Intelligent patrol task assignment method and system

Abstract

The application provides a kind of intelligent patrol task allocation method and system, comprising: step S1, obtaining multiple data;Step S2, define the actual distance of road network as clustering distance measure, according to the density of task point dynamically configure the parameter of spatial clustering algorithm, execute spatial clustering algorithm to generate task package, and calculate the comprehensive priority weight of each task package;Step S3, construct target optimization function, and define constraint condition, output allocation result;Step S4, real-time monitoring task execution progress, personnel state and new task access, when triggering redistribution condition, construct local optimization pool, update task execution list.The application can effectively overcome the disadvantages of fixed area and simple rule allocation, realize the global optimization configuration of multi-dimensional resources such as personnel, task, environment and the like through multi-factor fusion decision, so that it has the functions of multi-factor fusion, dynamic response and self-regulation, effectively improves the efficiency, continuity and reliability of patrol.

Description

Technical Field

[0001] This invention relates to the intersection of computer application technology and smart city management, and in particular to a method for allocating intelligent patrol tasks based on a combination of artificial intelligence, geographic information system and optimization algorithm, and further to an allocation system that uses the intelligent patrol task allocation method and system. Background Technology

[0002] With the continuous acceleration of urbanization, the scale of municipal infrastructure is becoming increasingly large, and the complexity and importance of its daily inspection and maintenance work have also increased significantly. However, the current allocation and management models for daily inspection tasks of facilities such as roads, manhole covers, and streetlights still have many bottlenecks, mainly reflected in the following two typical solutions.

[0003] First, a fixed-area manual allocation model.

[0004] This model relies on the historical experience of managers, statically dividing jurisdictional areas and assigning them to specific patrol personnel. While responsibilities are clearly defined, its inherent technical flaws are significant: route duplication and inefficiency; patrol personnel are confined to fixed boundaries, leading to overlapping coverage at the boundaries of adjacent areas and potential blind spots in other areas, resulting in a significant waste of human resources and time; lack of flexibility and adaptability; this model cannot effectively respond to unexpected tasks (such as emergency complaints from the public) or dynamic changes in personnel status (such as leave or resignation). When a regional supervisor is absent, their tasks are difficult to redistribute quickly and reasonably, resulting in gaps in patrol work and posing a threat to urban safety; high reliance on experience; the rationality of the entire task allocation system depends heavily on the personal experience of specific managers, lacking objective and scientific data support and decision-making models. Once core managers leave their posts, the efficiency of the entire dispatch system will drop sharply.

[0005] Second, a semi-automatic allocation mode based on simple rules.

[0006] Currently, some organizations have introduced basic information management tools to achieve electronic task assignment. This model typically employs simple decision-making logic; for example, it assigns tasks to the nearest patrol personnel based solely on the task's latitude and longitude information, following the "proximity principle." While this technical solution improves dispatch speed, its limitations prevent it from addressing fundamental problems: It ignores multi-dimensional constraints, treating only "geographical distance" as a single decision variable, completely disregarding crucial constraints such as the real-time location dynamics of patrol personnel, the impact of current traffic conditions on mobility efficiency, the matching relationship between personnel's professional skills and task types, and the balance of real-time workload among personnel; it fails to achieve global optimization, employing a proximity principle that only achieves local optimization, easily leading to overburdened patrol personnel in high-density task areas while those in low-density areas remain idle, failing to achieve optimal allocation of all resources and tasks at the system level, resulting in overall low efficiency; and it has poor dynamic response capabilities, lacking effective feedback and replanning mechanisms, unable to dynamically recover and redistribute tasks based on real-time changes in the situation (such as sudden traffic congestion on a road segment or sudden equipment failure of a patrol personnel), resulting in insufficient system robustness.

[0007] Clearly, neither of the two existing patrol allocation technologies can achieve multi-factor fusion, dynamic response, and self-adjustment. The efficiency, continuity, and reliability of patrols are poor, and they cannot adequately meet actual patrol needs. Summary of the Invention

[0008] The technical problem this invention aims to solve is to provide an intelligent patrol task allocation method that overcomes the inherent drawbacks of fixed-area and simple rule-based allocation. Through multi-factor fusion decision-making, it achieves global optimized allocation of multi-dimensional resources such as personnel, tasks, and environment, enabling it to possess multi-factor fusion, dynamic response, and self-adjustment functions, thereby improving the efficiency, continuity, and reliability of patrols. Furthermore, this invention provides an allocation system employing this intelligent patrol task allocation method and system.

