A method and system for collaborative optimization of environmental monitoring personnel allocation and vehicle routing considering sampling time.

Abstract

This invention provides a method for collaborative optimization of environmental monitoring personnel allocation and vehicle routes, considering sampling time, belonging to the field of logistics and route planning technology. This invention solves the technical problems of ignoring the dynamic changes in sampling time with the number of personnel and lacking a collaborative optimization mechanism for personnel and vehicles. The method includes: acquiring information on the monitoring center, monitoring points, personnel, and vehicles; constructing an environmental monitoring network graph; establishing a collaborative optimization model with the objective of minimizing the sum of vehicle usage cost, personnel cost, and travel distance cost, and including constraints such as time windows and capacity; solving the problem using a genetic algorithm that integrates neighborhood search, with chromosomes using double-layer encoding to represent personnel allocation and access order, and designing personnel selection probability operators and load balancing movement operators for local optimization. This invention can dynamically adjust the number of sampling personnel at monitoring points, achieve collaborative scheduling of personnel and vehicles, effectively balance various costs, and improve the execution efficiency and economic benefits of environmental monitoring tasks.

Description

Technical Field

[0001] This invention belongs to the field of logistics distribution and route planning technology, and relates to a method and system for coordinating the allocation of environmental monitoring personnel and vehicle routes, taking into account sampling time. Background Technology

[0002] Environmental monitoring, as a crucial component of environmental protection engineering, directly impacts the scientific rigor of environmental governance and decision-making through the efficiency of its execution and the quality of its monitoring data. In recent years, with the increasing demands for both the breadth of environmental monitoring network coverage and the timeliness of data in my country, the number and complexity of environmental monitoring tasks have significantly increased. Traditional scheduling and planning methods are proving inadequate when faced with multi-site, differentiated sampling tasks.

[0003] Currently, environmental monitoring task planning often focuses on vehicle route optimization or simply binding personnel to vehicles, failing to fully consider the crucial characteristic that sampling task operation time dynamically changes with the number of participants. In real-world scenarios, operation times vary significantly between different monitoring points, and the number of sampling personnel deployed to the same monitoring point significantly impacts sampling time. Existing methods often employ fixed personnel configurations or divide tasks by fixed areas, ignoring the trade-off between more personnel saving time but also incurring higher costs, making it difficult to achieve optimal resource allocation. Furthermore, existing models typically consider vehicle route optimization and personnel allocation separately, lacking a collaborative optimization mechanism between the two. This results in personnel configuration not being dynamically adjusted based on the actual vehicle route structure and task sequence.

[0004] From the perspective of improving the operational efficiency and economic benefits of environmental monitoring, if the planning method only focuses on the shortest path or single cost, while ignoring the dynamic nature of sampling time and the collaborative relationship between personnel, vehicles, and tasks, it will be difficult to cope with the complex and ever-changing actual monitoring needs, resulting in resource waste or task delays. Therefore, a method that considers the collaborative optimization of environmental monitoring personnel allocation and vehicle routes that takes into account sampling time is of great significance for improving the overall efficiency of environmental monitoring, controlling operating costs, and ensuring data timeliness. Summary of the Invention

[0005] In view of this, and considering that most existing technologies neglect the dynamic changes in environmental monitoring sampling task time with the number of personnel, and lack a mechanism for collaborative optimization of personnel allocation and vehicle routes, this invention provides a method and system for collaborative optimization of environmental monitoring personnel allocation and vehicle routes that takes into account sampling time. Under the premise of meeting actual constraints such as working time window, vehicle capacity, and total number of personnel, it simultaneously optimizes vehicle usage costs, personnel costs, and travel distance costs, so as to achieve efficient, economical, and feasible scheduling and route planning for monitoring tasks, thereby improving the overall efficiency and economic benefits of environmental monitoring operations.

[0006] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, this invention proposes a method for coordinating and optimizing the allocation of environmental monitoring personnel and vehicle routes, taking into account sampling time, comprising the following steps: 1. Input data to obtain environmental monitoring sampling task information within the planning period.

[0007] 1.1 Input the relevant information of the testing center, mainly including its geographical location and sampling operation time. The geographical location is obtained through geographic information systems such as Gaode Maps, specifying its latitude and longitude coordinates. The sampling operation time is based on the testing center's operation and management system, clearly defining the daily sampling operation periods, including the start time. With the latest end time .

[0008] 1.2. Input relevant information for all monitoring points. This mainly includes the geographical location of each monitoring point, the average sampling time, and the number of people assigned to each monitoring point. The geographical locations of the monitoring points were obtained using geographic information systems such as Gaode Maps, which provided their latitude and longitude coordinates. The average sampling time was determined based on past sampling experience.

[0009] 1.3. Input sampling personnel information. Enter the total number of sampling personnel. p Each sampling personnel was assigned a unique identification number. Furthermore, in accordance with the company's labor management system and relevant regulations, the maximum daily sampling time for each sampling personnel was clearly defined. and personnel usage costs .

[0010] 1.4 Input monitoring vehicle information. This involves querying and determining the key vehicle parameters used by the testing center, as well as vehicle operating costs. Cost per kilometer while the vehicle is in motion Maximum capacity per vehicle Q .

[0011] 2. Convert the input data format and construct a set of testing centers, a set of monitoring points, and a set of sampling personnel. Assign a unique number to each testing center, monitoring point, and vehicle, with the testing center numbered 0. c The monitoring points are numbered sequentially according to their entry order: 1, 2, 3... c Therefore, the set of detection centers can be represented as {0}, and the set of monitoring points can be represented as... C ={1,2,3...... c The center has a total of p A set of sampling personnel can be represented as a collection of sampling personnel performing monitoring tasks. P And | P |= pThe set of monitored vehicles is represented as V .

