Distributed cooperative scheduling system applied to global intelligent environmental sanitation

Abstract

The invention discloses a distributed cooperative scheduling system applied to global intelligent environmental sanitation, and particularly relates to the technical field of global scheduling. According to the method, the whole domain is divided into the sub-regions, the daily scene, the environmental protection scene and the quality scene are judged, each scene takes the minimization of the total operation cost, the minimization of the total carbon emission amount and the maximization of the operation coverage rate as optimization targets, the optimal solution coefficient is calculated in combination with the differentiation index weight, the problems of resource waste and service gap of a traditional one-sided scheduling mode are avoided, and the scheduling efficiency is improved. The scheduling scheme is more suitable for functional attributes and actual requirements of different areas, the multi-scene scheduling requirements are met, and the scheme accuracy is improved.

Description

Technical Field

[0001] This invention relates to the field of global scheduling technology, and more specifically, to a distributed collaborative scheduling system applied to global smart sanitation. Background Technology

[0002] Driven by both the acceleration of urbanization and the increasing demand for refined environmental sanitation management, smart sanitation across the entire region, as a core component of urban public services, directly impacts urban operating costs, ecological and environmental protection levels, and citizens' living experience through its scheduling efficiency and management quality.

[0003] However, the current traditional sanitation dispatching model still suffers from multi-dimensional technical bottlenecks, making it difficult to adapt to the management needs of comprehensive, dynamic, and multi-objective operations, including:

[0004] Existing sanitation scheduling is mostly planned in isolation based on administrative regions, without constructing differentiated scheduling logic for the functional attributes of different areas within the entire region. For example, daily cleaning in residential areas should prioritize controlling operating costs, operations in ecological protection areas should focus on reducing carbon emissions, and operations around large event venues should ensure high coverage. However, existing technologies only pursue operational efficiency or cost control, which cannot match the personalized optimization goals of daily operations, environmental protection, quality, and other scenarios. This leads to a disconnect between scheduling solutions and actual needs, resulting in both resource waste and service gaps.

[0005] Existing scheduling systems have a one-sided approach to handling constraints. They either strictly adhere to all constraints, resulting in no feasible solutions when resources are scarce, or they completely abandon constraints, pursuing only a single optimization objective, leading to operational violations or coordination failures, which further reduces the actual executability of scheduling solutions.

[0006] To address this, a distributed collaborative scheduling system for smart sanitation across the entire region has been launched. Summary of the Invention

[0007] To overcome the above-mentioned deficiencies of the prior art, embodiments of the present invention provide a distributed collaborative scheduling system for smart sanitation across the entire region.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A distributed collaborative scheduling system for smart sanitation across the entire region includes the following modules:

[0010] Scene recognition module: Divides the entire area into several sub-regions and identifies the set scene to which each sub-region belongs, including daily scenes, environmental protection scenes and quality scenes;

[0011] Distributed computing module: For each sub-region, a distributed computing architecture is used to construct and schedule computing tasks in combination with the relevant scenario. These tasks are then distributed to different computing nodes for parallel processing. Based on the optimization objectives of the current scenario, the optimal scheduling solution for each sub-region is obtained by selectively combining the set constraints.

[0012] Solution integration module: integrates the optimal solutions of each sub-region to construct a global scheduling scheme, detects conflicts and filters conflicting sub-regions to construct an evaluation region set, quantifies the degree of conflict and outputs the conflict level, and adjusts the global scheme based on the conflict level and priority index;

[0013] Solution Output Module: Outputs the final scheduling plan to the sanitation management platform in a visual format.

[0014] Specifically, the optimization goals for different scenarios are as follows:

[0015] The optimization goal for everyday scenarios is to minimize total operating costs;

[0016] The optimization objective for environmental protection scenarios is to minimize total carbon emissions;

[0017] The optimization objective for the quality scenario is to maximize job coverage.

[0018] Specifically, if the chosen constraint conditions are considered, the solution process is as follows:

[0019] For each sub-region scenario type, the optimization objective is used as the benchmark to solve and sort the solutions. The top three candidate solutions that meet all constraints are selected. The total operating cost, carbon emissions and operation coverage of each candidate solution are extracted. Based on the scenario-differentiated index weights, the parameters of the candidate solutions are normalized and weighted to calculate the optimal solution coefficient. The candidate solution with the highest optimal solution coefficient is selected as the optimal scheduling solution.