[0009] To address this, the present invention provides a method for allocating intelligent patrol tasks, comprising the following steps:

[0010] Step S1: Obtain multi-dimensional data, which includes data on the set of task points to be inspected, data on the set of inspection personnel, and real-time traffic data.

[0011] Step S2: First, define the actual distance of the road network as the clustering distance metric. Then, dynamically configure the parameters of the spatial clustering algorithm according to the task point density, execute the spatial clustering algorithm to generate task packages, and calculate the comprehensive priority weight of each task package.

[0012] Step S3: Construct the objective optimization function, define the constraints, and output the allocation results;

[0013] Step S4: Monitor task execution progress, personnel status, and new task access in real time. When a redistribution condition is triggered, build a local optimization pool and update the task execution list.

[0014] A further improvement of the present invention is that step S1 includes the following sub-steps:

[0015] Step S101: Obtain the set of task points to be inspected, P={p1,p2,...,p...}, from the management platform. n}, where n represents the total number of task points to be inspected. The data structure of each task point includes a unique task identifier, geographical latitude and longitude information, task type, priority, required skills, and task release time.

[0016] Step S102: The management platform and mobile terminal obtain the set of patrol personnel R={r1,r2, ...,r m}, m represents the total number of patrol personnel. The status information of each patrol personnel includes a unique personnel identifier, real-time location information, task type, priority, required skills, and task release time.

[0017] Step S103: Call the map open platform to obtain real-time traffic data.

[0018] A further improvement of the present invention is that step S2 includes the following sub-steps:

[0019] Step S201: Use the actual distance of the road network as the clustering distance metric, and apply the formula... Calculate task points With mission points Actual distance of road network between ,in, This represents the distance calculation function. and These represent the indices of different task points. and Indicates task point Latitude and longitude data, and Indicates task point Latitude and longitude data;

[0020] Step S202: Dynamically configure the parameters of the spatial clustering algorithm according to the task point density. The parameters of the spatial clustering algorithm include the neighborhood radius Eps and the minimum number of points MinPts.

[0021] Step S203: Execute the spatial clustering algorithm to aggregate discrete task points into a task package set. ;

[0022] Step S204, using the formula Calculate the i-th task package Comprehensive priority weight ,in, Represents a single task point The preset priority; i represents the index of the task package.

[0023] A further improvement of the present invention is that, in step S202, the task point density ρ is calculated using the formula ρ=n / S, where n represents the total number of task points to be inspected within the jurisdiction, and S represents the area of ​​the jurisdiction; when ρ≥10 points / square kilometer, the neighborhood radius Eps takes the lowest value of the preset radius range; when 5≤ρ<10 points / square kilometer, the neighborhood radius Eps takes the middle value of the preset radius range; when ρ<5 points / square kilometer, the neighborhood radius Eps takes the highest value of the preset radius range.

[0024] A further improvement of the present invention is that, in step S202, the range of the minimum number of points MinPts is preset, and the relationship between the minimum number of points MinPts and the task point density ρ is set to be positively correlated.

[0025] A further improvement of the present invention is that step S3 includes the following sub-steps:

[0026] Step S301, using the formula Construct the objective function for the global total allocation cost Where k represents the total number of task packages, i represents the index of the task package; m represents the total number of patrol personnel, and j represents the index of the patrol personnel; , and These represent the time cost weight, skill matching weight, and load balancing weight, respectively. Indicates patrol personnel Complete the task package The estimated total time; Indicates skill matching degree; Indicates the load balancing factor; Represent decision variables; Indicates task package Comprehensive priority weight;

[0027] Step S302 defines the following constraints: each task package must be assigned to only one patrol officer, and the number of task packages assigned to each patrol officer shall not exceed their maximum capacity. And skill mismatches should be prevented from being assigned;

[0028] Step S303, using the formula Output the optimal allocation matrix .

[0029] A further improvement of the present invention is that, in step S302, the formula is used... Define the constraint that each task package must be assigned to only one patrol officer. This means that the condition is true for all task packages.

[0030] Through formula Define the number of task packages assigned to each patrol officer as not exceeding their maximum capacity. This constraint, in which... This indicates that the agreement applies to all patrol personnel;

[0031] Through formula Define a constraint that prohibits assignment when skills do not match, where, This indicates that the agreement applies to all task packages and all patrol personnel.

