity, calculate the garbage truck's remaining load capacity. A203. Garbage collection points are divided into different priorities based on their overflow levels. The service time window for each garbage collection point is determined based on the type of garbage. The higher the priority, the shorter the corresponding service time window. A204. Calculate the real-time workload of each garbage truck based on the predicted arrival time of each garbage collection point for its unfinished mission, the remaining load capacity of the garbage truck, the priority of the garbage collection point, and the service time window; A205. Based on the real-time workload of each garbage truck, calculate the variance of the overall workload of the fleet and evaluate the balance of the overall workload distribution of the fleet. If the variance is greater than a preset variance threshold, it is determined that the overall workload distribution of the fleet is unbalanced.

4. A method for regional dispatching of garbage trucks according to claim 3, characterized in that: Step A203 includes: Based on the location of the collection point, determine the area type of the area where the collection point is located to determine the estimated waste generation rate of the collection point; the area type includes commercial area, residential area and industrial area; Based on the estimated waste generation rate and the current waste overflow level, predict the waste overflow level of each waste collection point within a preset time period in the future; Combined with the predicted overflow level and garbage type, fuzzy logic algorithm is used to determine the priority of each garbage collection point; According to the priority and garbage type of garbage collection points, the service time window of each garbage collection point is determined using a piecewise function.

5. A method for regional dispatching of garbage trucks according to claim 3, characterized in that: Step A204 includes: The time pressure index is calculated based on the predicted time for the garbage truck to arrive at each garbage collection point with unfinished tasks and the service time window of the garbage collection point; Calculate the remaining load ratio based on the remaining load capacity and maximum load of the garbage truck; According to the priority of the garbage collection point, a preset priority impact factor query table is searched to determine the priority impact factor corresponding to the priority; The real-time workload of each garbage truck is calculated based on the time pressure index, remaining load ratio and priority influencing factor.

6. A method for regional dispatching of garbage trucks according to claim 2, characterized in that: Step A3 includes: A301. For garbage trucks whose real-time workload exceeds the load threshold, analyze the dependencies between tasks in their current task sequence and construct a task dependency graph. The task dependency graph uses tasks as nodes and the dependencies between tasks as directed edges, which represent the order in which tasks are executed. A302. Based on the task dependency graph, starting from the end of the current task sequence, traverse the task nodes in reverse to identify the task to be transferred. If the current task node does not depend on any subsequent task node and meets the preset task transfer conditions, it is determined to be a task to be transferred; A303. According to the task type to be transferred and the processing function of the garbage truck, select the vehicle with the matching processing function from each garbage truck whose real-time workload does not exceed the load threshold; A304. According to the vehicle position of the vehicle matching the processing function and the location of the collection point corresponding to the task to be transferred, determine the accessibility of each processing function matching the vehicle to the garbage collection point corresponding to the task to be transferred; A305. Based on the accessibility judgment result, identify candidate vehicles that can receive the task to be transferred from vehicles with matching processing functions. Candidate vehicles must meet the following requirements: the accessibility judgment result is reachable, and the vehicle's remaining load capacity is greater than the amount of garbage corresponding to the task to be transferred.

7. A method for regional dispatching of garbage trucks according to claim 6, characterized in that: Step A304 includes: Get real-time traffic information; real-time traffic information includes traffic flow, average speed and congestion conditions of each road section; Calculate the minimum estimated travel time for each vehicle matching the processing function to reach the garbage collection point corresponding to the task to be transferred using the Dijkstra algorithm based on the vehicle position of the vehicle matching the processing function, the location of the collection point corresponding to the task to be transferred, and the real-time road condition information; If the minimum estimated driving time exceeds the preset time threshold, it is determined that the corresponding vehicle is unreachable to the garbage collection point corresponding to the task to be transferred; otherwise, it is determined that the corresponding vehicle is reachable to the garbage collection point corresponding to the task to be transferred.

8. The method for regional dispatching of garbage trucks according to claim 2, characterized in that: Step A4 includes: A401. Obtain the current task sequence of each candidate vehicle, extract the garbage collection point location and service time window corresponding to each task in the current task sequence; A402. For each candidate vehicle, insert the task to be transferred into a different position in its current task sequence to generate multiple alternative task sequences. Based on real-time traffic information and using a path optimization algorithm, calculate the total task duration for each alternative task sequence and the estimated completion time for each task in each alternative task sequence. A403. For each candidate task sequence, calculate the time window violation penalty based on the expected completion time of each task, the service time window of the garbage collection point, and the priority of the garbage collection point; A404. For each candidate mission sequence, an overload penalty is calculated based on the vehicle's load after completing the mission. If the vehicle does not exceed the preset load occupancy threshold, the overload penalty is zero. If the vehicle exceeds the preset load occupancy threshold, the overload penalty is proportional to the excess. A405. For each alternative task sequence, calculate the variance of the fleet's overall workload after task transfer as a measure of the fleet's overall workload imbalance; A406. Normalize the total mission duration, the time window violation penalty, the overload penalty, and the overall fleet load imbalance metric; A407. For each candidate task sequence of each candidate vehicle, the task transfer cost is calculated using the weighted average method based on the normalized total task duration, the normalized time window violation penalty, the normalized overload penalty, and the normalized overall fleet load imbalance measure.

9. A method for regional dispatching of garbage trucks according to claim 8, characterized in that: Step A402 includes: Obtain traffic light data for urban roads; traffic light data includes the location, timing plan, and cycle duration of traffic lights; For each candidate task sequence of each candidate vehicle, based on real-time road conditions and traffic light data, a path optimization algorithm is used to replan the vehicle's driving route in each candidate task sequence to obtain an updated candidate task sequence; Based on the updated alternative task sequences, the total task duration of each alternative task sequence and the estimated completion time of each task in each alternative task sequence are calculated.

10. A garbage truck regional dispatching system for dispatching garbage truck fleets within a city, characterized by: The system includes multiple vehicle-mounted terminals and a dispatching server. Each garbage transport vehicle is provided with a vehicle-mounted terminal, and each vehicle-mounted terminal is in communication with the dispatching server. The vehicle-mounted terminal is used to send the real-time status information of the garbage transport vehicle to the dispatch server; The scheduling server is provided with a scheduling program, and the scheduling program includes: The data acquisition module is used to obtain real-time status information of each garbage truck and each garbage collection point in the garbage transportation fleet during the process of the garbage transportation fleet performing pre-planned transportation tasks; The load assessment module is used to calculate the real-time workload of each garbage truck based on the real-time status information of each garbage truck and each garbage collection point, and evaluate the overall workload distribution balance of the fleet; an identification module for identifying, if the overall workload distribution of the fleet is uneven, tasks to be transferred that meet the transfer conditions from the current task sequences of garbage trucks whose real-time workloads exceed a load threshold, and identifying candidate vehicles in the fleet that can receive the tasks to be transferred based on the real-time status information of each garbage truck and each garbage collection point and the task type of the tasks to be transferred; A cost calculation module, configured to calculate the task transfer cost of each candidate vehicle receiving the task to be transferred; The solution generation module is used to determine the final task transfer solution based on the task transfer cost, generate a new task sequence and driving route for the relevant vehicles, and send it to the on-board terminal of the relevant vehicles.

Images