[0012] 3. Establish an environmental monitoring network.

[0013] 3.1 Constructing a monitoring network diagram G = ( N , E ), N Let be the set of nodes in this network. E Let be the set of arcs in this network.

[0014] 3.2 Determine the set of nodes N Monitoring network G The node set consists of a set of monitoring points and a set of detection center nodes, and can be represented as follows: N = C ∪{0}.

[0015] 3.3 Determine the arc set E Define nodes i and nodes j If the road segment between them is an arc, then the monitoring network... G The set of arcs is E = {( i , j )| i ∈ N , j ∈ N , i ≠ j}

[0016] 3.4 Determine the distances between different locations. and vehicle speed v Based on the latitude and longitude coordinates of each node, the actual distance between each node is obtained using tools such as Gaode Maps. (Unit: km). Simultaneously, the average vehicle speed was determined by combining historical traffic data and road network attributes. v (Unit: km / h)

[0017] 4. Establish a collaborative model for environmental monitoring personnel allocation and vehicle routing, taking into account sampling time, with the objective of minimizing total cost. The total cost is the sum of vehicle usage cost, personnel usage cost, and distance cost.

[0018] 4.1 The objective function of the model is: Minimize cost:

[0019] 4.2 The model constraints are as follows: (1) Each monitoring point must be visited once:

[0020] in, Indicates whether the vehicle departed from the monitoring point i Drive to the monitoring point j .when At that time, it indicates that the vehicle departed from the monitoring point. i Drive to the monitoring point j ,when At that time, it indicates that the vehicle did not leave the monitoring point. i Drive to the monitoring point j .

[0021] (2) The vehicle departs from and returns to the testing center:

[0022] in, Indicates whether the vehicle has traveled from inspection center 0 to monitoring point. j , Indicates whether the vehicle departed from the monitoring point i To the testing center 0.

[0023] (3) Flow balancing:

[0024] in, Indicates whether the vehicle departed from the monitoring point j Destination monitoring point k . (4) The actual number of vehicles used does not exceed the total number of available vehicles:

[0025] in, Indicates whether the vehicle has traveled from the testing center 0 to the monitoring point. i .

[0026] (5) Personnel number constraints at monitoring points:

[0027] in, Indicates monitoring point i The number of sampling personnel Q This represents the maximum number of passengers a vehicle can carry.

[0028] (6) Total number of personnel limit:

[0029] in, p This refers to the total number of sampling personnel at the testing center.

[0030] (7) The number of personnel assigned to all monitoring points on the same vehicle is the same:

[0031] in, Indicates monitoring point j The number of sampling personnel allocated. (8) The time of arrival at the next monitoring point shall not be earlier than the time of departure from the current monitoring point plus the travel time:

[0032] in, This indicates that the vehicle has arrived at the monitoring point. j At that moment, This indicates that the vehicle has left the monitoring point. i At that moment, Indicates monitoring point i to monitoring point j The driving distance.

[0033] (9) The time when the vehicle leaves the monitoring point is equal to the sum of its arrival time and the time required to complete sampling at that point:

[0034] in, This indicates that the vehicle has arrived at the monitoring point. i At that moment, Indicates monitoring point i Sampling time.

[0035] (10) The time when a vehicle arrives at each monitoring point shall not be earlier than the sampling start time, and the time when it leaves each monitoring point shall not be later than the sampling end time:

[0036]

[0037] in, Indicates the sampling start time. Indicates the sampling end time.

[0038] (11) All vehicles shall depart from the testing center at the designated sampling start time:

[0039] in, This indicates the time the vehicle leaves the testing center.

[0040] (12) Time capacity constraint:

[0041] in, This indicates the maximum sampling duration per day.

[0042] (13) Eliminating constraints in sub-loops:

[0043] in, , This represents an auxiliary variable.

[0044] 5. Use a genetic algorithm to solve the model. To solve the problem of coordinating the allocation of environmental monitoring personnel and vehicle routes while minimizing total cost and considering sampling time, this invention proposes a genetic algorithm that integrates neighborhood search. This algorithm uses a genetic algorithm as its main framework and optimizes the allocation of sampling personnel and monitoring points through a neighborhood search operator, guiding the population to evolve towards a more balanced direction.

[0045] 5.1 Chromosome Coding. A two-layer integer coding method is used. The upper layer of the chromosome coding represents the number of personnel allocated, and its length is equal to the total number of monitoring points. C |, the value for each gene locus is the number of sampling personnel assigned to that monitoring point, and the value range is [1, Q ],in Q This represents the maximum number of passengers the vehicle can carry. The lower-level code indicates the access order of the monitoring points, ranging from 1 to | C A permutation of | represents the order in which vehicles perform sampling tasks. This encoding method can directly support structural adjustments by the neighborhood search operator at the chromosome level.

[0046] 5.2, The size of the random generation is P The initial population. For each individual, traverse each personnel group, randomly generate an allocation scheme from monitoring points to several vehicles within that group, and ensure that each vehicle is assigned at least one monitoring point.

[0047] 5.3 Chromosome Decoding. First, based on the upper-level coding, the monitoring points are divided into different demand groups according to the allocated number of people. The order within each group is determined by the arrangement order of the lower-level coding. Then, each group is processed in sequence. Under the premise of satisfying the vehicle occupancy constraint and time window constraint, the monitoring point is inserted into the current vehicle, and the sampling time of each point is determined according to the allocated number of people, thus completing the vehicle grouping and path construction.

[0048] 5.4 Fitness function calculation. .

