A water conservancy and hydropower construction resource scheduling method and system

Abstract

The application discloses a water conservancy and hydropower construction resource scheduling method and system, relates to the technical field of construction resource scheduling, and comprises the following steps: vehicle utilization and team productivity are calculated according to construction data; a three-section code is constructed according to a preset construction task; a construction resource scheduling scheme is formed based on the three-section code; a conflict evaluation function and an objective function are constructed; an initial population is randomly generated based on the construction resource scheduling scheme; selection operation, crossover operation and mutation operation are sequentially executed on the initial population based on a genetic algorithm to obtain a first-generation population; an optimal fitness value is selected from individual fitness values calculated from the first-generation population based on the conflict evaluation function and the objective function; an iteration condition is preset; when the iteration condition is met, iteration operation is executed; when the iteration condition is not met, decoding operation is performed on an individual with the optimal fitness value in the first-generation population and conflict resolution is executed to obtain an optimal construction resource scheduling scheme, thereby improving the overall benefit of the scheduling scheme.

Description

Technical Field

[0001] This invention relates to the field of construction resource scheduling technology, and in particular to a method and system for scheduling water conservancy and hydropower construction resources. Background Technology

[0002] In recent years, with the promotion of the concept of smart construction sites, optimizing the scheduling of construction resources using information and intelligent technologies has become an important development direction for the industry. Traditional methods of scheduling and dispatching vehicles based on manual experience are difficult to cope with the dynamic changes in the construction site and the complex constraints of multi-resource coupling. This can easily lead to problems such as vehicle idleness, idle work in work teams, and conflicts in work processes, which restricts the improvement of overall construction efficiency. Therefore, developing an intelligent method that can comprehensively consider vehicle utilization and work team productivity and automatically generate efficient and conflict-free scheduling schemes is of great practical significance for promoting the refinement and intelligence of construction management in water conservancy and hydropower projects.

[0003] However, existing resource scheduling optimization techniques still have significant shortcomings. On the one hand, most studies or systems treat vehicle scheduling and personnel scheduling as independent problems, lacking integrated modeling and collaborative optimization of the closely coupled transportation-operation link. This leads to resource conflicts with poor coordination in actual implementation. On the other hand, even considering multi-resource collaboration, existing algorithms often fail to effectively balance scheduling benefits and scheme feasibility when facing complex spatiotemporal constraints and multi-objective optimization due to unreasonable coding methods or a single optimization mechanism. They either generate a large number of invalid solutions that violate actual constraints or get stuck in local optima, failing to obtain a globally superior and practical scheduling scheme. Summary of the Invention

[0004] The technical problem addressed by this invention is that existing resource scheduling optimization techniques still have significant shortcomings. On the one hand, most studies or systems treat vehicle scheduling and personnel scheduling as independent problems, lacking integrated modeling and collaborative optimization of the closely coupled transportation-operation link. This leads to resource conflicts with poor coordination in actual implementation. On the other hand, even considering multi-resource collaboration, existing algorithms often fail to effectively balance scheduling benefits and scheme feasibility when facing complex spatiotemporal constraints and multi-objective optimization due to unreasonable coding methods or a single optimization mechanism. They either generate a large number of invalid solutions that violate actual constraints or get stuck in local optima, failing to obtain a globally superior and practical scheduling scheme.

[0005] To solve the above technical problems, the present invention provides the following technical solution: a method for scheduling water conservancy and hydropower construction resources includes the following steps: step S1, calculating vehicle utilization rate and team productivity based on construction data;

[0006] Step S2: Construct a three-segment code based on the preset construction tasks, and form a construction resource scheduling scheme based on the three-segment code;

[0007] Step S3: Construct the conflict evaluation function and the objective function;

[0008] An initial population is randomly generated based on the construction resource scheduling scheme. The initial population is then subjected to selection, crossover and mutation operations in sequence using a genetic algorithm to obtain the first generation population.

[0009] Step S4: Select the optimal fitness value from the individual fitness values ​​calculated by the conflict evaluation function and objective function in the first generation population, and preset the iteration conditions. If the iteration conditions are met, the iteration operation is executed. If the iteration conditions are not met, the individual with the optimal fitness value in the first generation population is decoded and the conflict is resolved to obtain the optimal construction resource scheduling scheme.

[0010] As a preferred embodiment of the water conservancy and hydropower construction resource scheduling method of the present invention, step S1 involves collecting construction data, which includes vehicle work data and work group data.

[0011] Vehicle operational data includes total transport volume, number of transport trips, vehicle rated capacity, effective vehicle operating time, and planned working time.

