A cross-regional municipal road maintenance resource scheduling method and system

Abstract

The application discloses a kind of municipal road maintenance resource scheduling methods and systems across region, the scheduling method includes the following steps: collecting each regional road disease data, maintenance history record and real-time traffic flow data;Using analytic hierarchy process to construct judgment matrix, calculate the weight of road factor by square root method, according to the weight of the basis data after quantization, calculate the score of each regional maintenance urgency, divide regional maintenance grade;According to the maintenance grade of each region, divide the maintenance group with different priority;Statistical existing maintenance resources in each region, combined with the division result of maintenance group, predict the resource demand of each region based on historical maintenance resource consumption data;For resource shortage area, adopt greedy algorithm to select scheduling path, and develop a preliminary cross-regional resource scheduling scheme;Construct multi-objective cost evaluation algorithm, and iteratively optimize the scheme by combining particle swarm optimization algorithm.The application realizes efficient, accurate and low-cost scheduling of cross-regional municipal road maintenance resources.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of transportation, in particular to a cross-regional municipal road maintenance resource scheduling method and system. BACKGROUND

[0002] With the development of urban agglomeration and metropolitan area, the urban space range is continuously extended, the municipal road network breaks through the boundary of single administrative region, and forms a cross-regional and cross-provincial transportation interconnection system. The traditional maintenance mode of "each for oneself" based on administrative region is difficult to cope with the collaborative maintenance demand of cross-regional roads, and the maintenance standards and resource scheduling efficiency of the same road in different administrative regions are significantly different. The resources are not balanced among regions, the equipment is complete in economically developed regions, and the resources are scarce in underdeveloped regions, and the scheduling response is slow when operating across regions. Based on this, the present application provides a cross-regional municipal road maintenance resource scheduling method and system. SUMMARY

[0003] The present application provides a cross-regional municipal road maintenance resource scheduling method, characterized in that it comprises:

[0004] S10, collecting road disease data, maintenance history records and real-time traffic flow data of each region through Internet of Things sensors and municipal management system interfaces, preprocessing the collected data, and constructing a structured data set;

[0005] S20, constructing a judgment matrix based on road disease type, severity and traffic flow using the analytic hierarchy process, calculating the weight of each factor by the square root method, and calculating the emergency degree score of each region maintenance according to the quantized basic data and weight, and dividing the regional maintenance level;

[0006] S30, according to the maintenance level of each region, using clustering algorithm combined with regional geographical position and road network connectivity, dividing maintenance groups with different priorities, and drawing a maintenance priority heat map;

[0007] S40, statistics of existing maintenance resources in each region, combined with the division results of maintenance groups, taking historical maintenance resource consumption data as the benchmark, and using regression analysis method to predict the resource demand of each region;

[0008] S50, for the resource shortage region, using the greedy algorithm to preferentially select the scheduling path with the shortest distance and complementary resource type with the resource redundant region, setting the resource scheduling threshold, and formulating a preliminary cross-regional resource scheduling scheme;

[0009] S60, constructing a multi-objective cost evaluation algorithm, combining particle swarm optimization algorithm to iteratively optimize the scheduling path and resource allocation ratio of the preliminary scheme, and selecting the scheme with total cost lower than the set threshold as the final execution scheme.

[0010] The cross-regional municipal road maintenance resource scheduling method as described above, wherein the road disease data, maintenance history records and real-time traffic flow data of each region are collected through Internet of Things sensors and municipal management system interfaces, the collected data is preprocessed, and a structured data set is constructed. Specifically, the following sub-steps are included:

[0011] The distributed optical fiber sensor deployed on the road surface collects crack development dynamic data, and the laser radar carried by the unmanned aerial vehicle scans to obtain pit depth and area data;

[0012] Access the municipal management system database to extract maintenance history records for the past three years, integrate real-time traffic flow interface data from the traffic department, and collect vehicle flow and speed information for each regional road section;

[0013] The box plot analysis method is applied to the collected road disease data to identify and eliminate outliers, and the isolated forest algorithm is used to detect sudden traffic flow anomalies for traffic flow data.

[0014] The cross-regional municipal road maintenance resource scheduling method as described above, wherein a judgment matrix is constructed based on road disease type, severity and traffic flow using the analytic hierarchy process, the weight of each factor is calculated by the square root method, and the regional maintenance urgency score is calculated based on the quantified basic data and weight, and the regional maintenance level is divided. Specifically, the following sub-steps are included:

[0015] Establish a factor level relationship index system to evaluate the road maintenance urgency, and determine the key evaluation factors of each level of road;

[0016] Based on the factor scoring, a pairwise comparison judgment matrix is constructed, the weight of each factor is calculated by the square root method, and the consistency test is used to ensure the rationality of the weight;

[0017] Multiply and accumulate the quantified factor data and weight of each region to calculate the regional maintenance urgency score and divide the regional maintenance level.