[0032] A further improvement of the present invention is that, between steps S302 and S303, a genetic algorithm parameter configuration step is included, which optimizes the objective function by setting the genetic algorithm parameters. The model; the genetic algorithm parameters include population size, crossover probability, mutation probability, termination generation, and fitness function.

[0033] A further improvement of the present invention is that step S4 includes the following sub-steps:

[0034] Step S401: Real-time monitoring. When any reassignment condition is triggered, the reassignment process is started. The reassignment conditions include the deviation between the actual start time and the planned time of the task package exceeding a preset threshold Δt, the patrol personnel actively reporting an anomaly through a mobile terminal, and the system receiving a new high-priority task.

[0035] Step S402: Construct a local optimization pool using a new task package set C' and a new personnel set R'. The new task package set C' includes task packages that have not yet started execution, task packages that have started but are not 50% complete, and newly added emergency task packages. The new personnel set R' includes patrol personnel who are on duty and have not reached their maximum load, as well as patrol personnel who were originally assigned to affected task packages.

[0036] Step S403, reuse the objective function from step S301. Calculate the cost of the new task package set C' and the new personnel set R', output the locally optimal allocation scheme, and update the task execution list.

[0037] The present invention also provides a method for allocating intelligent patrol tasks, which employs the intelligent patrol task allocation method described above and includes:

[0038] The multi-source data acquisition module acquires multi-source data through the management platform, mobile terminals, and map open platform;

[0039] The task package clustering generation module first defines the actual distance of the road network as the clustering distance metric, then dynamically configures the parameters of the spatial clustering algorithm according to the density of task points, executes the spatial clustering algorithm to generate task packages, and calculates the comprehensive priority weight of each task package.

[0040] The multi-factor matching and optimal allocation module constructs the objective optimization function, defines constraints, and outputs the allocation results.

[0041] The dynamic monitoring and redistribution module monitors task execution progress, personnel status, and new task access in real time. When redistribution conditions are triggered, a local optimization pool is built and the task execution list is updated.

[0042] Compared with existing technologies, the beneficial effects of this invention are as follows: First, multi-dimensional data is acquired. Then, the actual distance of the road network is defined as the clustering distance metric. The parameters of the spatial clustering algorithm are dynamically configured according to the task point density. The spatial clustering algorithm is executed to generate task packages, and the comprehensive priority weight of each task package is calculated. Next, an objective optimization function is constructed, constraints are defined, and the allocation results are output. Finally, the task execution progress, personnel status, and new task access are monitored in real time. When a reassignment condition is triggered, a local optimization pool is constructed, and the task execution list is updated. Therefore, this invention can effectively overcome the drawbacks of fixed area and simple rule allocation. Through multi-factor fusion decision-making, it achieves global optimization configuration of multi-dimensional resources such as personnel, tasks, and environment, enabling it to have multi-factor fusion, dynamic response, and self-adjustment functions, effectively improving the efficiency, continuity, and reliability of patrols. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the workflow of one embodiment of the present invention. Detailed Implementation

[0044] In the description of this invention, if directional descriptions are involved, such as "up," "down," "front," "back," "left," "right," etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, it is only for the convenience of describing the invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. If a technical feature is referred to as "set," "fixed," "connected," or "installed" on another technical feature, it can be directly set, fixed, or connected to the other technical feature, or it can be indirectly set, fixed, connected, or installed on the other technical feature.

[0045] In the description of this invention, the term "several" means one or more; the term "multiple" means two or more; the terms "greater than," "less than," and "exceeding" are all understood to exclude the stated number; and the terms "above," "below," and "within" are all understood to include the stated number. The terms "first," "second," etc., are understood to be used only to distinguish identical or similar technical feature names, and should not be construed as implying / indicating the relative importance of the technical features, the number of technical features, or the sequential relationship between the technical features.

[0046] The preferred embodiments of the present invention will now be described in further detail with reference to the accompanying drawings.

[0047] like Figure 1 As shown, this embodiment provides a method for allocating intelligent patrol tasks, including the following steps:

[0048] Step S1: Obtain multi-dimensional data, which includes data on the set of task points to be inspected, data on the set of inspection personnel, and real-time traffic data.

[0049] Step S2: First, define the actual distance of the road network as the clustering distance metric. Then, dynamically configure the parameters of the spatial clustering algorithm according to the task point density, execute the spatial clustering algorithm to generate task packages, and calculate the comprehensive priority weight of each task package.