[0020] Specifically, if the constraints are abandoned, the solution process is as follows:

[0021] Based on the optimization objective, the top three candidate solutions are solved and ranked, and the optimal solution coefficient is calculated. At the same time, the degree of violation of the flexible constraints by the candidate solutions is analyzed, and the timeout duration of tasks, jobs, and collaboration is determined. After normalization, the violation coefficient is calculated in combination with the weights. The adjustment coefficient is obtained based on the mapping rule between the violation coefficient and the adjustment coefficient. The optimal solution coefficient and the adjustment coefficient are multiplied to obtain the limit coefficient. The candidate solution with the highest limit coefficient is selected as the optimal scheduling solution.

[0022] Specifically, flexible constraints include:

[0023] Task service time window, maximum continuous working time for personnel, and time difference for multi-resource collaboration.

[0024] Specifically, the collision detection process is as follows:

[0025] Resource allocation conflict: The same vehicle or personnel is assigned to two or more sub-areas and the work periods overlap;

[0026] Job assignment conflict: The job routes of adjacent sub-regions overlap spatially on the map;

[0027] Time scheduling conflict: The same shared facility is used by two or more sub-areas and the usage time interval is shorter than the preset time interval.

[0028] Specifically, the process for outputting the conflict level is as follows:

[0029] The conflict ratio of the number of sub-regions within the evaluation region set to the total number of sub-regions is used. A conflict ratio higher than the preset conflict threshold is considered a high conflict level, while a ratio lower than the preset conflict threshold is considered a low conflict level.

[0030] Specifically, the global scheduling scheme is adaptively adjusted based on the conflict level to output the final scheduling scheme, as follows:

[0031] If the conflict level is high, the global scheduling scheme will be marked and sent to the administrator for adjustment before the final scheduling scheme is output.

[0032] If the conflict level is low, the conflict sub-regions are classified according to the conflict type, the optimal solution coefficients and the scenarios to which the sub-regions of the same conflict type belong are extracted, the basic evaluation value of the scenario is set, and the optimal solution coefficients and the basic evaluation value are weighted according to the preset weights to calculate the priority index.

[0033] Resource allocation conflict: Prioritize allocating conflicting resources to the sub-region with the highest priority index, and re-solve the remaining sub-regions;

[0034] Job assignment conflict: Prioritize assigning overlapping routes to the sub-region with the highest priority index, and remove overlapping routes from the remaining sub-regions;

[0035] Time scheduling conflict: Prioritize allocating the right to use shared facilities to the sub-area with the highest priority index, and postpone the use of the remaining sub-areas.

[0036] The technical effects and advantages of this invention are as follows:

[0037] (1) By dividing the entire domain into sub-regions and determining the three scenarios of daily, environmental protection and quality, each scenario is optimized with the goal of minimizing total operating cost, minimizing total carbon emissions and maximizing operation coverage. The optimal solution coefficient is calculated by combining the weight of differentiated indicators, which avoids the resource waste and service gap problems of the traditional one-sided scheduling mode, makes the scheduling scheme more in line with the functional attributes and actual needs of different regions, adapts to the scheduling needs of multiple scenarios, and improves the accuracy of the scheme.

[0038] (2) By adopting a distributed computing architecture, the sub-regional scheduling tasks are distributed to different nodes for parallel processing, which solves the problems of low efficiency and congestion in traditional centralized computing and can quickly respond to sudden sanitation needs. At the same time, by combining the two solution modes of constraint conditions and abandoning constraint conditions, the degree of violation of flexible constraints is quantitatively evaluated and the optimization coefficient is calculated. This avoids the infeasibility of strictly following the constraints and prevents the violation of the operation caused by completely abandoning the constraints, thereby improving the feasibility of the scheduling scheme.

[0039] (3) By accurately detecting three types of conflicts—resource allocation, task allocation, and time planning—high and low conflict levels are classified based on the conflict ratio and handled differently. High conflict levels are submitted to management personnel for adjustment, while low conflict levels are allocated resources, routes, and facility usage rights based on priority index. This avoids the inefficiency and bias of manual conflict investigation, ensures coordinated and orderly sanitation operations across the entire area, and reduces efficiency losses caused by resource competition and route overlap. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the distributed collaborative scheduling system of the present invention applied to intelligent sanitation across the entire region. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] like Figure 1 As shown, the distributed collaborative scheduling system modules applied to the whole-area smart sanitation system are as follows:

[0043] The scene recognition module is used to divide the entire area into several sub-regions according to geographical regions, and to set the current scene for each sub-region. Scenes include daily scenes, environmental protection scenes and quality scenes.