[0018] The cross-regional municipal road maintenance resource scheduling method as described above, wherein according to the regional maintenance level, the clustering algorithm is used in combination with the regional geographical location and road network connectivity to divide the maintenance groups with different priorities, and draw the maintenance priority heat map. Specifically, the following sub-steps are included:

[0019] Taking the maintenance urgency score as the basic variable and introducing the area and population density as auxiliary variables, the density clustering algorithm is used to automatically identify the density connected regional clusters and determine the number of clusters;

[0020] The clustering result is combined with a geographic information system to analyze the road network connectivity and distribution of traffic hubs in each region, and regions directly connected by the road network and having similar maintenance needs are preferentially merged. For regions with fuzzy boundaries, the clustering boundaries are adjusted through manual intervention.

[0021] The cross-regional municipal road maintenance resource scheduling method as described above, wherein the existing maintenance resources of each region are counted, and the resource demand of each region is predicted by using regression analysis based on historical maintenance resource consumption data in combination with the maintenance group division result. The method includes the following sub-steps:

[0022] The maintenance resources of each region are counted through a municipal department asset management system to establish a resource account and record the resource usage status.

[0023] The resource demand of each region in the next three months is predicted based on historical maintenance resource consumption data and in combination with future regional planning.

[0024] The resource supply-demand ratio of each region is calculated to determine the resource shortage regions and further analyze the resource gap types.

[0025] The cross-regional municipal road maintenance resource scheduling method as described above, wherein for the resource shortage regions, a greedy algorithm is used to preferentially select the scheduling path with the shortest distance and complementary resource types from the resource redundant regions, a resource scheduling threshold is set, and a preliminary cross-regional resource scheduling scheme is developed.

[0026] The method includes the following sub-steps:

[0027] A resource type matching matrix is established to compare the personnel skills, equipment functions, and material types of the resource shortage regions and the redundant regions, calculate the matching degree score, and determine the potential scheduling routes in combination with the shortest path algorithm between regions.

[0028] Starting from the region pair with the highest matching degree and the shortest distance, resources are gradually allocated according to the greedy algorithm. In the allocation process, the resource status of each region is updated in real time to ensure that the remaining resources in the dispatched region meet the basic maintenance needs.

[0029] Based on the resource allocation results, a detailed scheduling list is developed, and a resource scheduling flowchart is drawn.

[0030] The cross-regional municipal road maintenance resource scheduling method as described above, wherein a multi-objective cost evaluation algorithm is constructed, and the scheduling path and resource allocation ratio of the preliminary scheme are iteratively optimized by combining the particle swarm optimization algorithm to select a scheme with a total cost reduced below a set threshold as the final execution scheme. The method includes the following sub-steps:

[0031] A multi-objective cost evaluation algorithm is constructed based on transportation cost, time cost, and labor cost.

[0032] The particle swarm optimization algorithm is used, two optimization dimensions of path optimization and resource allocation ratio adjustment are set, and through iterative calculation, the scheduling scheme is updated constantly to find the solution space with the minimum total cost.

[0033] The application also provides a cross-regional municipal road maintenance resource scheduling system, comprising:

[0034] The acquisition and processing module acquires road disease data, maintenance history records and real-time traffic flow data of each region through Internet of Things sensors and municipal management system interfaces, pre-processes the collected data, and constructs a structured data set.

[0035] The emergency score module constructs a judgment matrix based on road disease types, severity and traffic flow using the analytic hierarchy process, calculates the weight of each factor through the square root method, and calculates the emergency degree score of each region maintenance according to the quantitative basic data and weight, and divides the regional maintenance level.

[0036] The division maintenance group module divides maintenance groups with different priorities according to the regional maintenance level, uses clustering algorithm combined with regional geographical location and road network connectivity, and draws a maintenance priority heat map.

[0037] The prediction demand module combines the maintenance group division result with the historical maintenance resource consumption data to predict the resource demand of each region by using regression analysis method.

[0038] The scheduling scheme module adopts the greedy algorithm to preferentially select the scheduling path with the shortest distance and complementary resource type in the resource redundant region for the resource shortage region, sets a resource scheduling threshold, and formulates a preliminary cross-regional resource scheduling scheme.

[0039] The optimization scheme module constructs a multi-objective cost evaluation algorithm, iteratively optimizes the scheduling path and resource allocation ratio of the preliminary scheme by combining the particle swarm optimization algorithm, and selects the scheme with the total cost lower than the set threshold as the final execution scheme.

[0040] The application has the following beneficial effects: the application realizes efficient, accurate and low-cost scheduling of cross-regional municipal road maintenance resources. BRIEF DESCRIPTION OF DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments described in the application, and other drawings can also be obtained by those skilled in the art according to these drawings.

[0042] Figure 1is a cross-regional municipal road maintenance resource scheduling method flowchart provided by the embodiment one of the application.