[0050] Step S3: Construct the objective optimization function, define the constraints, and output the allocation results;

[0051] Step S4: Monitor task execution progress, personnel status, and new task access in real time. When a redistribution condition is triggered, build a local optimization pool and update the task execution list.

[0052] In this embodiment, step S1 is used to acquire multi-source data. Preferably, step S1 includes the following sub-steps:

[0053] Step S101: Obtain the set of task points to be inspected, P={p1,p2,...,p...}, from the management platform. n}, where n represents the total number of task points to be inspected. The data structure of each task point includes a unique task identifier, geographical latitude and longitude information, task type, priority, required skills, and task release time.

[0054] Step S102: The management platform and mobile terminal obtain the set of patrol personnel R={r1,r2, ...,r m}, m represents the total number of patrol personnel. The status information of each patrol personnel includes a unique personnel identifier, real-time location information, task type, priority, required skills, and task release time.

[0055] Step S103: Call the map open platform to obtain real-time traffic data.

[0056] In step S101 of this embodiment, each task point The preferred definition of the data structure is as follows:

[0057] Parameter name Data types illustrate Example ID String Unique Task Identifier "TS-20240501-001" Lon / Lat floating-point numbers Geographic latitude and longitude (accurate to 0.0001°) 116.403874, 39.914885 Type Enumeration value Task type Road surface damage / Missing manhole covers / Streetlight malfunction / Pipeline leaks Level Integer Priority (1-5, 5 being the highest) 4 (Higher priority) Skill_required gather Required skills (multiple selections allowed) {"Routine", "High-altitude operations"} RublishTime Timestamp Task release time 1714521600（2024-05-01 00:00:00）

[0058] In step S102 of this embodiment, each patrol personnel The preferred definition of the data structure is as follows:

[0059] Parameter name Data types illustrate Example ID String Unique Personnel Identifier "RP-20240105-003" Lon / Lat floating-point numbers Real-time GPS location (accuracy to 0.0001°) 116.405874, 39.913885 Type Enumeration value Task type Road surface damage / Missing manhole covers / Streetlight malfunction / Pipeline leaks Level Integer Priority (1-5, 5 being the highest) 4 (Higher priority) Skill_required gather Required skills (multiple selections allowed) {"Routine", "High-altitude operations"} RublishTime Timestamp Task release time 1714521600（2024-05-01 00:00:00）

[0060] In step S103 of this embodiment, it is preferable to obtain real-time traffic data by calling the "Route Planning API" (such as version V4.0) of Gaode Map Open Platform. The request parameters and returned data are as follows: the request parameters include: starting point latitude and longitude (Lon_s, Lat_s), ending point latitude and longitude (Lon_e, Lat_e), mode of travel (walking / driving / cycling), and real-time traffic conditions. The returned data includes: travel distance Droad (meters), estimated travel time Ttrave (seconds), and traffic congestion level C. The traffic congestion level C includes levels 0-5 by default, where level 0 is smooth and level 5 is severely congested.

[0061] In this embodiment, step S2 is the task package clustering generation step, which is used to aggregate discrete task points into geographically compact "task packages" to improve the efficiency of patrol paths. It is preferably divided into steps S201-S204, using the improved DBSCAN spatial clustering algorithm.

[0062] Preferably, step S2 in this embodiment includes the following sub-steps:

[0063] Step S201: Use the actual distance of the road network as the clustering distance metric, and apply the formula... Calculate task points With mission points Actual distance of road network between The unit is meters by default; among which, This represents the distance calculation function, preferably the distance calculation function of the Amap route planning API, with the task point as the input. With mission points The latitude and longitude of these two points are output as the actual driving distance; and These represent the indices of different task points. and Indicates task point Latitude and longitude data, and Indicates task point Latitude and longitude data;

[0064] Step S202: Dynamically configure the parameters of the spatial clustering algorithm according to the task point density. The parameters of the spatial clustering algorithm include the neighborhood radius Eps and the minimum number of points MinPts.

[0065] Step S203: Execute the spatial clustering algorithm to aggregate discrete task points into a task package set. ;

[0066] Step S204, using the formula Calculate the i-th task package Comprehensive priority weight ,in, Represents a single task point The preset priority; i represents the index of the task package.

[0067] In this embodiment, step S201 is used to define the distance metric method. Instead of using the traditional Euclidean distance, the actual distance of the road network is used as the clustering distance metric to better meet the actual inspection needs.