[0044] Additional explanations: the division criteria include geographical boundaries (such as administrative boundaries, rivers, and roads), road network distribution (such as the density of main roads and the distribution of streets and alleys), and sanitation resource density (such as the location of garbage transfer stations and the distribution of sanitation workers). In addition, the distribution calculation module can dynamically adjust the scope of sub-regions based on real-time data (such as a sudden large accumulation of garbage in a certain sub-region), for example, by incorporating parts of adjacent sub-regions into this region to enhance resource allocation capabilities. Scene recognition can be automatically triggered by user input or preset conditions. For example, daily scenes can be automatically recognized based on the date, environmental protection scenes can be set based on notices issued by environmental protection departments, and quality scenes can be set based on activity information.

[0045] The distributed computing module is used to construct and schedule computing tasks for each sub-region using a distributed computing architecture combined with the current scenario. The scheduling computing tasks of each sub-region are allocated to different computing nodes for parallel processing. The algorithm parameters are adjusted according to the constraints of different scenarios. Each computing node selects and combines the set constraints based on the optimization goal of the current scenario to solve for the optimal scheduling solution of each sub-region.

[0046] Additional notes: When solving for the optimal scheduling solution for each sub-region, constraints are imposed, including but not limited to:

[0047] Vehicle capacity restrictions: For garbage trucks, water trucks, etc., the restriction is that "the loading volume of a single operation shall not be less than or equal to the vehicle's rated capacity". For example, a 10-ton garbage truck shall not carry more than 10 tons of garbage to avoid overloading violations.

[0048] Task service time window: For specific areas (such as main roads and areas around schools), the constraints are "work start time ≥ earliest allowed time, work end time ≤ latest deadline"; for example, main road cleaning should be completed between 5:00 and 7:00 to avoid affecting traffic during the morning rush hour;

[0049] Maximum continuous working time for personnel: "single person's single continuous working time ≤ 4 hours, and daily cumulative working time ≤ 8 hours" to avoid fatigue work;

[0050] Multi-resource collaborative matching: For collaborative operation scenarios such as sweeper trucks and garbage trucks, water trucks and washing sweeper trucks, the constraint is "the time difference between the preceding and following processes ≤ a preset threshold". For example, after a sweeper truck completes the cleaning of a certain section of road, a garbage truck needs to arrive within 30 minutes to avoid secondary accumulation of garbage.

[0051] Collaborative operation rules: For cross-regional operations (such as district boundaries), the constraint is that "the operation range of adjacent areas does not overlap and is fully covered". For example, the road section at the junction of area A and area B must be clearly assigned to one of them to avoid missed sweeps or duplicate operations.

[0052] Specifically:

[0053] The optimization goal for everyday scenarios is to minimize total operating costs;

[0054] The optimization objective for environmental protection scenarios is to minimize total carbon emissions;

[0055] The optimization objective for the quality scenario is to maximize job coverage;

[0056] Total operating cost = fuel cost + labor cost + fixed cost; where fuel cost = operating mileage × unit fuel consumption × fuel price; labor cost = operating hours × hourly wage; fixed cost = daily amortization of vehicle depreciation + maintenance cost;

[0057] Total carbon emissions = (Operating mileage of fuel-powered vehicles × unit carbon emission coefficient) + (Operating mileage of new energy vehicles × unit emission reduction coefficient) (Note: Coefficient for fuel-powered sweepers = 2.6 kg CO2 / km, emission reduction coefficient for new energy vehicles = 2.0 kg CO2 / km).

[0058] Coverage rate = (Actual number of operation points / Total number of emergency task points) × 100%;

[0059] If the set constraints are selected, each computing node will solve and sort the solutions based on the optimization objective for the scene type of the sub-region, and select the top three candidate solutions that satisfy all constraints; if there are fewer than three candidate solutions, all feasible solutions will be selected.