[0043] Figure 2 is a cross-regional municipal road maintenance resource scheduling system schematic diagram provided by the embodiment two of the application. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are some of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by a person skilled in the art without creative work fall within the protection scope of the application.

[0045] Embodiment one

[0046] As shown in the figure, the embodiment one of the application provides a cross-regional municipal road maintenance resource scheduling method, which comprises: Figure 1

[0047] S10, collecting road disease data, maintenance history records and real-time traffic flow data of each region through Internet of Things sensors and municipal management system interfaces, pre-processing the collected data, and constructing a structured data set.

[0048] S11, collecting crack development dynamic data through distributed optical fiber sensors deployed on the road surface, and obtaining pit depth and area data by using a laser radar carried by an unmanned aerial vehicle.

[0049] Distributed optical fiber sensors are laid on the road surface at intervals of 50 meters, real-time monitoring of road surface strain is realized based on the principle of fiber Bragg grating, and strain data is converted into crack width, length and development rate through a strain-crack relationship model. The laser radar carried by the unmanned aerial vehicle scans the road surface at a resolution of 10 centimeters, and after removing noise points by using a point cloud filtering algorithm, the depth, area and volume of the pit area are obtained through a point cloud three-dimensional reconstruction technology.

[0050] S12, accessing the municipal management system database, extracting maintenance history records in the past three years, integrating real-time traffic flow interface data of the traffic department, and collecting vehicle flow and vehicle speed information of each regional road section.

[0051] Accessing the municipal management system database, extracting road maintenance records of each region in the past three years, including repair time, resource input amount and road disease type, and resources including manpower, equipment and materials. Calling a data interface of the traffic department, obtaining vehicle flow, average speed, number of lanes and vehicle type distribution ratio of each road section every 15 minutes, small vehicle ratio, medium vehicle ratio and large vehicle ratio, and recording the collection time stamp.

[0052] ​S13, apply the box plot analysis method to the collected road disease data to identify and eliminate outliers; for traffic flow data, use the isolated forest algorithm to detect sudden traffic anomaly points.

[0053] For road disease data, the box plot analysis method is used to identify outliers. First, calculate the quartiles Q3, Q1 of the data to obtain the interquartile range IQR = Q3-Q1, and then introduce a correction coefficient α according to the data fluctuation, and the data exceeding Q3+α×1.5×IQR or below Q1-α×1.5×IQR is considered as an outlier and is removed.

[0054] For traffic flow data, based on the isolated forest algorithm, a dynamic threshold model is constructed by combining historical same-period flow data to identify abnormal points. Let the current data point flow be q cur , the historical same-period average flow be , and the standard deviation be σ his , the standardized score is calculated according to the formula , and when S σ is greater than the dynamic threshold, the data point is determined to be abnormal. The dynamic threshold function simulates the change rule of traffic flow under different time scales to generate a threshold that fits the actual traffic conditions. The specific formula is T = f(t season , t weekday , t hour ), where t season is the season, t weekday is the week, t hour is the hour, and f represents a function relationship that maps the three independent variables of season, week, and hour into a threshold for judging whether the traffic flow data is abnormal through specific calculation rules and logic.

[0055] For missing maintenance history data, use the random forest algorithm to predict missing values based on the maintenance rules of similar roads.

[0056] The processed data is classified and coded according to region, time, and data type to construct a standardized data table for subsequent analysis.

[0057] S20, based on road disease type, severity, and traffic flow, use the analytic hierarchy process to construct a judgment matrix, calculate the weight of each factor by the square root method, and calculate the regional maintenance urgency score based on the quantitative basic data and weight to divide the regional maintenance level.

[0058] S21, establish a factor level relationship index system for road maintenance emergency assessment to determine the key evaluation factors of each level of road.

[0059] The evaluation of road maintenance emergency degree involves many complex factors, and it is difficult and confusing to evaluate directly. By establishing a hierarchical relationship, the complex problem is decomposed into different levels of sub-problems. The hierarchical relationship is divided into target layer, criterion layer and index layer. The target layer is the maintenance emergency degree, the criterion layer is divided into disease, traffic, economy and other categories, and the index layer includes 10 key evaluation factors such as road disease severity, traffic flow impact, regional economic importance, surrounding facility density, maintenance history frequency, road grade, accident frequency, environmental protection requirements, policy priority and public complaint volume.

[0060] S22, based on the factor scoring, a judgment matrix of pairwise comparison is constructed, the square root method is used to calculate the weight of each factor, and the consistency test is used to ensure the rationality of the weight.

[0061] According to the scoring of the expert using 1-9 scale method for pairwise comparison of each factor, a judgment matrix A=(a ij ) n×n , a ij is the element of the i-th row and the j-th column in the judgment matrix A, which represents the scoring value when the i-th evaluation factor and the j-th evaluation factor are compared in importance, and n represents the order of the judgment matrix. The square root method is used to calculate the weight of the factor, the product of each row element of the matrix is calculated first Then calculate its n-th root The expert authority coefficient γ i is introduced, and the weight vector is obtained by weighted normalization of The rationality of the weight is tested by the consistency test formula. When the test result is less than the set threshold, the weight is considered reasonable.