[0068] Step S202 in this embodiment is used to dynamically configure algorithm parameters. Preferably, in step S202 of this embodiment, the task point density ρ is calculated using the formula ρ=n / S, where n represents the total number of task points to be inspected within the jurisdiction, and S represents the area of ​​the jurisdiction (the unit is square kilometers by default). When ρ≥10 points / square kilometer, the neighborhood radius Eps takes the lowest value of the preset radius range; when 5≤ρ<10 points / square kilometer, the neighborhood radius Eps takes the middle value of the preset radius range; when ρ<5 points / square kilometer, the neighborhood radius Eps takes the highest value of the preset radius range. The preset radius range refers to the range of values ​​for the neighborhood radius Eps set in advance, which can be set according to the actual situation, and the default is 300-800 meters.

[0069] In step S202 of this embodiment, the range of the minimum number of points MinPts is preset, and the relationship between the minimum number of points MinPts and the task point density ρ is set to be positively correlated, that is, the higher the task point density ρ, the larger the value of the minimum number of points MinPts. The following example illustrates a dynamic configuration example of the algorithm parameters.

[0070] Parameter name data illustrate Example Neighborhood radius Eps 300-800 meters When the task point density ρ ≥ 10 points / km², use 300 meters; when 5 ≤ ρ < 10, use 500 meters; when ρ < 5, use 800 meters. 500 meters Minimum points MinPts 2-5 The higher the task point density, the larger the value. 3

[0071] In step S203 of this embodiment, the DBSCAN spatial clustering algorithm is preferably used. This spatial clustering algorithm is existing technology, so it will not be described in detail here.

[0072] Step S204 in this embodiment is used to implement the task package. The comprehensive priority calculation formula is as follows: Among the various parameters, The data type is a floating-point number, representing the overall priority weight of the i-th task package (core output result). The default value range is 1-5, which is consistent with the value range of the task point priority Level, such as 3.2 and 4.5. The data type is an integer, representing the sum of priorities of all task points within the i-th task package. Assume the task package... If there are 3 task points with priorities of 4, 5, and 3, then the sum is 4 + 5 + 3 = 12. Represents a single task point The preset priority is used to reflect the urgency and importance of the task, such as 1-5 levels, where 1 = normal, 2 = low, 3 = medium, 4 = high, and 5 = urgent; i represents the index of the task package. Represents a task point object, a task package. Individual inspection task points, such as "missing manhole covers" or "streetlight malfunctions." Task package The set used to represent the i-th task package is preferably a geographically compact task set generated by the DBSCAN spatial clustering algorithm in step S203. For certificates, it represents the total number of task points contained in the i-th task package. The absolute value symbol here represents "count". The value range is ≥1, such as 2 or 5. The minimum number of points MinPts during clustering is ≥2. Noise points are 1 when they are grouped separately.

[0073] Step S3 in this embodiment is a multi-factor matching and optimal allocation step, which is a core decision-making step. By constructing and solving a multi-objective optimization model, the globally optimal allocation is achieved. Preferably, step S3 in this embodiment includes the following steps S301 to S303.

[0074] Step S301, using the formula Construct the objective function for the global total allocation cost The objective function To minimize the total cost. Here, k represents the total number of task packages, and i represents the index of the task package. m represents the total number of patrol personnel, and j represents the index of the patrol personnel. ; , and These represent the time cost weight, skill matching weight, and load balancing weight, respectively. Indicates patrol personnel Complete the task package The estimated total time; Indicates skill matching degree; Indicates the load balancing factor; Represent decision variables; Indicates task package The overall priority weight.

[0075] More specifically, the objective function The parameters are explained below:

[0076] Parameter name Data types Definitions Z floating-point numbers Total global allocation cost (unitless, used for relative comparison) k Integer Total number of task packages m Integer Total number of patrol personnel α, β, γ floating-point numbers The weighting coefficients range from [0,1] and satisfy α+β+γ=1. Optimal values ​​are: α=0.5 (time cost weight), β=0.3 (skill matching weight), and γ=0.2 (load balancing weight). Integer personnel Complete the task package The estimated total time (in minutes) is calculated using the following formula: Integer personnel Complete the task package The estimated total time (in minutes) is calculated using TSP route planning based on real-time traffic conditions. Integer Task package The formula for the task time (in minutes) is: The default time for each task is 2 minutes, which can be adjusted according to the task type. Boolean value Skill matching express: (Personnel skills cover the requirements of the task package) Indicates a mismatch (at this time) (Tending to infinity, achieving hard constraints) Integer Load balancing factor, formula: This indicates the assigned patrol personnel. load Boolean value Decision variables Indicates task package Assigned to personnel , Indicates no allocation floating-point numbers Task package Comprehensive priority weight

[0077] In step S302 of this embodiment, the defined constraints include: each task package must be assigned to only one patrol officer, and the number of task packages assigned to each patrol officer does not exceed their maximum capacity. And skill mismatches should be prevented from being assigned.