[0049] 5.5. The upper-level coding corresponds to the number of personnel allocated, using a two-point crossover. During operation, two crossover points are randomly selected, and the gene fragments of the two parent individuals between these two points are directly exchanged, thus generating two offspring individuals. The lower-level coding corresponds to the monitoring point access order, using sequential crossover. During operation, two crossover points are first randomly selected, and the monitoring point sequence fragment of one parent individual within that interval is directly copied to the corresponding position in the offspring; then, monitoring points that have not yet appeared in the offspring are extracted from the other parent individual in the original order and filled sequentially along the remaining empty spaces in the offspring.

[0050] 5.6. The upper-level coding uses exchange mutation, randomly selecting two monitoring point locations and swapping the corresponding personnel allocation numbers. The lower-level coding uses reverse transcription mutation, randomly selecting two locations and completely reversing the access order of the monitoring points between these two locations, thereby generating a new path structure.

[0051] 5.7. Integrate two neighborhood search operators to carry out local search optimization.

[0052] 5.7.1 A personnel selection probability operator is used to dynamically allocate sampling personnel at monitoring points. Since sampling time decreases non-linearly with the number of people allocated, different personnel allocation probabilities should be matched to monitoring points with different workloads. This method divides the monitoring point into three intervals based on the single-person sampling time and presets the personnel selection probability for each interval, making local adjustments to the personnel allocation while keeping the vehicle path unchanged. The specific steps of the personnel selection probability operator are as follows: Step 1: Traverse all monitoring points i ∈ C The system uses a set probability to determine whether to adjust personnel at the monitoring point.

[0053] Step 2: For the selected monitoring points i Based on the time it takes for an individual to complete a sampling task t i s The number of people at that point will be randomly reassigned. y i n .like t i s < t 1, then with p 1 1 The probability allocation for 1 person p 1 2 The probability allocation for 2 people p 1 3 The probability is allocated to 3 people; if t 1≤ t i s ≤ t2, then with p 2 1 The probability allocation for 1 person p 2 2 The probability allocation for 2 people p 2 3 The probability is allocated to 3 people; if t i s > t 2, then with p 3 1 The probability allocation for 1 person p 3 2 The probability allocation for 2 people p 3 3 The probability is allocated to 3 people.

[0054] Step 3: After updating the personnel allocation, keep the current vehicle routes unchanged and adjust the allocation based on the new number of personnel. y i n Recalculate the monitoring point i Sampling time t i And check that the operating time of each vehicle does not exceed T max With vehicle passenger capacity y i n No more than Q .

[0055] Step 4: If the adjusted total operating costs decrease and all constraints are met, accept the adjustment and update. y i = y i n Otherwise, revert to the original personnel allocation plan. y i .

[0056] Step 5: Repeat the above process until all monitoring points have been adjusted once or the maximum number of iterations has been reached.

[0057] 5.7.2 A load balancing mover is used to optimize the task load between vehicles. Since vehicle load is positively correlated with the number of monitoring points it visits, this method identifies the vehicles with the highest and lowest loads by statistically analyzing the number of monitoring points served by each vehicle in the sampling task. Then, a small number of monitoring point tasks are transferred between these two types of vehicles to optimize the search, making it easier to obtain candidate solutions that perform better in terms of vehicle load balance. The specific steps of the load balancing mover are as follows: Step 1: Calculate the statistics for each vehicle in the current plan. v ∈V Number of monitoring points accessed | C v | Identify the vehicles that visited the monitoring points the most. v max and the fewest number of vehicles v min .

[0058] Step 2: From the vehicle v max monitoring point set C v max One monitoring point is selected in sequence. i ∈ C v max Try to take it from C v max Remove from the path and insert into the vehicle. v min Any feasible position in the path.

[0059] Step 3: Set up monitoring points i Personnel allocation number y i Adjusted to vehicles v min Current staffing standards y v min Recalculate the total operating cost of the two vehicles. If the total cost decreases after the relocation, and the total number of personnel does not exceed the total number of available personnel, then... p The working time of each vehicle shall not exceed T max If so, accept the move and update. C v max , C v min as well as y i Otherwise, revert to the original plan.

[0060] Step 4: Repeat the above process until no further improvement can be made or the maximum number of iterations is reached.

[0061] 5.8. Individual selection is carried out using a tournament selection method based on the inverse of fitness combined with an elite retention strategy.

[0062] The algorithm is set to terminate when the population reaches the maximum number of iterations, thus obtaining the final monitoring path scheme.

[0063] 6. Output planning scheme. Mainly includes: (1) Detailed monitoring path, clearly showing the vehicle monitoring path and the order of monitoring point visits and the number of sampling personnel on each vehicle in tabular form, and calculating the total working time of each vehicle in the cycle; (2) Core indicators: vehicle usage cost, driving cost, and personnel cost.

[0064] Secondly, this invention proposes a collaborative optimization system for environmental monitoring personnel allocation and vehicle routing that considers sampling time, comprising the following modules: 1) Data input module: Acquires and stores sampling task data, including information on the testing center, monitoring points, vehicles, and sampling personnel.

[0065] 2) Data processing module: Converts the format of the input data and constructs a set of monitoring point nodes, a set of detection center nodes, a set of vehicles, and a set of sampling personnel.

[0066] 3) Delivery network establishment module: Based on the basic information dataset, the addresses of the detection center and all monitoring points are abstracted into network nodes to construct an environmental monitoring network diagram.

[0067] 4) Model building module: Based on the environmental monitoring network diagram and basic information dataset, complete the setting of the objective function and constraints of the model, and establish an environmental monitoring sampling task path planning model.

[0068] 5) Optimization and solution module: The genetic algorithm with fusion neighborhood search is used to solve the environmental monitoring sampling task path model. Through iterative optimization, the environmental monitoring sampling task path scheme is finally obtained.