[0060] Additional explanation,

[0061] Transform the sanitation scheduling problem into an optimization problem;

[0062] For example, in each sub-region, the sanitation scheduling problem can be abstracted into a multi-objective, multi-constraint optimization problem. Its core is to find a set of optimal task allocation schemes so that the optimization objectives (cost, carbon emissions, coverage) are optimal under the premise of satisfying the constraints.

[0063] (1) Definition of decision variables;

[0064] Suppose that in a subregion:

[0065] N task points (such as garbage collection points);

[0066] M operating vehicles;

[0067] K workers;

[0068] The decision variable can then be defined as:

[0069] : Indicates whether task point i is executed by vehicle j;

[0070] : Indicates whether person k is assigned to vehicle j;

[0071] : Departure time of vehicle j;

[0072] The route (sequence) of vehicle j;

[0073] Based on the above definition of decision variables, combined with the objective functions and constraints of the different scenarios mentioned above;

[0074] The key parameters of the particle swarm optimization (PSO) algorithm used are as follows:

[0075] Population size: 50-100 particles;

[0076] Maximum number of iterations: 200~500;

[0077] Learning factors: c1=1.5, c2=1.5 (individual and social learning weights);

[0078] Inertia weight: w=0.7 (dynamically decreasing, decreasing by 0.005 per generation);

[0079] Velocity range: [-v_max, v_max], where v_max = 20% of the particle's position range;

[0080] Location range: set according to the type of decision variable (0 / 1 or continuous value);

[0081] Each particle represents a complete scheduling scheme, and its position vector is designed as follows:

[0082] The first N×M dimensions represent the allocation relationship between task points and vehicles (0 / 1).

[0083] The middle K×M dimensions represent the allocation relationship between personnel and vehicles (0 / 1).

[0084] The last M dimensions represent the departure time of each vehicle (continuous values).

[0085] The route sequence can be encoded using integers to represent the order in which task points are visited;

[0086] The fitness function is a transformation of the objective function, for example:

[0087] In everyday scenarios: Adaptability = -Total operating cost (PSO defaults to minimization, taking a negative sign).

[0088] Environmental scenario: Adaptability = −Total carbon emissions;

[0089] Quality Scenario: Fit = Coverage;

[0090] Then proceed with the iterative process:

[0091] Initialization: Randomly generate a set of particles (i.e., the initial scheduling scheme).

[0092] Assess fitness: Calculate the fitness value for each particle;

[0093] Adjust the particle state according to the velocity and position update formula;

[0094] Update the individual optimal and global optimal;

[0095] Continue until the maximum number of iterations is reached or convergence occurs;

[0096] The sanitation scheduling problem was modeled as an optimization problem that can be handled by particle swarm optimization.

[0097] Daily scenarios (core objective: minimize total operating costs)

[0098] Solving and sorting: The computing nodes solve the sub-region scheduling scheme through optimization algorithms (such as particle swarm optimization), output all feasible solutions that meet the constraints (vehicle capacity, time window, etc.), and sort them in ascending order of "total operating cost".

[0099] Select the top three solutions: Select the three solutions with the lowest cost as candidate solutions, denoted as S1, S2, and S3, where S1 has the lowest cost, followed by S2, and then S3.

[0100] Environmental protection scenario (core objective: minimizing total carbon emissions)

[0101] Solution and sorting: The computation node takes "total carbon emissions" as the core optimization objective and outputs all feasible solutions that meet the constraints, sorted by "total carbon emissions" from smallest to largest (the smaller the emissions, the better).

[0102] The top three solutions are selected as candidate solutions, denoted as E1, E2, and E3, with E1 having the lowest emissions, followed by E2, and then E3.

[0103] Quality Scenario (Core Objective: Maximize Job Coverage)

[0104] Solution and sorting: The computation node takes "job coverage" as the core optimization objective, outputs all feasible solutions that meet the constraints, and sorts them from largest to smallest "coverage" (the higher the coverage, the better).

[0105] Select the top three solutions: Select the top three solutions with the highest coverage as candidate solutions, denoted as R1, R2, and R3. Among them, R1 has the highest coverage, followed by R2, and then R3.

[0106] For each candidate solution selected, multi-objective parameters are extracted, including total operating cost, carbon emissions, and operational coverage.