[0078] More specifically, in step S302 of this embodiment, the formula is used... Define the constraint that each task package must be assigned to only one patrol officer. This represents the summation operator, which iterates through "all patrol personnel" and sums the results, from the 1st patrol personnel to the mth patrol personnel. This means that the condition applies to all task packages.

[0079] Through formula Define the number of task packages assigned to each patrol officer as not exceeding their maximum capacity. This constraint, in which... This applies to all patrol personnel. Maximum capacity. This represents the maximum task package capacity of the j-th patrol personnel. It is set according to the patrol personnel's skill proficiency, working hours, and configuration settings. The default value is 5, indicating strong skills; it can also be set to 6-7, indicating weak skills; or it can be set to 3-4, indicating part-time work.

[0080] Through formula Define a constraint that prohibits assignment when skills do not match, where, This indicates that the agreement applies to all task packages and all patrol personnel.

[0081] Step S303, using the formula Output the optimal allocation matrix Preferably, a task execution list is generated for each patrol member. The task execution list includes: task package ID, task point coordinates, patrol order (TSP path planning result), estimated start / end time, and required skills.

[0082] Preferably, in this embodiment, a genetic algorithm parameter configuration step is included between steps S302 and S303, whereby the objective function is optimized by setting the genetic algorithm parameters. The model; the genetic algorithm parameters include population size, crossover probability, mutation probability, termination generation, and fitness function. Preferably, the parameters are set as follows:

[0083] Parameter name Range of values Preferred value illustrate Population size 50-200 100 Number of chromosomes in each generation Crossover probability 0.7-0.9 0.8 The probability of chromosome crossing over Mutation probability 0.01-0.1 0.05 Probability of chromosome mutation operations Termination Algebra 100-500 200 Maximum number of iterations required for algorithm to terminate fitness function --- Fit(Z) = 1 / 1 + Z The higher the fitness value, the better the corresponding allocation scheme. Parameter name Range of values Preferred value illustrate

[0084] Step S4 in this embodiment is used to implement dynamic monitoring and redistribution to ensure the robustness of the system.

[0085] Preferably, step S4 in this embodiment includes the following sub-steps:

[0086] Step S401: Real-time monitoring. When any reassignment condition is triggered, the reassignment process is started. The reassignment conditions include the deviation between the actual start time and the planned time of the task package exceeding a preset threshold Δt, the patrol personnel actively reporting an anomaly through a mobile terminal, and the system receiving a new high-priority task.

[0087] Step S402: Construct a local optimization pool using a new task package set C' and a new personnel set R'. The new task package set C' includes task packages that have not yet started execution, task packages that have started but are not 50% complete, and newly added emergency task packages. The new personnel set R' includes patrol personnel whose status is "on duty" and who have not reached their maximum load, as well as patrol personnel whose original assignments were affected by the task packages.

[0088] Step S403, reuse the objective function from step S301. Calculate the cost of the new task package set C' and the new personnel set R', output the locally optimal allocation scheme, and update the task execution list.

[0089] In step S401 of this embodiment, the preferred start-up redistribution process is as follows:

[0090] Triggering conditions Detailed Explanation Threshold setting Time Deviation Trigger The actual start time of the task package deviates from the planned time by more than the threshold Δt. Level 5 task: Δt = 5 minutes; Level 4 task: Δt = 10 minutes; Level 1-3 tasks: Δt = 15 minutes Error reporting triggered Inspectors proactively report anomalies via the app. Equipment malfunction, physical discomfort, unexpected situations on site, etc. Emergency mission triggered The system received a new high-priority task. New task Level ≥ 4

[0091] Preferably, in step S403 of this embodiment, the genetic algorithm parameters are adjusted as follows: population size = 50, termination generation = 100, to improve the calculation speed; output the local optimal allocation scheme, update the task list of affected personnel, and keep the execution plan of unaffected tasks unchanged.