[0069] 6) Solution Output Module: Outputs the final delivery route solution information, including the monitoring routes of each vehicle, vehicle operation time, monitoring point access order, cost statistics, etc.

[0070] The beneficial effects of this invention are as follows: This invention, based on the environmental monitoring sampling task scheduling scenario, fully considers sampling operation time and proposes a method and system for optimizing the allocation of environmental monitoring personnel and vehicle routes while taking into account sampling time. Compared with existing planning methods that focus on a single dimension or use static sampling time assumptions, this invention, by constructing an optimization model that integrates human and vehicle collaboration and designing corresponding solution algorithms, can more effectively balance personnel costs, vehicle costs, and travel costs while meeting various practical constraints, minimizing total operating costs, and improving task execution efficiency and resource utilization. This has positive practical value for the scientific and refined operation management of the environmental monitoring industry.

[0071] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0072] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein: Figure 1 This is an overall flowchart of the method for coordinating and optimizing the allocation of environmental monitoring personnel and vehicle routes, taking into account sampling time, as described in this invention. Figure 2 This is a flowchart illustrating the solution process of the genetic algorithm that integrates neighborhood search in this invention. Figure 3 This is a schematic diagram of the monitoring point distribution in one embodiment of the present invention. Detailed Implementation

[0073] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0074] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0075] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do 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. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present invention. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0076] Figure 1 This is an overall flowchart of the method for coordinating and optimizing the allocation of environmental monitoring personnel and vehicle routes, taking into account sampling time, as described in this invention. Figure 2 This is a flowchart illustrating the solution process of the genetic algorithm that integrates neighborhood search in this invention. Taking an environmental monitoring center in a certain city as an example, this invention provides a method for collaborative optimization of environmental monitoring personnel allocation and vehicle routes, considering sampling time, and includes the following steps: 1. Input data to obtain information on the planned sampling task.

[0077] 1.1 Input the relevant information of the testing center, mainly including its geographical location and sampling operation time. The geographical location is obtained by obtaining its latitude and longitude coordinates through geographic information systems such as Gaode Map; the sampling operation time is based on the testing center's operation and management system, which clearly plans the sampling time slots that can be carried out each day: the vehicle will depart from the testing center at 9:00 am every day, and all sampling tasks must be completed before 5:00 pm on the same day.

[0078] 1.2 Enter the relevant information for all monitoring points, such as Figure 3 As shown, this includes the geographical location of each monitoring point, the average sampling time, and the number of people assigned to each monitoring point. y i = (1, 2, 3). The geographical location of the monitoring point is obtained using the same geographic information system as the detection center, which is also used to obtain its latitude and longitude coordinates; the average sampling time is determined based on historical sampling data and combined with the geographical environment of the monitoring point.

[0079] 1.3 Input sampling personnel information. Enter the total number of personnel participating in the sampling. p =30, and a unique identification number is assigned to each sampling personnel. At the same time, in accordance with the company's labor management system and relevant regulations, the maximum daily sampling time for each sampling personnel is clearly defined. =480 minutes and personnel usage costs =200 yuan.

[0080] 1.4 Input Monitoring Vehicle Information. The key parameters of the monitoring vehicles used by the testing center are determined through querying, mainly including: vehicle operating costs. =100 yuan, the cost per kilometer of vehicle travel. =2 yuan / km, maximum capacity per vehicle Q =3 people.

[0081] 2. Convert the input data format to construct a set of testing centers, a set of monitoring points, a set of vehicles, and a set of sampling personnel. Assign a unique number to each testing center, monitoring point, and vehicle, with the testing center numbered 0, and the 40 monitoring points numbered sequentially from 1 to 40 according to their entry order. Therefore, the set of testing centers can be represented as {0}, and the set of monitoring points can be represented as... C ={1,2,3......40}, the center has a total of 30 sampling personnel to carry out sampling tasks, and the set of sampling personnel can be represented as... P And | P |=30.

[0082] 3. Establish an environmental monitoring network.

[0083] 3.1 Constructing a monitoring network diagram G = ( N , E ), N Let be the set of nodes in this network. E Let be the set of arcs in this network.

[0084] 3.2 Determine the set of nodes N Monitoring network G The node set consists of a set of monitoring points and a set of detection center nodes, and can be represented as follows: N = C ∪{0}.

[0085] 3.3 Determining the Arc Set E Define nodes i and nodes j If the road segment between them is an arc, then the monitoring network... G The set of arcs is E ={( i , j )| i ∈ N , j ∈N, i ≠ j}

[0086] 3.4 Determine the distances between different locations and vehicle speed v Based on the latitude and longitude coordinates of each node, the actual distance between each node is obtained using tools such as Gaode Maps. (Unit: km). Simultaneously, the average vehicle speed was determined by combining historical traffic data and road network attributes. v =60 (unit: km / h).

[0087] 4. Establish a collaborative model for environmental monitoring personnel allocation and vehicle routing, taking into account sampling time, with the objective of minimizing total cost. The total cost is the sum of vehicle usage cost, personnel usage cost, and distance cost.

[0088] 4.1 The objective function of the model is: Minimize cost:

[0089] 4.2 The model constraints are as follows: (1) Each monitoring point must be visited once:

[0090] in, Indicates whether the vehicle departed from the monitoring point i Drive to the monitoring point j .when At that time, it indicates that the vehicle departed from the monitoring point. i Drive to the monitoring point j ,when At that time, it indicates that the vehicle did not leave the monitoring point. i Drive to the monitoring point j .

[0091] (2) The vehicle departs from and returns to the testing center:

[0092] in, Indicates whether the vehicle has traveled from the testing center 0 to the monitoring point. j , Indicates whether the vehicle departed from the monitoring point i To the testing center 0.