[0107] Differentiated index weights are set for different scenarios. The multi-objective parameters of the top three candidate solutions in the sub-region are comprehensively processed in combination with the corresponding scenario, and the optimal solution coefficients of the top three candidate solutions in the sub-region are output.

[0108] The multi-objective parameters of the top three candidate solutions in the sub-region are comprehensively processed in conjunction with the corresponding scenario, specifically as follows:

[0109] After identifying the scenario to which the current sub-region belongs, extract the corresponding scenario's indicator weights, including cost weight, carbon emission weight, and operational coverage weight;

[0110] After normalizing the total operating cost, carbon emissions, and operational coverage for each candidate solution in the sub-region, the results are substituted into the formula. After weighting, the optimal coefficients of the candidate solutions are obtained; where A, B, and C represent the total operating cost, carbon emissions, and operational coverage of each candidate solution after normalization, respectively. These represent cost weight, carbon emission weight, and operational coverage weight, respectively.

[0111] For example, the cost weight, carbon emission weight, and operation coverage weight are set as follows for different scenarios:

[0112] In everyday scenarios: cost weight (0.6), carbon emission weight (0.2), and operational coverage weight (0.2).

[0113] Environmental scenario: cost weight (0.2), carbon emission weight (0.6), and operation coverage weight (0.2).

[0114] Quality scenario: cost weight (0.2), carbon emission weight (0.2), and operation coverage weight (0.6).

[0115] For each group of candidate solutions calculated for each sub-region, the candidate solution with the highest optimal solution coefficient is selected as the optimal scheduling solution under the constraints.

[0116] Taking "Sub-area B of a certain city's high-tech zone" as a specific case, this sub-area includes 5 garbage collection points (denoted as P1-P5, with garbage volumes of 2 tons, 1.5 tons, 3 tons, 2.5 tons, and 1 ton respectively), 3 operating vehicles (2 fuel-powered garbage trucks C1 / C2, rated capacity 10 tons, unit fuel consumption 12L / 100km; 1 new energy garbage truck C3, rated capacity 8 tons, no fuel consumption), and 2 operators (A / B, hourly wage 20 yuan, single operation ≤4 hours). The road network mileage is as follows: C1 to P1-P5 total mileage 25km, C2 to P1-P5 total mileage 22km, C3 to P1-P5 total mileage 28km. The details are illustrated using a typical daily scenario:

[0117] The particle swarm optimization algorithm is used to solve the problem using computing nodes, and three candidate solutions that satisfy the constraints (vehicle capacity ≤ 10 tons, personnel duration ≤ 4 hours, and cleaning time window 6:00-8:00) are output, denoted as S1, S2 and S3.

[0118] Calculate the total operating cost, carbon emissions, and operational coverage for the three candidate solutions:

[0119] After normalizing S1, S2, and S3, the normalization method adopts the linear proportional transformation method. Combined with the differentiated weights of daily scenarios, the optimal solution coefficient is calculated, and the solution with the highest optimal solution coefficient is selected as the optimal scheduling solution.

[0120] Supplement the calculation process of the optimal solution coefficients:

[0121] A linear proportional transformation method is used. Total operating cost (A) and carbon emissions (B) are cost-based indicators, and their normalized values ​​are obtained using a linear proportional transformation method. For example, after calculation:

[0122] S1 normalization parameters: (A: 0.764, B: 1.000, C: 1.000);

[0123] S2 normalization parameters: (A:1.000, B:0.837, C:1.000);

[0124] S3 normalization parameters: (A: 0.000, B: 0.000, C: 1.000;

[0125] Substituting the weights for everyday scenarios (cost 0.6, carbon emissions 0.2, coverage 0.2), calculate the optimal solution coefficient:

[0126] S1 optimal solution coefficient = 0.764 × 0.6 + 1.000 × 0.2 + 1.000 × 0.2 = 0.858;

[0127] S2 optimal solution coefficient = 1.000 × 0.6 + 0.837 × 0.2 + 1.000 × 0.2 = 0.967;

[0128] S3 optimal solution coefficient = 0.000 × 0.6 + 0.000 × 0.2 + 1.000 × 0.2 = 0.200;

[0129] After comparison, S2 has the highest optimization coefficient (0.967), so S2 is selected as the optimal scheduling solution for this sub-region.