[0092] This embodiment also provides a method for allocating intelligent patrol tasks, which employs the intelligent patrol task allocation method described above and includes:

[0093] The multi-source data acquisition module acquires multi-source data through the management platform, mobile terminals, and map open platform;

[0094] The task package clustering generation module first defines the actual distance of the road network as the clustering distance metric, then dynamically configures the parameters of the spatial clustering algorithm according to the density of task points, executes the spatial clustering algorithm to generate task packages, and calculates the comprehensive priority weight of each task package.

[0095] The multi-factor matching and optimal allocation module constructs the objective optimization function, defines constraints, and outputs the allocation results.

[0096] The dynamic monitoring and redistribution module monitors task execution progress, personnel status, and new task access in real time. When redistribution conditions are triggered, a local optimization pool is built and the task execution list is updated.

[0097] In summary, this embodiment first acquires multi-dimensional data, then defines the actual distance of the road network as the clustering distance metric, and dynamically configures the parameters of the spatial clustering algorithm based on the task point density. The spatial clustering algorithm is then executed to generate task packages, and the comprehensive priority weight of each task package is calculated. Next, an objective optimization function is constructed, constraints are defined, and the allocation results are output. Finally, the task execution progress, personnel status, and new task access are monitored in real time. When a reassignment condition is triggered, a local optimization pool is constructed, and the task execution list is updated. This embodiment effectively overcomes the drawbacks of fixed-area and simple rule-based allocation. Through multi-factor fusion decision-making, it achieves global optimization and configuration of multi-dimensional resources such as personnel, tasks, and environment, enabling it to possess multi-factor fusion, dynamic response, and self-adjustment functions, effectively improving the efficiency, continuity, and reliability of patrols.

[0098] More specifically, this embodiment overcomes the mechanical and inefficient problems of fixed area allocation mode, breaks through the limitations of static area division in the prior art, avoids repetitive patrol routes and coverage blind spots, and reduces the system's over-reliance on the experience of specific management personnel; it overcomes the one-sidedness and local optimization problems of simple rule allocation mode, and integrates multiple factors such as geographical distance, real-time personnel location, real-time traffic conditions, personnel professional skills, and workload balancing during task allocation, thereby achieving global optimal resource allocation rather than local proximity principle; it improves the dynamic adaptability and robustness of the patrol task allocation system, enabling the system to perceive changes on site in real time (such as sudden tasks, personnel status changes, and traffic congestion), and has the ability to quickly and automatically replan and dynamically adjust unfinished tasks, ensuring the continuity, integrity, and reliability of patrol work.

[0099] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for allocating intelligent patrol tasks, characterized in that, Includes the following steps: Step S1: Obtain multi-dimensional data, which includes data on the set of task points to be inspected, data on the set of inspection personnel, and real-time traffic data. Step S2: First, define the actual distance of the road network as the clustering distance metric. Then, dynamically configure the parameters of the spatial clustering algorithm according to the task point density, execute the spatial clustering algorithm to generate task packages, and calculate the comprehensive priority weight of each task package. Step S3: Construct the objective optimization function, define the constraints, and output the allocation results; Step S4: Monitor task execution progress, personnel status, and new task access in real time. When a redistribution condition is triggered, build a local optimization pool and update the task execution list.

2. The method for allocating intelligent patrol tasks according to claim 1, characterized in that, Step S1 includes the following sub-steps: Step S101: Obtain the set of task points to be inspected, P={p1,p2,...,p...}, from the management platform. n }, where n represents the total number of task points to be inspected. The data structure of each task point includes a unique task identifier, geographical latitude and longitude information, task type, priority, required skills, and task release time. Step S102: The management platform and mobile terminal obtain the set of patrol personnel R={r1,r2, ...,r m }, m represents the total number of patrol personnel. The status information of each patrol personnel includes a unique personnel identifier, real-time location information, task type, priority, required skills, and task release time. Step S103: Call the map open platform to obtain real-time traffic data.

3. The method for allocating intelligent patrol tasks according to claim 1 or 2, characterized in that, Step S2 includes the following sub-steps: Step S201: Use the actual distance of the road network as the clustering distance metric, and apply the formula... Calculate task points With mission points Actual distance of road network between ,in, This represents the distance calculation function. and These represent the indices of different task points. and Indicates task point Latitude and longitude data, and Indicates task point Latitude and longitude data; Step S202: Dynamically configure the parameters of the spatial clustering algorithm according to the task point density. The parameters of the spatial clustering algorithm include the neighborhood radius Eps and the minimum number of points MinPts. Step S203: Execute the spatial clustering algorithm to aggregate discrete task points into a task package set. ; Step S204, using the formula Calculate the i-th task package Comprehensive priority weight ,in, Represents a single task point The preset priority; i represents the index of the task package.