[0012] Team data includes team output, number of personnel involved, working hours, number of certified workers, and total number of team members;

[0013] Calculate vehicle utilization rate and team productivity by analyzing vehicle working data and team data;

[0014] The calculation process of the vehicle utilization rate is as follows: the ratio of the total transportation volume to the number of transportation trips is recorded as the first ratio, and the first ratio is the average loading volume per trip.

[0015] The ratio of average single-trip loading to vehicle rated capacity is recorded as the second ratio, the ratio of effective vehicle operating time to planned working time is recorded as the third ratio, and the product of the second ratio and the third ratio is recorded as the first product, which is the vehicle utilization rate.

[0016] The calculation process for the team's productivity is as follows: the product of the number of people involved and the working time is recorded as the second product, which is the total working hours; the ratio of the team's output to the second product is recorded as the fourth ratio; and the ratio of the fourth ratio to the total number of people in the team is recorded as the fifth ratio. The fifth ratio is the team's productivity.

[0017] As a preferred embodiment of the water conservancy and hydropower construction resource scheduling method of the present invention, step S2 involves constructing a three-segment code based on the construction task and forming a construction resource scheduling scheme based on the three-segment code.

[0018] The construction tasks include vehicle transportation tasks, work group operation sequence tasks, and task duration adaptation tasks.

[0019] Each construction task corresponds to a unique construction task ID. Vehicle transportation tasks, work group operation sequence tasks, and task duration adaptation tasks under the same construction task have the same ID.

[0020] The three-segment coding includes the vehicle transportation allocation segment, the team operation sequence segment, and the task duration adaptation segment.

[0021] The three-segment encoding is binary, in which the number of encoding gene bits is the same for the vehicle transportation allocation segment, the team operation sequence segment, and the task duration adaptation segment.

[0022] The vehicle transportation allocation section, the team operation sequence section, and the task duration adaptation section correspond to the vehicle transportation task, the team operation sequence task, and the task duration adaptation task, respectively.

[0023] Step S2 includes steps S201, S202 and S203;

[0024] The construction logic of step S201 is as follows: for the vehicle transportation allocation segment, each gene bit corresponds to one vehicle transportation task in ascending order of vehicle transportation task number, thus obtaining the vehicle transportation allocation coding sequence.

[0025] One vehicle transportation task corresponds to one vehicle number, and one vehicle number corresponds to one vehicle.

[0026] The construction logic of step S202 is as follows: for the work sequence segment of the work group, the gene position corresponds to one work sequence task in the order of the work sequence task number from smallest to largest.

[0027] Each gene locus is a unique work group task sequence code. Work group tasks are sorted according to the binary natural size order of the work group task sequence code. When multiple work group task sequences have the same work group task sequence code, the work group task sequence with the smaller work group task sequence number is selected for priority execution, thus obtaining the work group task sequence code sequence.

[0028] The construction logic of step S203 is as follows: for the task duration adaptation segment, the gene bit corresponds to one task duration adaptation task in the order of task duration adaptation task number, and the gene bit is a unique task duration code.

[0029] The preset duration range is divided into two continuous and non-overlapping duration intervals. The duration intervals are arranged in ascending order of their lower limits. The task duration codes are mapped to the arranged duration intervals in binary natural size order. Through the one-to-one correspondence between the task duration code sequence and the duration interval, the duration interval of the corresponding task duration is determined, and the task duration adaptation code sequence is obtained.

[0030] Based on the vehicle transportation allocation code sequence corresponding to the vehicle transportation allocation segment, the team operation sequence code sequence corresponding to the team operation sequence segment, and the task duration adaptation code sequence corresponding to the task duration adaptation segment, they are sorted in sequence to form a construction resource scheduling scheme.

[0031] As a preferred embodiment of the water conservancy and hydropower construction resource scheduling method of the present invention, step S3 involves constructing a conflict assessment function and an objective function.

[0032] The expression for the conflict evaluation function is:

[0033] ;

[0034] ;

[0035] ;

[0036] ;

[0037] ;

[0038] in, This represents the conflict assessment function value, where 'a' represents the conflict value for repeated vehicle assignment. This indicates the conflict value in the work sequence of the work group. This indicates a conflict value for task duration adaptation. , , These correspond to the conflict value weights for vehicle duplicate allocation, work group operation sequence, and task duration adaptation, respectively.