[0093] (3) Flow balancing:

[0094] in, Indicates whether the vehicle departed from the monitoring point j Destination monitoring point k .

[0095] (4) The actual number of vehicles used does not exceed the total number of available vehicles:

[0096] in, Indicates whether the vehicle has traveled from the testing center 0 to the monitoring point. i .

[0097] (5) Personnel number constraints at monitoring points:

[0098] in, Indicates monitoring point i The number of sampling personnel Q This represents the maximum number of passengers a vehicle can carry.

[0099] (6) Total number of personnel limit:

[0100] in, p This refers to the total number of sampling personnel at the testing center.

[0101] (7) The number of personnel assigned to all monitoring points on the same vehicle is the same:

[0102] in, Indicates monitoring point j The number of sampling personnel allocated.

[0103] (8) The time of arrival at the next monitoring point shall not be earlier than the time of departure from the current monitoring point plus the travel time:

[0104] in, This indicates that the vehicle has arrived at the monitoring point. j At that moment, This indicates that the vehicle has left the monitoring point. i At that moment, Indicates monitoring point i to monitoring point j The driving distance.

[0105] (9) The time when the vehicle leaves the monitoring point is equal to the sum of its arrival time and the time required to complete sampling at that point:

[0106] in, This indicates that the vehicle has arrived at the monitoring point. i At that moment, Indicates monitoring point i Sampling time.

[0107] (10) The time when a vehicle arrives at each monitoring point shall not be earlier than the sampling start time, and the time when it leaves each monitoring point shall not be later than the sampling end time:

[0108]

[0109] in, Indicates the sampling start time. Indicates the sampling end time.

[0110] (11) All vehicles shall depart from the testing center at the designated sampling start time:

[0111] in, This indicates the time the vehicle leaves the testing center.

[0112] (12) Time capacity constraint:

[0113] in, This indicates the maximum sampling duration per day.

[0114] (13) Eliminating constraints in sub-loops:

[0115] in, , This represents an auxiliary variable.

[0116] 5. Use a genetic algorithm to solve the model. To solve the problem of coordinating the allocation of environmental monitoring personnel and vehicle routes while minimizing total cost and considering sampling time, this invention proposes a genetic algorithm that integrates neighborhood search. This algorithm uses a genetic algorithm as its main framework and optimizes the allocation of sampling personnel and monitoring points through a neighborhood search operator, guiding the population to evolve towards a more balanced direction.

[0117] 5.1 Chromosome Coding. A two-layer integer coding method is used. The upper layer of the chromosome coding represents the number of personnel allocated, and its length is equal to the total number of monitoring points. C |, the value for each gene locus is the number of sampling personnel assigned to that monitoring point, and the value range is [1, Q ],in Q This represents the maximum number of passengers the vehicle can carry. The lower-level code indicates the access order of the monitoring points, ranging from 1 to | C A permutation of | represents the order in which vehicles perform sampling tasks. This encoding method can directly support structural adjustments by the neighborhood search operator at the chromosome level.

[0118] 5.2 The random generation size is P The initial population. Population size. P The value is 100. For each individual, each personnel group is traversed, and a random allocation scheme from monitoring points to several vehicles within that group is generated, ensuring that each vehicle is assigned at least one monitoring point.

[0119] 5.3 Chromosome Decoding. First, based on the upper-level coding, the monitoring points are divided into different demand groups according to the number of people allocated. The order within each group is determined by the arrangement order of the lower-level coding. Then, each group is processed in sequence. Under the premise of satisfying the vehicle occupancy constraint and time window constraint, the monitoring point is inserted into the current vehicle, and the sampling time of each point is determined according to the number of people allocated, thus completing the vehicle grouping and path construction.

[0120] 5.4 Fitness function calculation. .

[0121] 5.5 The upper-level coding corresponds to the number of personnel allocated, using two-point crossover. During operation, two crossover points are randomly selected, and the gene fragments of the two parent individuals between these two points are directly exchanged, thus generating two offspring individuals. The lower-level coding corresponds to the monitoring point access order, using sequential crossover. During operation, two crossover points are first randomly selected, and the sequential fragment of the monitoring point in that interval from one parent individual is directly copied to the corresponding position in the offspring; then, monitoring points that have not yet appeared in the offspring are extracted from the other parent individual in the original order and filled sequentially along the remaining empty positions in the offspring. The crossover rate is 0.8. 5.6 The upper-level coding employs exchange mutation, randomly selecting two monitoring point locations and swapping their corresponding personnel allocation numbers. The lower-level coding employs reverse transcription mutation, randomly selecting two locations and completely reversing the visiting order of the monitoring points between these two locations, thereby generating a new path structure. The mutation rate is 0.15.

[0122] 5.7 Integrating two neighborhood search operators to perform local search optimization.

[0123] 5.7.1 A personnel selection probability operator is used to dynamically allocate sampling personnel at monitoring points. Since sampling time decreases non-linearly with the number of people allocated, different personnel allocation probabilities should be matched to monitoring points with different workloads. This method divides the monitoring point into three intervals based on the single-person sampling time and presets the personnel selection probability for each interval, making local adjustments to the personnel allocation while keeping the vehicle path unchanged. The specific steps of the personnel selection probability operator are as follows: Step 1: Traverse all monitoring points i ∈ C The system uses a set probability to determine whether to adjust personnel at the monitoring point.