4. The method for allocating intelligent patrol tasks according to claim 3, characterized in that, In step S202, the task point density ρ is calculated using the formula ρ=n / S, where n represents the total number of task points to be inspected within the jurisdiction, and S represents the area of ​​the jurisdiction. When ρ≥10 points / km², the neighborhood radius Eps takes the lowest value within the preset radius range. When 5≤ρ<10 points / km², the neighborhood radius Eps takes the middle value within the preset radius range. When ρ<5 points / km², the neighborhood radius Eps takes the highest value within the preset radius range.

5. The method for allocating intelligent patrol tasks according to claim 4, characterized in that, In step S202, the range of the minimum number of points MinPts is preset, and the relationship between the minimum number of points MinPts and the task point density ρ is set to be positively correlated.

6. The method for allocating intelligent patrol tasks according to claim 1 or 2, characterized in that, Step S3 includes the following sub-steps: Step S301, using the formula Construct the objective function for the global total allocation cost Where k represents the total number of task packages, i represents the index of the task package; m represents the total number of patrol personnel, and j represents the index of the patrol personnel; , and These represent the time cost weight, skill matching weight, and load balancing weight, respectively. Indicates patrol personnel Complete the task package The estimated total time; Indicates skill matching degree; Indicates the load balancing factor; Represent decision variables; Indicates task package Comprehensive priority weight; Step S302 defines the following constraints: each task package must be assigned to only one patrol officer, and the number of task packages assigned to each patrol officer shall not exceed their maximum capacity. And skill mismatches should be prevented from being assigned; Step S303, using the formula Output the optimal allocation matrix .

7. The method for allocating intelligent patrol tasks according to claim 6, characterized in that, In step S302, the formula is used. Define the constraint that each task package must be assigned to only one patrol officer. This means that the condition is true for all task packages. Through formula Define the number of task packages assigned to each patrol officer as not exceeding their maximum capacity. This constraint, in which... This indicates that the agreement applies to all patrol personnel; Through formula Define a constraint that prohibits assignment when skills do not match, where, This indicates that the agreement applies to all task packages and all patrol personnel.

8. The method for allocating intelligent patrol tasks according to claim 6, characterized in that, Between steps S302 and S303, a genetic algorithm parameter configuration step is also included, which optimizes the objective function by setting the genetic algorithm parameters. The model; the genetic algorithm parameters include population size, crossover probability, mutation probability, termination generation, and fitness function.

9. The method for allocating intelligent patrol tasks according to claim 6, characterized in that, Step S4 includes the following sub-steps: Step S401: Real-time monitoring. When any reassignment condition is triggered, the reassignment process is started. The reassignment conditions include the deviation between the actual start time and the planned time of the task package exceeding a preset threshold Δt, the patrol personnel actively reporting an anomaly through a mobile terminal, and the system receiving a new high-priority task. Step S402: Construct a local optimization pool using a new task package set C' and a new personnel set R'. The new task package set C' includes task packages that have not yet started execution, task packages that have started but are not 50% complete, and newly added emergency task packages. The new personnel set R' includes patrol personnel who are on duty and have not reached their maximum load, as well as patrol personnel who were originally assigned to affected task packages. Step S403, reuse the objective function from step S301. Calculate the cost of the new task package set C' and the new personnel set R', output the locally optimal allocation scheme, and update the task execution list.

10. A method for allocating intelligent patrol tasks, characterized in that, The method for allocating intelligent patrol tasks as described in any one of claims 1 to 9 is adopted, and includes: The multi-source data acquisition module acquires multi-source data through the management platform, mobile terminals, and map open platform; The task package clustering generation module first defines the actual distance of the road network as the clustering distance metric, then dynamically configures the parameters of the spatial clustering algorithm according to the density of task points, executes the spatial clustering algorithm to generate task packages, and calculates the comprehensive priority weight of each task package. The multi-factor matching and optimal allocation module constructs the objective optimization function, defines constraints, and outputs the allocation results. The dynamic monitoring and redistribution module monitors task execution progress, personnel status, and new task access in real time. When redistribution conditions are triggered, a local optimization pool is built and the task execution list is updated.

Images