[0062] S23, multiply the quantified factor data of each region by the weight and accumulate, calculate the maintenance emergency degree score of each region, and divide the regional maintenance level.

[0063] The calculation formula of the maintenance emergency degree score of each region is Where δ t is the weight correction coefficient of different time periods, W i is the weight of the i-th factor, and x i is the quantified score of the i-th factor. According to the score, the natural break point classification method is used to automatically divide the emergency maintenance area, the regular maintenance area and the observation area.

[0064] S30, according to the maintenance level of each region, use clustering algorithm combined with regional geographical location and road network connectivity to divide maintenance groups with different priorities, and draw maintenance priority heat map.

[0065] S31, taking the maintenance emergency degree score as a basic variable, introducing the area and population density as auxiliary variables, using the density clustering algorithm to automatically identify the density connected regional cluster, and determining the clustering number.

[0066] Taking the maintenance emergency degree score as a basic variable, introducing the area, population density, and economic activity as auxiliary variables, forming a four-dimensional data space, and calculating the economic activity by weighting the regional gross product and the number of enterprises. Using the density clustering algorithm, setting the dynamic neighborhood radius as The minimum sample number is MinPts, which is dynamically adjusted according to the data density. ε0 is the initial radius, θ is the adjustment coefficient, is the average distance from the current data point to the neighborhood point, and d avg is the global average distance. The algorithm starts from any unvisited point, searches for points in its ε neighborhood, and if the number of neighborhood points is greater than or equal to MinPts, a new cluster is formed, and the neighborhood points are added to the cluster. Continue to search the neighborhood from unvisited points in the cluster, and continuously expand the cluster, until all points are visited, automatically identify the density connected regional cluster, and determine the clustering number.

[0067] S32, combine the clustering results with geographic information system, analyze the road network connectivity and traffic hub distribution of each region, and preferentially merge the regions with direct road network connection and similar maintenance needs; for the boundary fuzzy region, adjust the clustering boundary through manual intervention.

[0068] Import the clustering results into the geographic information system, and use the road network analysis tool to calculate the road network connectivity between regions. The connectivity calculation formula is where C is the road network connectivity, L is the actual number of connected roads between regions, τ is the road capacity correction coefficient determined according to road grade and lane number, N is the total number of regions, μ is the weight coefficient, w avg is the average width of the connected road, and w max is the maximum width of all roads. Preferentially merge the regions with road network connectivity greater than the set threshold and similar maintenance needs, and similar maintenance needs refer to the emergency degree score difference less than 10. For the boundary fuzzy region, manually view satellite images, road planning maps, and traffic flow heat maps, and combine the ant colony algorithm to optimize the boundary adjustment path to achieve accurate division.

[0069] S40, statistics of existing maintenance resources in each region, combined with the maintenance group division results, taking the historical maintenance resource consumption data as the benchmark, using regression analysis method to predict the resource demand of each region.

[0070] S41, through the municipal department asset management system, statistics of each regional maintenance resource, establishment of resource account, recording of resource use state.

[0071] Through the municipal department asset management system, export each regional maintenance personnel information table, statistics road maintenance, mechanical operation, electrical maintenance and other different professional personnel quantity, personnel skill level distribution, work experience length distribution. Obtain equipment account, record equipment model, purchase time, use time, current state, maintenance record, equipment perfect rate, equipment perfect rate is calculated through historical failure data and maintenance. Statistics commonly used materials including asphalt, sandstone, cement and other inventory quantity, warehousing time, shelf life, material loss rate, material loss rate is determined according to storage conditions and time. Establish dynamic resource account, real-time update resource use state, including personnel attendance, equipment scheduling record, material out-of-stock quantity, resource idle rate and other information.

[0072] S42, according to historical maintenance resource consumption data, combined with regional future planning, predict the resource demand of each region in the next three months.

[0073] The specific prediction formula is D future is the predicted value of the maintenance resource demand of a region in the next three months, is the monthly average amount of maintenance resource consumption in the past three years of the region, reflecting the historical resource consumption level, ω l is the influence weight of newly built road length on resource demand, which is determined by regression analysis according to the correlation between newly built road and resource consumption in historical data, and the value range is between 0 and 1; Δl is the total length of planned newly built roads in the next three months, reflecting the incremental influence of newly built road projects on resource demand, l total is the total length of the existing roads in the region, Δq is the growth of traffic flow in the next three months, which is obtained through the traffic growth prediction data of the traffic department, q avg is the monthly average traffic flow of the region in the past year, ω q is the influence weight of traffic flow change on resource demand, which is also determined based on regression analysis of historical data; y is the number of other factors affecting resource demand in regional future planning, such as large-scale activities, policy requirements for road upgrading and reconstruction, etc. is the correction coefficient of the ith other influencing factor on resource demand, which is determined according to the size, urgency and other comprehensive evaluation of the factor, and positive number indicates increased demand and negative number indicates reduced demand; w i is the weight of the ith other influencing factor.