[0124] Step 2: For the selected monitoring points i Based on the time it takes for an individual to complete a sampling task t i s The number of people at that point will be randomly reassigned. y i n .like t is < t 1, then with p 1 1 The probability allocation for 1 person p 1 2 The probability allocation for 2 people p 1 3 The probability is allocated to 3 people; if t 1≤ t i s ≤ t 2, then with p 2 1 The probability allocation for 1 person p 2 2 The probability allocation for 2 people p 2 3 The probability is allocated to 3 people; if t i s > t 2, then with p 3 1 The probability allocation for 1 person p 3 2 The probability allocation for 2 people p 3 3 The probability is allocated to 3 people.

[0125] Step 3: After updating the personnel allocation, keep the current vehicle routes unchanged and adjust the allocation based on the new number of personnel. y i n Recalculate the monitoring point i Sampling time t i And check that the operating time of each vehicle does not exceed T max With vehicle passenger capacity y i n No more than Q .

[0126] Step 4: If the adjusted total operating costs decrease and all constraints are met, accept the adjustment and update. y i = y i n Otherwise, revert to the original personnel allocation plan. y i .

[0127] Step 5: Repeat the above process until all monitoring points have been adjusted once or the maximum number of iterations has been reached.

[0128] 5.7.2 A load balancing mover is used to optimize the task load between vehicles. Since vehicle load is positively correlated with the number of monitoring points it visits, this method identifies the vehicles with the highest and lowest loads by statistically analyzing the number of monitoring points served by each vehicle in the sampling task. Then, a small number of monitoring point tasks are transferred between these two types of vehicles to optimize the search, making it easier to obtain candidate solutions that perform better in terms of vehicle load balance. The specific steps of the load balancing mover are as follows: Step 1: Calculate the statistics for each vehicle in the current plan. v ∈ V Number of monitoring points accessed | C v | Identify the vehicles that visited the monitoring points the most. v max and the fewest number of vehicles v min .

[0129] Step 2: From the vehicle v max monitoring point set C v max One monitoring point is selected in sequence. i ∈ C v max Try to take it from C v max Remove from the path and insert into the vehicle. v min Any feasible position in the path.

[0130] Step 3: Set up monitoring points i Personnel allocation number y i Adjusted to vehicles v min Current staffing standards y v min Recalculate the total operating cost of the two vehicles. If the total cost decreases after the relocation, and the total number of personnel does not exceed the total number of available personnel, then... p The working time of each vehicle shall not exceed T max If so, accept the move and update. C v max , C v min as well as y i Otherwise, revert to the original plan.

[0131] Step 4: Repeat the above process until no further improvement can be made or the maximum number of iterations is reached. 5.8 A tournament selection method based on the inverse of fitness combined with an elite retention strategy is adopted for individual selection, and 5 optimal individuals are retained in each generation to directly enter the next generation.

[0132] The algorithm is set to terminate when the population has iterated to 200 generations, thus obtaining the final monitoring path scheme.

[0133] 6. Output the planning scheme. This mainly includes: core environmental monitoring indicators: vehicle usage cost of 400 yuan, travel cost of 1692.65 yuan, personnel cost of 2000 yuan, as well as sampling vehicles, number of samplers, optimal sampling route, sampling duration, and vehicle travel time. Table 1 shows the monitoring scheme in this embodiment.

[0134] Table 1 Monitoring Plan

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for coordinating and optimizing the allocation of environmental monitoring personnel and vehicle routes, considering sampling time, characterized in that, The method specifically includes the following steps: S1. Obtain environmental monitoring sampling task information within the planning period; the information includes information on the testing center, monitoring points, sampling personnel, and monitoring vehicles; S2. Convert the format of the information data obtained in step S1, construct a set of detection centers, a set of monitoring points, a set of sampling personnel, and a set of monitoring vehicles, and construct an environmental monitoring network diagram with detection centers and monitoring points as nodes and road segments between nodes as arcs. S3. Establish a collaborative optimization model for environmental monitoring personnel allocation and vehicle routing, taking into account sampling time, with the objective of minimizing total cost; wherein the total cost includes vehicle usage cost, personnel usage cost, and vehicle travel distance cost; the constraints of the model include at least: each monitoring point is visited once, vehicles depart from and return from the monitoring center, traffic balance, the number of personnel at monitoring points does not exceed the maximum capacity of the vehicle, the total number of personnel used does not exceed the total number of sampling personnel, the number of personnel allocated to all monitoring points on the same vehicle is the same, and time window constraints; S4. A genetic algorithm with fusion neighborhood search is used to solve the model established in step S3 to obtain the optimal environmental monitoring personnel allocation and vehicle route collaborative optimization scheme; wherein, the chromosome of the genetic algorithm adopts a two-layer coding structure, the upper layer coding represents the number of sampling personnel allocated to each monitoring point, and the lower layer coding represents the access order of the monitoring points; the neighborhood search includes a personnel selection probability operator for optimizing personnel allocation and a load balancing movement operator for optimizing the task load between vehicles; S5. Output the optimal solution obtained in step S4. The solution shall include at least the monitoring path of each vehicle, the order of access to monitoring points, the number of sampling personnel assigned to each vehicle, and cost statistics.

2. The method for collaborative optimization of environmental monitoring personnel allocation and vehicle routes considering sampling time, as described in claim 1, is characterized in that... In step S1, the testing center information includes its geographical location and sampling operation time. The geographical location is obtained through a geographic information system, including map software, which provides its latitude and longitude coordinates. The sampling operation time is determined according to the testing center's operation and management system, specifying the daily sampling operation periods, including the start time. With the latest end time ; The monitoring point information includes the geographical location of each monitoring point, the average sampling time, and the number of people assigned to each monitoring point. The geographical location of the monitoring points is obtained through a geographic information system, including map software, which provides their latitude and longitude coordinates; the average sampling time is determined by past sampling experience. The sampling personnel information includes the total number of sampling personnel entered. p Each sampling personnel is assigned a unique identification number; the maximum daily sampling time for each personnel is also included. and personnel usage costs ; The monitored vehicle information includes key vehicle parameters determined by querying the testing center, and vehicle usage costs. Cost per kilometer while the vehicle is in motion Maximum capacity per vehicle Q .