[0130] If the constraints set are abandoned, each computing node will solve and sort the sub-regions based on the scene type and the optimization objective, select the top three candidate solutions and calculate the optimal solution coefficients.

[0131] Meanwhile, the degree of constraint violation of the top three candidate solutions is analyzed. First, the constraints marked as flexible are extracted from the set constraints as violable conditions. Violable conditions include task service time window, maximum continuous operation time of personnel, and multi-resource collaboration time difference.

[0132] The top three candidate solutions are compared with the constraints in terms of task service time window, maximum continuous working time of personnel, and multi-resource collaboration time difference to determine the task timeout, operation timeout, and collaboration timeout of the top three candidate solutions.

[0133] For example, in a task service time window, the constraint is that the job must be completed between 6:00 and 8:00 (i.e., the end time is ≤ 8:00). If the candidate solution is estimated to be later than 8:00 in the task service time window, the timeout duration is counted as the task timeout duration.

[0134] The task timeout duration, job timeout duration, and collaboration timeout duration of the top three candidate solutions are combined and processed to obtain the violation coefficients of the top three candidate solutions.

[0135] Task timeout duration: in minutes, indicating the duration by which the task ends beyond the preset time window;

[0136] Overtime period: in minutes, representing the portion of continuous work exceeding the maximum allowed time;

[0137] Collaboration timeout duration: in minutes, representing the portion of the collaborative operation time difference that exceeds the threshold;

[0138] That is, set the task timeout duration, job timeout duration, and collaboration timeout duration with their corresponding task weights, job weights, and collaboration weights, respectively. After normalizing the task timeout duration, job timeout duration, and collaboration timeout duration, multiply them by their respective set weights, and then sum them to obtain the violation coefficient.

[0139] The normalization process uses a linear scaling transformation method;

[0140] For example, the normalized task timeout, job timeout, and collaboration timeout are 0.75, 0.667, and 0.667, respectively.

[0141] The task weight is set to 0.4, the job weight to 0.3, and the collaboration weight to 0.3.

[0142] The violation coefficient is 0.75×0.4 + 0.667×0.3 + 0.667×0.3 = 0.710.

[0143] Establish a mapping rule between violation coefficients and adjustment coefficients. After converting the violation coefficients of candidate solutions into adjustment coefficients, multiply them with the corresponding optimal solution coefficients to obtain the limiting coefficients of the top three candidate solutions.

[0144] Additional explanation: This involves pre-setting coefficient intervals corresponding to the pre-defined violation coefficients, with each interval corresponding to an adjustment coefficient. The adjustment coefficient is limited to a range of 1.0-0.5. A lower violation coefficient indicates a higher probability of a match of 1.0. Example:

[0145] If the coefficient range is 0-0.2, then the corresponding adjustment coefficient is 1.0;

[0146] If the coefficient range is 0.2-0.5, then the corresponding adjustment coefficient is 0.9;

[0147] If the coefficient range is 0.5-0.8, then the corresponding adjustment coefficient is 0.7;

[0148] The coefficient range is (0.8-1.0), and the corresponding adjustment coefficient is 0.6.

[0149] If the coefficient range is greater than 1.0, the corresponding adjustment coefficient is 0.5.

[0150] For each group of candidate solutions calculated for each sub-region, the candidate solution with the highest constraint coefficient is selected as the optimal scheduling solution under the constraints.

[0151] The scheme integration module is used to integrate the optimal solutions of each sub-region to construct a global scheduling scheme, detect conflicts in the global scheduling scheme, screen the sub-regions with conflicts to construct an evaluation region set, quantify and comprehensively process the conflict degree of the sub-regions in the evaluation region set and output the conflict level, and adaptively adjust the global scheduling scheme based on the conflict level to output the final scheduling scheme; the conflict level division unit divides the conflicts into high conflict level and low conflict level.

[0152] The current approach compares resource allocation (such as a water truck being assigned to two sub-areas), work routes (such as two sweeping routes overlapping in the boundary area), and time planning (such as two sub-areas using the same waste transfer station at the same time).

[0153] Specifically:

[0154] Extract resource allocation data, job allocation data, and time planning data from the optimal solutions in each sub-region;

[0155] If the same vehicle or personnel is assigned to two or more sub-areas and the work periods overlap, it is considered a resource allocation conflict.