[0074] Residual test is performed on the prediction results to ensure that the error is within an acceptable range.

[0075] S43, calculate the resource supply and demand ratio of each region, determine the resource shortage area, and further analyze the resource gap type.

[0076] The resource supply-demand ratio of each region is calculated, considering the time effectiveness and importance difference of resources, and the calculation formula is Wherein, R is the comprehensive supply-demand ratio, D i is the predicted demand of the ith resource, S i is the existing amount of the ith resource, is the importance weight of the ith resource, and k is the number of resource categories. For the regions with a ratio greater than a set threshold, the resource gap type is further analyzed, the resource gap evaluation matrix is established, and factors such as supply-demand ratio, resource urgency, and resource substitutability are considered to determine the specific resource shortage type, including manpower shortage, equipment shortage, and material shortage. For the lightly short of regions, a four-level resource early warning classification based on risk level is established, and different response strategies such as emergency procurement, regional allocation, temporary recruitment, and shared leasing are developed.

[0077] S50, for the resource shortage region, a greedy algorithm is used to preferentially select the dispatch path with the shortest distance and complementary resource type with the resource redundant region, a resource dispatch threshold is set, and a preliminary cross-regional resource dispatch scheme is developed.

[0078] S51, a resource type matching matrix is established, the personnel skills, equipment functions, and material types of the resource shortage region and the redundant region are compared, and the matching degree score is calculated; combined with the shortest path algorithm between regions, the potential dispatch route is determined.

[0079] A multi-dimensional resource matching matrix M=(m ij ) b×c×e is established, where b is the number of resource shortage regions, c is the number of resource redundant regions, and e is the resource category, including personnel, equipment, and materials. For personnel skill matching, a skill-task matching degree algorithm is constructed, considering factors such as skill level, work experience, and professional relevance, and the personnel matching degree score is calculated by formula , where w s is the skill level weight, p ijs is the demand proportion of the ith shortage region to the jth redundant region for the s-level skill personnel, and r ijs is the supply proportion of the ith shortage region to the jth redundant region for the s-level skill personnel.

[0080] For equipment function matching, the similarity of equipment model parameters and demand parameters is compared, a device performance attenuation coefficient is introduced, and the equipment matching degree score is calculated by formula , where n para is the number of device parameters, h ijk is the similarity of the kth parameter of the equipment of the ith shortage region to the jth redundant region, and β ijk is the performance attenuation coefficient of the kth parameter of the equipment of the jth redundant region. The material type directly calculates the matching degree according to the number of the same type. The matrix element mijv is the matching degree score of the ith shortage area and the jth redundancy area for the vth type of resource. According to the multi-dimensional resource matching matrix result and the device function matching result, combined with the inter-area shortest path algorithm, the optimal path between areas is calculated with road distance, travel time, road condition level as the comprehensive weight, to determine the potential dispatch route.

[0081] S52, starting from the area pair with the highest matching degree and the shortest distance, gradually allocate resources according to the greedy algorithm, and update the resource state of each area in real time during the allocation process to ensure that the remaining resources of the dispatched area meet the basic maintenance needs.

[0082] Starting from the area pair with the highest matching degree and the optimal comprehensive path in the multi-dimensional resource matching matrix, resource allocation is performed according to the greedy algorithm. In the allocation process, the resource state of each area is updated in real time, considering factors such as resource transportation loss and deployment time cost, and setting a dynamic resource reservation threshold for the dispatched area T base is the basic reservation ratio, η is the adjustment coefficient, d yun is the transportation distance, d max is the maximum transportation distance, to ensure that the dispatched area meets the basic maintenance needs. If the resource demand of a certain area is not fully met, continue to allocate resources from the suboptimal matching area pair, while considering resource complementarity and synergistic effect, until the emergency needs of all shortage areas are basically met.

[0083] S53, according to the resource allocation result, develop a detailed dispatch list and draw a resource dispatch flowchart.

[0084] According to the resource allocation result, develop a detailed dispatch list, including the number of personnel and their specialties, the type and number of equipment, the type and weight of materials; arrange transportation vehicles, considering the weight, volume, transportation time limit, road weight and height limit requirements, select appropriate vehicle models and transportation batches through a vehicle scheduling optimization model based on genetic algorithm; the estimated arrival time is based on factors such as optimal path distance, real-time traffic conditions, vehicle speed limit, and rest time during the journey. Draw a resource dispatch flowchart, label key nodes such as resource dispatch location, transportation route, transit station, and dispatch location, and clearly define the responsibilities of each node, the time requirements for task completion, the resource handover process, and the quality acceptance standards.