3. The method for collaborative optimization of environmental monitoring personnel allocation and vehicle routes considering sampling time, as described in claim 2, is characterized in that... In step S2, the process of converting the information data format and constructing the sets of testing centers, monitoring points, sampling personnel, and monitoring vehicles specifically includes assigning unique numbers to each testing center, monitoring point, and vehicle, where the testing center number is 0. c The monitoring points are numbered sequentially according to their entry order: 1, 2, 3... c Therefore, the set of testing centers is represented by {0}, and the set of monitoring points is represented by... C ={1,2,3...... c The center has a total of p A set of sampling personnel is represented as follows: P And | P |= p The set of monitored vehicles is represented as V ; The construction of the environmental monitoring network diagram, with detection centers and monitoring points as nodes and road segments between nodes as arcs, specifically includes: 1) Construct a monitoring network diagram G = ( N , E ), N Let be the set of nodes in this network. E Let be the set of arcs in this network; 2) Determine the set of nodes N Monitoring network G The node set consists of a set of monitoring points and a set of detection center nodes, represented as... N = C ∪{0}; 3) Determine the arc set E Define nodes i and nodes j If the road segment between them is an arc, then the monitoring network... G The set of arcs is E ={( i , j )| i ∈ N , j ∈ N , i ≠ j }; 4) Determine the distances between different locations. and vehicle speed v Based on the latitude and longitude coordinates of each node, the actual distance between each node is obtained through map system tools. Simultaneously, the average vehicle speed was determined by combining historical traffic data and road network attributes. v .

4. The method for collaborative optimization of environmental monitoring personnel allocation and vehicle routes considering sampling time, as described in claim 3, is characterized in that... In step S3, an environmental monitoring personnel allocation and vehicle routing coordination model considering sampling time is established with the objective of minimizing the total cost. The total cost is the sum of vehicle usage cost, personnel usage cost, and distance cost. The model's objective function is: Minimize cost: in: Indicates minimizing cost. This indicates the cost of using the vehicle. i , j Indicates the monitoring point number. C Represents the set of monitoring points. E Let the set of arcs in this network be... Indicates whether the vehicle has traveled from inspection center 0 to monitoring point. i , Indicates the cost of using personnel. Indicates monitoring point i The number of sampling personnel This indicates the cost per kilometer the vehicle travels. Indicates from monitoring point i Drive to the monitoring point j The distance between them Indicates whether the vehicle departed from the monitoring point i Drive to the monitoring point j ; The constraints of the model include: each monitoring point must be visited once; vehicles depart from and return to the monitoring center; flow balance; the actual number of vehicles used does not exceed the total number of available vehicles; personnel constraints at monitoring points; total personnel limit; the same number of personnel assigned to all monitoring points on the same vehicle; the time of arrival at the next monitoring point is no earlier than the time of departure from the current monitoring point plus the travel time; the time a vehicle leaves the monitoring point is equal to the sum of the arrival time and the time required to complete sampling at that point; the time a vehicle arrives at each monitoring point is no earlier than the sampling start time, and the time it leaves each monitoring point is no later than the sampling end time; the departure time of all vehicles from the monitoring center is the specified sampling start time; time capacity constraints; sub-loop elimination constraints.

5. The method for coordinating and optimizing the allocation of environmental monitoring personnel and vehicle routes considering sampling time, as described in claim 4, is characterized in that... In step S4, a genetic algorithm incorporating neighborhood search is employed. This algorithm uses a genetic algorithm as its main framework and optimizes the allocation of sampling personnel and monitoring points through neighborhood search operators, guiding the population to evolve towards a more balanced direction. Specifically, it includes the following steps: S41. Chromosome coding: A two-layer integer coding method is used; the upper layer of the chromosome coding represents the number of personnel allocated, and its length is equal to the total number of monitoring points. C |, The value for each gene locus is the number of sampling personnel assigned to that monitoring point, and the value range is [1, Q ],in Q The maximum number of passengers allowed in the vehicle; the lower-level code represents the access order of the monitoring points, from 1 to | C A permutation of | represents the order in which vehicles perform sampling tasks; S42, the size of the random generation is P The initial population: For each individual, traverse each group of personnel, randomly generate an allocation scheme from monitoring points to several vehicles within that group, and ensure that each vehicle is assigned to at least one monitoring point; S43. Chromosome Decoding and Fitness Function Calculation: First, based on the upper-level coding, monitoring points are divided into different demand groups according to the allocated number of people. The order within each group is determined by the arrangement order of the lower-level coding. Then, each group is processed sequentially. Under the premise of satisfying vehicle occupancy constraints and time window constraints, monitoring points are inserted into the current vehicle, and the sampling time for each point is determined according to the allocated number of people, completing vehicle grouping and path construction; the fitness function is calculated as follows: ;in: Represents the fitness value. Indicates minimizing cost; S44. The upper-level coding corresponds to the number of personnel allocated, using two-point crossover. During operation, two crossover points are randomly selected, and the gene fragments of the two parent individuals are directly exchanged between these two points to generate two offspring individuals. The lower-level coding corresponds to the access order of monitoring points, using sequential crossover. During operation, two crossover points are first randomly selected, and the sequential fragment of the monitoring point in the interval of one parent individual is directly copied to the corresponding position in the offspring. Then, monitoring points that have not yet appeared in the offspring are extracted from the other parent individual in the original order and filled in sequentially along the remaining empty spaces in the offspring. S45. The upper-level coding uses exchange mutation, randomly selecting two monitoring point locations and exchanging the corresponding personnel allocation numbers; the lower-level coding uses reverse transcription mutation, randomly selecting two locations and completely reversing the access order of the monitoring points between these two locations, thereby generating a new path structure. S46. Integrate two neighborhood search operators to perform local search optimization; S47. Individual selection is performed using a tournament selection method based on the inverse of fitness combined with an elite retention strategy; the termination condition of the algorithm is set to stop when the population reaches the maximum number of iterations, thus obtaining the final monitoring path scheme.