[0039] Indicates the total number of vehicles. Indicates the vehicle number. Indicates the duration of vehicle overlap;

[0040] This represents the m-th task number in the work group's operational sequence. This indicates the total number of tasks in the work group's work sequence. This represents the end time of the m-th task in the order of work group operations. Indicates the first The start time of each work group's work sequence tasks;

[0041] This indicates that the task duration is adapted to the task number. This indicates that the task duration is adapted to the total number of tasks. This represents the theoretically shortest duration for a duration-adaptive task. This indicates that the task duration is adapted to the total task capacity, and s represents the maximum capacity of each vehicle. This indicates the planned allocation duration for the i-th task to adapt to the task's duration.

[0042] The objective function expression is: ;

[0043] Where Y represents the scheduling efficiency value, B represents vehicle utilization rate, and A represents team productivity. The corresponding weight representing vehicle utilization rate. This represents the weight corresponding to the productivity of the work team, and and The sum is 1.

[0044] As a preferred embodiment of the water conservancy and hydropower construction resource scheduling method of the present invention, in step S3, an initial population is randomly generated based on the construction resource scheduling scheme, and selection operation, crossover operation and mutation operation are sequentially performed on the initial population based on the genetic algorithm to generate the first generation population;

[0045] Step S3 includes steps S301 and S302;

[0046] Step S301: Based on the vehicle transportation allocation coding sequence, the shift operation sequence coding sequence, and the task duration adaptation coding sequence, the initial population is generated using a random initialization method.

[0047] Each individual in the initial population includes the vehicle transportation allocation coding sequence, the shift operation sequence coding sequence, and the task duration adaptation coding sequence. The number of gene positions in the vehicle transportation allocation coding sequence, the shift operation sequence coding sequence, and the task duration adaptation coding sequence are the same. During the random initialization process, each gene position is independently assigned a corresponding three-segment code.

[0048] The expression for calculating the fitness value of the initial population is: ;

[0049] in, Indicates the fitness value. This represents the penalty coefficient.

[0050] As a preferred embodiment of the water conservancy and hydropower construction resource scheduling method of the present invention, step S302, the logic of performing crossover and mutation operations on the initial population based on the genetic algorithm is as follows: N individuals are selected from the initial population as the set of individuals to be operated on based on the genetic algorithm.

[0051] For the set of individuals to be operated on, crossover and mutation operations are performed based on the adaptive crossover and adaptive mutation probabilities of each individual in the set, according to the genetic algorithm.

[0052] The expression for the adaptive crossover probability is:

[0053] ;

[0054] The expression for the adaptive mutation probability is:

[0055] ;

[0056] in, Indicates the adaptive crossover probability. This represents the adaptive mutation probability. This represents the maximum fitness value of an individual in the current population. This represents the average fitness value of all individuals in the current population. This indicates that the parent individual with the higher fitness value in the crossover operation has a higher fitness value. This represents the fitness value of the individual after the mutation operation. , , and All are constants;

[0057] The individuals that have undergone crossover and mutation operations are replaced with the individuals in the initial population to obtain the first generation population.

[0058] As a preferred embodiment of the water conservancy and hydropower construction resource scheduling method of the present invention, in step S4, the optimal fitness value is selected from the individual fitness values ​​calculated by the first generation population based on the conflict evaluation function and the objective function, and the iteration conditions are preset. When the iteration conditions are met, the iteration operation is performed. When the iteration conditions are not met, the individual with the optimal fitness value in the first generation population is decoded and the conflict is resolved to obtain the optimal construction resource scheduling scheme.

[0059] The selection logic for the optimal fitness value is as follows: the fitness value of each individual in the first generation population is compared, and the maximum value among the fitness values ​​is selected as the optimal fitness value.

[0060] Step S4 includes steps S401, S402 and S403;

[0061] Step S401: After each generation T population is generated, calculate the fitness value of individuals in the generation T population, and obtain the optimal fitness value of individuals in the generation T population according to the selection logic of the optimal fitness value.

[0062] The preset iteration condition is specifically to compare the optimal fitness value with the optimal fitness value of an individual in the (T-1)th generation population;

[0063] If the optimal fitness value of an individual in the T-th generation population is greater than the optimal fitness value of an individual in the (T-1)-th generation population, then the optimal fitness value of the individual in the (T-1)-th generation population is updated to the optimal fitness value of the individual in the T-th generation population, and the stall generation counter is reset to zero; otherwise, the stall generation counter is incremented by one.

[0064] If the stall generation counter does not reach the preset threshold, the Tth generation population is used as the parent population for the next iteration, T=T+1, and the Tth generation population is used as the parent population. Then, the logic for the crossover and mutation operations in step S302 is returned.

[0065] If the stall generation counter reaches a preset threshold, then the individual with the best fitness value in the Tth generation population is decoded and conflict resolution is performed.