[0085] S60, construct a multi-objective cost evaluation algorithm, and use a particle swarm optimization algorithm to iteratively optimize the dispatch path and resource allocation ratio of the preliminary scheme, and select a scheme with a total cost lower than a set threshold as the final execution scheme.

[0086] S61, construct a multi-objective cost evaluation algorithm based on transportation cost, time cost, and labor cost.

[0087] The transportation cost is divided into fuel cost, toll cost, and vehicle wear cost. The fuel cost is calculated according to the transportation distance, the fuel consumption per unit distance of the vehicle, the unit price of fuel, and the flatness of the route. The toll cost is calculated according to the number of toll sections, the toll standard, and the toll discount. The vehicle wear cost is determined according to the vehicle failure.

[0088] The time cost is determined according to the basic transportation time, the road congestion time, and the transportation urgency.

[0089] The labor cost is determined according to the number of workers, the working time, the hourly wage, and the overtime allowance.

[0090] In S62, the particle swarm optimization algorithm is used to set two optimization dimensions of path optimization and resource allocation ratio adjustment. Through iterative calculation, the scheduling scheme is continuously updated to find the solution space with the minimum total cost.

[0091] The resource allocation ratio and the transportation path selection in the scheduling scheme are taken as the position parameters of the particles. The allocation ratio is the allocation proportion of personnel, equipment, and materials in different regions, and the transportation path selection is the specific transportation route combination between regions. The particle swarm size is set to N, and the position and velocity of each particle are initialized in a multi-dimensional space. The dimensions in the multi-dimensional space include the resource allocation dimension and the path selection dimension. The value range of the position parameter is determined according to the actual total amount of resources and the regional road network, and the velocity parameter is set in a reasonable interval to control the step size of particle search.

[0092] The total cost C calculated by the multi-objective cost evaluation algorithm is taken as the fitness function of the particle, that is, the lower the total cost of the scheduling scheme corresponding to the particle position, the higher the fitness. The total cost corresponding to each particle position is calculated in each iteration to evaluate the quality of the particle.

[0093] In each iteration, each particle updates its velocity and position according to its historical optimal position and the group historical optimal position. In the updating process, the total cost corresponding to the new position is calculated in real time, and the particle's historical optimal position and the group historical optimal position are updated. The iteration continues until the maximum number of iterations is reached or the total cost converges to a satisfactory range, and finally the particle position with the minimum total cost is obtained, that is, the optimal scheduling scheme.

[0094] The optimized scheduling scheme is substituted into the cost evaluation algorithm to recalculate the total cost. The scheme whose total cost is reduced by more than a certain threshold is taken as the final scheme. If there is no scheme that fully meets the standard, the optimization algorithm parameters are adjusted or the cost evaluation algorithm is reexamined, and optimization and verification are performed again until a reliable scheduling scheme that meets the requirements is obtained.

[0095] Example Two

[0096] As Figure 2As shown, the second embodiment of the present application provides a cross-regional municipal road maintenance resource scheduling system, comprising:

[0097] The acquisition and processing module acquires road disease data, maintenance history records and real-time traffic flow data of each region through Internet of Things sensors and municipal management system interfaces, pre-processes the collected data, and constructs a structured data set. Specifically, it is divided into the following sub-modules:

[0098] The road data submodule acquires crack development dynamic data through distributed optical fiber sensors deployed on the road surface, and uses a laser radar carried by a UAV to scan to obtain pit depth and area data.

[0099] The traffic data submodule accesses the municipal management system database, extracts maintenance history records for the past three years, integrates real-time traffic flow interface data from the traffic department, and collects vehicle flow and speed information for each regional road segment.

[0100] The anomaly detection and deletion submodule applies the box plot analysis method to the collected road disease data to identify and eliminate outliers; for traffic flow data, the isolated forest algorithm is used to detect sudden traffic anomaly points.

[0101] The emergency score module constructs a judgment matrix based on road disease type, severity, and traffic flow using the analytic hierarchy process, calculates the weight of each factor using the square root method, and calculates the emergency degree score of each regional maintenance based on the quantified basic data and weights, and divides the regional maintenance level. Specifically, it is divided into the following sub-modules:

[0102] The factor determination submodule establishes a factor level relationship index system for road maintenance emergency degree evaluation to determine the key evaluation factors of each level road.

[0103] The factor weight submodule constructs a pairwise comparison judgment matrix based on factor scoring, calculates the weight of each factor using the square root method, and ensures the rationality of the weight through consistency test.

[0104] The division submodule multiplies and accumulates the quantified factor data and weights of each region to calculate the emergency degree score of each regional maintenance and divide the regional maintenance level.