6. The method for collaborative optimization of environmental monitoring personnel allocation and vehicle routes considering sampling time, as described in claim 5, is characterized in that... In step S46, the fusion of the two neighborhood search operators to perform local search optimization specifically includes: 1) Dynamically allocate sampling personnel at monitoring points using a personnel selection probability operator: Divide the monitoring point into three intervals based on the individual sampling time, and preset the personnel selection probability for each interval. Adjust the personnel allocation locally while keeping the vehicle route unchanged. 2) Optimize the task load between vehicles by using a load balancing mover: By counting the number of monitoring points served by each vehicle in the sampling task, identify the vehicle with the highest load and the vehicle with the lowest load, and then optimize the transfer of a small number of monitoring point tasks between the two types of vehicles, making it easier to obtain candidate solutions that perform better in terms of vehicle load balance index.

7. The method for collaborative optimization of environmental monitoring personnel allocation and vehicle routes considering sampling time, as described in claim 6, is characterized in that... The specific steps for personnel to select probability operators are as follows: 1.1) Traverse all monitoring points i ∈ C The probability is set to determine whether personnel adjustments should be made at the monitoring point; 1.2) For the selected monitoring points i Based on the time it takes for an individual to complete a sampling task t i s The number of people at that point will be randomly reassigned. y i n ; like t i s < t 1, then with p 1 1 The probability allocation for 1 person p 1 2 The probability allocation for 2 people p 1 3 The probability is allocated to 3 people; if t 1≤ t i s ≤ t 2, then with p 2 1 The probability allocation for 1 person p 2 2 The probability allocation for 2 people p 2 3 The probability is allocated to 3 people; if t i s > t 2, then with p 3 1 The probability allocation for 1 person p 3 2 The probability allocation for 2 people p 3 3 The probability is allocated to 3 people; 1.3) After updating personnel allocation, keep the current vehicle routes unchanged and adjust the allocation based on the new number of personnel. y i n Recalculate the monitoring point i Sampling time t i And check that the operating time of each vehicle does not exceed T max With vehicle passenger capacity y i n No more than Q ; 1.4) If the adjusted total operating costs decrease and all constraints are met, then the adjustment is accepted and updated. y i = y i n ; Otherwise, revert to the original personnel allocation plan. y i ; 1.5) Repeat the above process until all monitoring points have completed one adjustment or the maximum number of iterations has been reached.

8. The method for collaborative optimization of environmental monitoring personnel allocation and vehicle routes considering sampling time, as described in claim 6, is characterized in that... The specific steps of the load balancing shift operator are as follows: 2.1) Statistics on each vehicle in the current plan v ∈ V Number of monitoring points accessed | C v | Identify the vehicles that visited the monitoring points the most. v max and the fewest number of vehicles v min ; 2.2) From the vehicle v max monitoring point set C v max One monitoring point is selected in sequence. i ∈ C v max Try to get it from C v max Remove from the path and insert into the vehicle. v min Any feasible position in the path; 2.3) Monitoring points i Personnel allocation number y i Adjusted to vehicles v min Current staffing standards y v min Recalculate the total operating cost of the two vehicles. If the total cost decreases after the relocation, and the total number of personnel does not exceed the total number of available personnel, then... p The working time of each vehicle shall not exceed T max If so, accept the move and update. C v max , C v min as well as y i Otherwise, revert to the original plan; 2.4) Repeat the above process until no further improvement can be made or the maximum number of iterations is reached.

9. The method for collaborative optimization of environmental monitoring personnel allocation and vehicle routes considering sampling time, as described in claim 1, is characterized in that... In step S5, the output planning scheme mainly includes: (1) detailed monitoring path, which clearly shows the vehicle monitoring path, the order of access to monitoring points and the number of sampling personnel on each vehicle in the form of a table, and calculates the total working time of each vehicle in the cycle; (2) core indicators: vehicle usage cost, driving cost and personnel cost.

10. A system for collaborative optimization of environmental monitoring personnel allocation and vehicle routing considering sampling time, characterized in that, The system employs the method as described in any one of claims 1 to 9, and the system comprises the following modules: 1) Data Input Module: Acquires and stores sampling task data, including information on the testing center, monitoring points, vehicles, and sampling personnel; 2) Data processing module: Converts the format of the input data and constructs a set of monitoring point nodes, a set of detection center nodes, a set of vehicles, and a set of sampling personnel; 3) Delivery network establishment module: Based on the aforementioned basic information dataset, the addresses of the testing center and all monitoring points are abstracted into network nodes to construct an environmental monitoring network diagram; 4) Model building module: Based on the environmental monitoring network diagram and basic information dataset, complete the setting of the objective function and constraints of the model, and establish an environmental monitoring sampling task path planning model; 5) Optimization and solution module: The genetic algorithm with fusion neighborhood search is used to solve the environmental monitoring sampling task path model. Through iterative optimization, the environmental monitoring sampling task path scheme is finally obtained. 6) Solution Output Module: Outputs the final delivery route solution information, including the monitoring routes of each vehicle, vehicle operation time, monitoring point access order, cost statistics, etc.

Images