[0066] As a preferred embodiment of the water conservancy and hydropower construction resource scheduling method of the present invention, in step S402, the individual with the best fitness value in the Tth generation population is decoded to obtain the corresponding construction resource scheduling scheme.

[0067] The decoding process specifically includes decoding the vehicle transportation allocation coding sequence, the shift operation sequence coding sequence, and the task duration adaptation coding sequence in the three-segment coding.

[0068] Decode the vehicle transportation allocation code sequence, read the vehicle transportation allocation code sequence of the vehicle transportation allocation segment, convert the vehicle transportation allocation code sequence into vehicle number, use the vehicle number as matrix element, and construct the vehicle transportation task-vehicle number matrix, denoted as matrix L;

[0069] Each row of the matrix L corresponds to a vehicle transportation task, and each column corresponds to a vehicle number.

[0070] Decode the work order encoding sequence of the work group, read the work order encoding sequence of the work group sequence segment, convert the work order encoding sequence of the work group sequence into work order task number, use the work order task number as matrix element, construct the work group task-work order task encoding matrix, denoted as matrix S;

[0071] Each row of the matrix S corresponds to a work group's work sequence task, and each column corresponds to a work group's work sequence task code.

[0072] Decode the task duration adaptation encoding sequence, read the task duration adaptation encoding sequence of the task duration adaptation segment, convert the task duration adaptation encoding sequence into a duration interval according to the one-to-one correspondence between the task duration adaptation encoding sequence and the duration interval, and use the duration interval as matrix elements to construct a task duration adaptation task-duration interval matrix, denoted as matrix X;

[0073] Each row of matrix X corresponds to a task duration adaptation task, and each column corresponds to a duration interval.

[0074] Based on the IDs of vehicle transportation tasks, work group operation sequence tasks, and task duration adaptation tasks, matrices L, S, and X are associated to obtain the vehicle number, work group operation sequence code, and duration range corresponding to the construction resource scheduling scheme.

[0075] As a preferred embodiment of the water conservancy and hydropower construction resource scheduling method of the present invention, step S403, based on the construction resource scheduling scheme obtained by decoding, performs conflict resolution, specifically including extracting the initial decision variable vector of the construction resource scheduling scheme;

[0076] The initial decision variable vector is a set of key parameters extracted from the construction resource scheduling scheme, specifically including the vehicle number corresponding to each vehicle transportation task, the work sequence code corresponding to each shift work sequence task, and the duration interval corresponding to each task duration adaptation task.

[0077] Using the construction task ID as an index, the vehicle number, work sequence code, and duration range corresponding to the same construction task ID are used as sub-vectors;

[0078] Using the initial decision variable vector as the initial search point of the simulated annealing algorithm, local optimization is performed to obtain the local optimal solution;

[0079] The local optimal solution is re-executed sequentially through the construction logic of step S201, step S202, and step S203 to convert the local optimal solution into a three-segment encoding format, denoted as the local optimal individual code.

[0080] Then, replace any ordinary individual in the current T-generation population, except for the individual with the best fitness value, with the locally optimal individual code.

[0081] When the preset iteration termination condition is met, the individual with the best fitness value in the Tth generation population is output, and the individual is decoded to obtain the optimal construction resource scheduling scheme.

[0082] The specific iteration termination condition is that the stall generation counter reaches a preset threshold, and the individual with the best fitness value in the T-th generation population has performed conflict resolution.

[0083] A water conservancy and hydropower construction resource scheduling system includes a calculation module, a construction module, a generation module, and a scheduling module;

[0084] The calculation module calculates vehicle utilization and work team productivity based on construction data.

[0085] The module constructs a three-segment code based on the preset construction tasks, and forms a construction resource scheduling scheme based on the three-segment code;

[0086] Generate modules to construct conflict evaluation functions and objective functions;

[0087] An initial population is randomly generated based on the construction resource scheduling scheme. The initial population is then subjected to selection, crossover and mutation operations in sequence using a genetic algorithm to obtain the first generation population.

[0088] The scheduling module selects the optimal fitness value from the individual fitness values ​​calculated by the conflict evaluation function and objective function in the first generation population, and presets the iteration conditions. When the iteration conditions are met, the iteration operation is executed. When the iteration conditions are not met, the individual with the optimal fitness value in the first generation population is decoded and the conflict is resolved to obtain the optimal construction resource scheduling scheme.