[0105] The division maintenance group module divides maintenance groups with different priorities according to the regional maintenance level, uses clustering algorithms combined with regional geographic location and road network connectivity, and draws a maintenance priority heat map. Specifically, it is divided into the following sub-modules:

[0106] The clustering submodule takes the maintenance emergency degree score as the basic variable, introduces the area and population density as auxiliary variables, uses the density clustering algorithm to automatically identify the density connected regional clusters, and determines the number of clusters.

[0107] Group division sub-module: Combine the clustering results with geographic information system, analyze the road network connectivity and traffic hub distribution of each region, and preferentially merge regions with direct road network connection and similar maintenance needs. For regions with fuzzy boundaries, adjust the clustering boundaries through manual intervention.

[0108] Predictive demand module: Based on the historical maintenance resource consumption data, and combined with the maintenance group division results, use regression analysis to predict the resource demand of each region. Specifically, it is divided into the following sub-modules:

[0109] Statistical resource sub-module: Through the municipal department asset management system, statistical resource of each region is counted, and resource ledger is established to record resource usage status.

[0110] Prediction sub-module: According to historical maintenance resource consumption data, combined with future regional planning, predict the resource demand of each region in the next three months.

[0111] Shortage area determination sub-module: Calculate the resource supply-demand ratio of each region, determine the resource shortage area, and further analyze the resource gap type.

[0112] Dispatching scheme module: For resource shortage areas, use the greedy algorithm to preferentially select the dispatching path with the shortest distance and complementary resource type from the resource redundant area, set the resource dispatching threshold, and develop a preliminary cross-regional resource dispatching scheme. Specifically, it is divided into the following sub-modules:

[0113] Matching sub-module: Establish a resource type matching matrix, compare the personnel skills, equipment functions, and material types of resource shortage areas and redundant areas, calculate the matching degree score, and determine the potential dispatching route combined with the shortest path algorithm between regions.

[0114] Allocation sub-module: Starting from the region pair with the highest matching degree and the shortest distance, gradually allocate resources according to the greedy algorithm. In the allocation process, update the resource status of each region in real time to ensure that the remaining resources in the dispatched region meet the basic maintenance needs.

[0115] Dispatching sub-module: Based on the resource allocation results, develop a detailed dispatching list and draw a resource dispatching flowchart.

[0116] Optimization scheme module: Construct a multi-objective cost evaluation algorithm, and use the particle swarm optimization algorithm to iteratively optimize the dispatching path and resource allocation ratio of the preliminary scheme. Select the scheme with a total cost reduction below the set threshold as the final execution scheme. Specifically, it is divided into the following sub-modules:

[0117] Cost evaluation sub-module: Construct a multi-objective cost evaluation algorithm based on transportation cost, time cost, and labor cost.

[0118] Iteration optimization submodule: particle swarm optimization algorithm is used, two optimization dimensions of path optimization and resource allocation ratio adjustment are set, and the scheduling scheme is updated through iteration calculation to find the solution space with minimum total cost.

[0119] The above specific embodiments further explain the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modification, equivalent replacement, improvement, etc. made on the basis of the technical solutions of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for scheduling municipal road maintenance resources across regions, characterized in that, Comprise: S10, through the Internet of Things sensor, municipal management system interface collection each area road disease data, maintenance history record and real-time traffic flow data, pre-processing of collected data, structured data set is built; S20, based on road disease type, severity, traffic flow, use analytic hierarchy process to build judgment matrix, through the square root method to calculate the weight of each factor, according to the weight of the quantized basic data and weight, calculate the score of each area maintenance emergency, and divide the regional maintenance level; S30, according to the maintenance level of each area, use clustering algorithm combined with regional geographical location and road network connectivity, divide the maintenance group with different priority, and draw the maintenance priority heat map; S40, statistics of each area existing maintenance resources, combined with the division results of maintenance group, take the historical maintenance resource consumption data as the benchmark, use regression analysis method to predict the resource demand of each area; S50, for resource shortage area, use greedy algorithm to select the shortest dispatch path with complementary resources in the resource redundant area, set the resource scheduling threshold, and develop the preliminary cross-regional resource scheduling scheme; S60, construct multi-objective cost evaluation algorithm, combined with particle swarm optimization algorithm to iterate and optimize the dispatch path and resource allocation ratio of the preliminary scheme, select the scheme with total cost lower than the set threshold as the final execution scheme.

2. The method of claim 1, wherein, Through the Internet of Things sensor, municipal management system interface collection each area road disease data, maintenance history record and real-time traffic flow data, pre-processing of collected data, structured data set is built, which is divided into the following substeps: Through the distributed optical fiber sensor deployed on the road surface to collect crack development dynamic data, use unmanned aerial vehicle to scan the depth and area data of potholes by laser radar; Access to municipal management system database, extract maintenance history record in the past three years, integrate real-time traffic flow interface data of traffic department, collect vehicle flow and speed information of each area road section; Apply box plot analysis method to the collected road disease data, identify and eliminate outliers, and use isolated forest algorithm to detect sudden traffic anomaly points for traffic flow data.