[0156] If the work routes of adjacent sub-regions overlap spatially on the map, it is determined to be a work assignment conflict;

[0157] If a shared facility is used by two or more sub-areas and the usage time interval is less than the preset time interval, it is considered a time planning conflict. The preset time interval is set by the facility's processing capacity limit. Shared facilities include transfer stations or water supply points.

[0158] The number of sub-regions within the statistical evaluation region set is calculated, and their proportion in the total number of sub-regions is used as the conflict proportion.

[0159] If the conflict ratio is higher than the preset conflict threshold, it is judged as a high conflict level, and the global scheduling scheme is marked as being sent to the management personnel for adjustment before the final scheduling scheme is output.

[0160] If the conflict ratio is lower than the preset conflict threshold ratio, it is judged as a low conflict level. Sub-regions with conflicts in the evaluation area set are classified according to the conflict type, which includes resource allocation conflict, task allocation conflict and time planning conflict.

[0161] Extract the optimal solution coefficients and their respective scenarios for each sub-region belonging to the same conflict type; additionally, if there are no limiting constraints in the process of solving the optimal solution, extract the limiting coefficients.

[0162] A basic assessment value is set for each different scenario; the basic assessment value for the environmental protection scenario is greater than that for the quality scenario, which is greater than that for the daily scenario. The range of the basic assessment value is limited to 1.159-1.269. The specific value is set by technical personnel and can be dynamically adjusted later.

[0163] After weighted logical processing of the optimal solution coefficients and basic evaluation values ​​of sub-regions belonging to the same conflict type, priority indices of different sub-regions under the same conflict type are obtained.

[0164] That is, set the coefficient weight and evaluation weight corresponding to the optimal solution coefficient and the basic evaluation value respectively, multiply the optimal solution coefficient and the basic evaluation value of the sub-region by the corresponding coefficient weight and evaluation weight, and then sum them to obtain the priority index of different sub-regions under the same conflict type;

[0165] To elaborate further, under the same conflict type, a higher optimal solution coefficient indicates a higher degree of performance of the optimal solution in the sub-region, and resources should be allocated to sub-regions with higher optimal solution performance. The basic evaluation value represents the scenario priority; the higher the scenario priority, the more resources should be allocated to sub-regions with higher scenario priority. The priority index of the two sets of values ​​reflects the priority of resource allocation to different sub-regions under the same conflict type.

[0166] For each sub-region under resource allocation conflict, the conflicting resources are allocated to the sub-region with the highest priority index first. The remaining sub-regions are removed from the resource allocation and the optimal solution is recalculated or the backup resources are called.

[0167] For sub-regions with conflicting task assignments, the overlapping route portions are assigned to the sub-region with the highest priority index first, and the overlapping route portions are removed from the remaining sub-regions.

[0168] As a supplementary explanation, since the overlapping route portion is assigned to the sub-region with the highest priority index, the overall results of various indicators are still higher than those of other sub-regions, so it is given priority in the allocation.

[0169] For sub-regions with time planning conflicts, the right to use shared facilities will be allocated to the sub-region with the highest priority index, and the remaining sub-regions will have their use delayed.

[0170] The solution output module outputs the final scheduling plan to the sanitation management platform in a visual format (such as electronic map and work list) to guide actual operations;

[0171] The specific values ​​of the various constraint thresholds (such as conflict threshold ratio, preset time interval of shared facilities) and weight parameters (such as scenario index weight, basic evaluation value, and various weights) involved in this system are examples set by those skilled in the art based on actual conditions.

[0172] The above formulas are all dimensionless calculations. Dimensionless calculations can be performed using various methods such as standardization, which will not be elaborated here. The formulas are derived from software simulations based on a large amount of collected data, and the preset parameters in the formulas can be set by those skilled in the art according to the actual situation.

[0173] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, ATA hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state ATA hard disk.

[0174] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0175] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0176] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0177] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0178] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0179] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable ATA hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0180] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A distributed collaborative scheduling system applied to intelligent sanitation across the entire region, characterized in that, Includes the following modules: Scene recognition module: Divides the entire area into several sub-regions and identifies the set scene to which each sub-region belongs, including daily scenes, environmental protection scenes and quality scenes; Distributed computing module: For each sub-region, a distributed computing architecture is used to construct and schedule computing tasks in combination with the relevant scenario. These tasks are then distributed to different computing nodes for parallel processing. Based on the optimization objectives of the current scenario, the optimal scheduling solution for each sub-region is obtained by selectively combining the set constraints. Solution integration module: integrates the optimal solutions of each sub-region to construct a global scheduling scheme, detects conflicts and filters conflicting sub-regions to construct an evaluation region set, quantifies the degree of conflict and outputs the conflict level, and adjusts the global scheme based on the conflict level and priority index; Solution Output Module: Outputs the final scheduling plan to the sanitation management platform in a visual format.