3. The method of claim 1, wherein, Based on road disease type, severity, traffic flow, use analytic hierarchy process to build judgment matrix, through the square root method to calculate the weight of each factor, according to the weight of the quantized basic data and weight, calculate the score of each area maintenance emergency, and divide the regional maintenance level, which is divided into the following substeps: Establish the index system of factor level relationship for road maintenance emergency evaluation, determine the key evaluation factors of each level road; Based on factor scoring, build a judgment matrix for pairwise comparison, use square root method to calculate the weight of each factor, and ensure the rationality of weight through consistency test; Multiply and accumulate the quantized factor data and weight of each area, calculate the score of each area maintenance emergency, and divide the regional maintenance level.

4. The method of claim 1, wherein, According to the maintenance level of each area, use clustering algorithm combined with regional geographical location and road network connectivity, divide the maintenance group with different priority, and draw the maintenance priority heat map, which is divided into the following substeps: With the maintenance emergency score as the basis variable, the area and population density as the auxiliary variables, the density clustering algorithm is used to automatically identify the density connected regional cluster and determine the number of clusters; The clustering results are combined with geographic information system to analyze the road network connectivity and traffic hub distribution of each region, and the regions with direct connectivity and similar maintenance needs are preferentially merged. For the fuzzy boundary region, the clustering boundary is adjusted through manual intervention.

5. The method of claim 1, wherein, The existing maintenance resources of each region are counted, and the resource demand of each region is predicted by regression analysis method based on historical maintenance resource consumption data. The specific steps are as follows: Through the municipal department asset management system, the maintenance resources of each region are counted, and the resource account is established to record the resource usage status; According to the historical maintenance resource consumption data and the future planning of the region, the resource demand of each region in the next three months is predicted; The resource supply and demand ratio of each region is calculated to determine the resource shortage area, and the resource gap type is further analyzed.

6. The method of claim 1, wherein, For the resource shortage area, the greedy algorithm is used to preferentially select the dispatch path with the shortest distance and complementary resource type from the resource redundant area, set the resource dispatch threshold, and develop the preliminary cross-regional resource dispatch scheme, The specific steps are as follows: A resource type matching matrix is established to compare the personnel skills, equipment functions and material types of the resource shortage area and the redundant area, calculate the matching degree score, and determine the potential dispatch route by combining the shortest path algorithm between regions; Starting from the region pair with the highest matching degree and the shortest distance, the resources are allocated step by step according to the greedy algorithm. In the allocation process, the resource status of each region is updated in real time to ensure that the remaining resources of the dispatched region meet the basic maintenance needs; According to the resource allocation results, a detailed dispatch list is developed, and a resource dispatch flowchart is drawn.

7. The method of claim 1, wherein, A multi-objective cost evaluation algorithm is constructed, and the particle swarm optimization algorithm is used to iteratively optimize the dispatch path and resource allocation ratio of the preliminary scheme. The scheme with a total cost lower than the set threshold is selected as the final execution scheme. The specific steps are as follows: A multi-objective cost evaluation algorithm is constructed based on transportation cost, time cost and labor cost; The particle swarm optimization algorithm is used to set two optimization dimensions of path optimization and resource allocation ratio adjustment. Through iterative calculation, the dispatch scheme is constantly updated to find the solution space with the minimum total cost.

8. A cross-jurisdictional municipal road maintenance resource dispatch system, characterized by, It includes: Data acquisition and processing module: Collect road disease data, maintenance history records and real-time traffic flow data of each region through Internet of Things sensors and municipal management system interface, preprocess the collected data, and construct a structured data set; Emergency score module: Based on road disease type, severity and traffic flow, a judgment matrix is constructed using the analytic hierarchy process, the weight of each factor is calculated by square root method, and the maintenance emergency score of each region is calculated based on the quantitative basic data and weight, and the regional maintenance level is divided; The division of maintenance group module: According to the maintenance level of each region, the clustering algorithm is used to combine the regional geographical location and road network connectivity to divide the maintenance groups with different priorities, and draw the maintenance priority heat map; The prediction demand module: the existing maintenance resources of each region are counted, the historical maintenance resource consumption data is taken as a benchmark, and a regression analysis method is used to predict the resource demand of each region according to the division result of the maintenance group; The scheduling scheme module: for the resource shortage region, a greedy algorithm is used to preferentially select the scheduling path with the shortest distance and complementary resource type from the resource redundant region, a resource scheduling threshold is set, and a preliminary cross-regional resource scheduling scheme is developed; The optimization scheme module: a multi-objective cost evaluation algorithm is constructed, the scheduling path and resource allocation ratio of the preliminary scheme are iteratively optimized by combining a particle swarm optimization algorithm, and a scheme with a total cost lower than a set threshold is selected as the final execution scheme.

Images