2. The distributed collaborative scheduling system for smart sanitation across the entire region as described in claim 1, characterized in that, The specific optimization goals for different scenarios are as follows: The optimization goal for everyday scenarios is to minimize total operating costs; The optimization objective for environmental protection scenarios is to minimize total carbon emissions; The optimization objective for the quality scenario is to maximize job coverage.

3. The distributed collaborative scheduling system for smart sanitation across the entire region as described in claim 2, characterized in that, If we choose to combine the set constraints, the specific solution process is as follows: For each sub-region scenario type, the optimization objective is used as the benchmark to solve and sort the solutions. The top three candidate solutions that meet all constraints are selected. The total operating cost, carbon emissions and operation coverage of each candidate solution are extracted. Based on the scenario-differentiated index weights, the parameters of the candidate solutions are normalized and weighted to calculate the optimal solution coefficient. The candidate solution with the highest optimal solution coefficient is selected as the optimal scheduling solution.

4. The distributed collaborative scheduling system for smart sanitation across the entire region as described in claim 2, characterized in that, If the constraints are abandoned, the specific solution process is as follows: Based on the optimization objective, the top three candidate solutions are solved and ranked, and the optimal solution coefficient is calculated. At the same time, the degree of violation of the flexible constraints by the candidate solutions is analyzed, and the timeout duration of tasks, jobs, and collaboration is determined. After normalization, the violation coefficient is calculated in combination with the weights. The adjustment coefficient is obtained based on the mapping rule between the violation coefficient and the adjustment coefficient. The optimal solution coefficient and the adjustment coefficient are multiplied to obtain the limit coefficient. The candidate solution with the highest limit coefficient is selected as the optimal scheduling solution.

5. The distributed collaborative scheduling system for smart sanitation across the entire region as described in claim 4, characterized in that, Flexible constraints include: Task service time window, maximum continuous working time for personnel, and time difference for multi-resource collaboration.

6. The distributed collaborative scheduling system for smart sanitation across the entire region as described in claim 1, characterized in that, The specific process of conflict detection is as follows: Resource allocation conflict: The same vehicle or personnel is assigned to two or more sub-areas and the work periods overlap; Job assignment conflict: The job routes of adjacent sub-regions overlap spatially on the map; Time scheduling conflict: The same shared facility is used by two or more sub-areas and the usage time interval is shorter than the preset time interval.

7. The distributed collaborative scheduling system for smart sanitation across the entire region as described in claim 1, characterized in that, The specific process for outputting the conflict level is as follows: The conflict ratio of the number of sub-regions within the evaluation region set to the total number of sub-regions is used. A conflict ratio higher than the preset conflict threshold is considered a high conflict level, while a ratio lower than the preset conflict threshold is considered a low conflict level.

8. The distributed collaborative scheduling system for smart sanitation across the entire region according to claim 7, characterized in that, The global scheduling scheme is adaptively adjusted based on the conflict level to output the final scheduling scheme, specifically: If the conflict level is high, the global scheduling scheme will be marked and sent to the administrator for adjustment before the final scheduling scheme is output. If the conflict level is low, the conflict sub-regions are classified according to the conflict type, the optimal solution coefficients and the scenarios to which the sub-regions of the same conflict type belong are extracted, the basic evaluation value of the scenario is set, and the optimal solution coefficients and the basic evaluation value are weighted according to the preset weights to calculate the priority index. Resource allocation conflict: Prioritize allocating conflicting resources to the sub-region with the highest priority index, and re-solve the remaining sub-regions; Job assignment conflict: Prioritize assigning overlapping routes to the sub-region with the highest priority index, and remove overlapping routes from the remaining sub-regions; Time scheduling conflict: Prioritize allocating the right to use shared facilities to the sub-area with the highest priority index, and postpone the use of the remaining sub-areas.

Images