[0089] The beneficial effects of this invention are as follows: By designing a three-segment unified coding structure associated with task IDs, the problem of coordination and coupling representation of vehicle and work team resources in scheduling schemes is fundamentally solved. This coding integrates and associates vehicle allocation, work sequence, and task duration information for the same construction task, ensuring the logical integrity and direct executability of the scheduling scheme. It overcomes the inherent defects of traditional methods, such as resource planning disconnection and frequent timing conflicts caused by separate coding or modeling. Based on this coding structure, an adaptive genetic algorithm is used for global optimization. Its crossover and mutation probabilities can be dynamically adjusted according to individual fitness, effectively balancing the breadth and depth of the search. This avoids the problems of fixed parameters and premature convergence of traditional algorithms, thus more efficiently searching for high-quality scheduling schemes that comprehensively improve vehicle utilization and work team productivity. In addition, the simulated annealing algorithm is introduced to resolve conflicts of local optima that occur during the optimization process, further ensuring the high feasibility of the final output scheme. This series of innovative technologies with three-segment coding as the core have jointly achieved a reliable and intelligent solution to the multi-resource collaborative scheduling problem in water conservancy and hydropower projects, significantly improving the overall efficiency and practical value of the scheduling scheme. Attached Figure Description

[0090] Figure 1 This is a flowchart illustrating the steps of a water conservancy and hydropower construction resource scheduling method according to an embodiment of the present invention.

[0091] Figure 2 This is a basic flowchart of a water conservancy and hydropower construction resource scheduling system provided in one embodiment of the present invention. Detailed Implementation

[0092] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0093] Example 1, referring to Figure 1 As an embodiment of the present invention, a method for scheduling water conservancy and hydropower construction resources is provided, including the following steps: Step S1, calculating vehicle utilization rate and team productivity based on construction data;

[0094] Step S2: Construct a three-segment code based on the preset construction tasks, and form a construction resource scheduling scheme based on the three-segment code;

[0095] Step S3: Construct the conflict evaluation function and the objective function;

[0096] An initial population is randomly generated based on the construction resource scheduling scheme. The initial population is then subjected to selection, crossover and mutation operations in sequence using a genetic algorithm to obtain the first generation population.

[0097] Step S4: Select the optimal fitness value from the individual fitness values ​​calculated by the conflict evaluation function and objective function in the first generation population, and preset the iteration conditions. If the iteration conditions are met, the iteration operation is executed. If the iteration conditions are not met, the individual with the optimal fitness value in the first generation population is decoded and the conflict is resolved to obtain the optimal construction resource scheduling scheme.

[0098] In one embodiment, this scheme is an integrated collaborative scheduling scheme that addresses the comprehensive resource scheduling problem common in water conservancy and hydropower projects, which involves continuous construction activities with vehicle transportation and on-site team operations as the core components. Based on vehicle work data and team data from the construction data, it calculates vehicle utilization and team productivity, providing core indicators for optimization. Furthermore, it innovatively employs a three-segment coding method to model construction tasks in an integrated manner. This three-segment coding is a chromosome data structure used in genetic algorithms, not a traditional mathematical function. It integrates and correlates the three types of decision information corresponding to each construction task—vehicle allocation, team operation sequence, and planned task duration—through binary coding. This represents a complete resource scheduling scheme as an individual that can be directly processed by the algorithm. By integrating vehicle allocation, team operation sequence, and task duration into a unified individual representation, a complete resource scheduling scheme is directly generated. Based on this, a fitness function is constructed to simultaneously evaluate scheme conflicts and comprehensive benefits. The degree function is composed of a conflict evaluation function and an objective function. The conflict evaluation function is a penalty function used to quantify resource conflicts and logical contradictions in the scheduling scheme, such as overlapping vehicle times and reversed work processes. The higher the value, the worse the feasibility of the scheme. Its purpose is to automatically identify and penalize infeasible solutions during the optimization process. The objective function is a benefit function used to quantify the overall performance of the scheduling scheme in terms of resource utilization efficiency, integrating vehicle utilization rate and team productivity. The higher the value, the better the economic benefit of the scheme. Its purpose is to define a clear direction for maximizing benefits for the optimization algorithm. Together, the two functions guide the algorithm to search for efficient and feasible scheduling schemes and generate an initial scheme population based on this. Iterative optimization is performed through an adaptive genetic algorithm, including selection, crossover, and mutation operations, to gradually evolve into a better scheduling scheme. Finally, an iteration termination condition is set, the optimal individual obtained by evolution is decoded, and conflict resolution is performed, thereby outputting an optimal construction resource scheduling scheme that balances high resource utilization efficiency and strong feasibility.

[0099] Step S1: Collect construction data, including vehicle working data and work team data;

[0100] Vehicle operational data includes total transport volume, number of transport trips, vehicle rated capacity, effective vehicle operating time, and planned working time.

[0101] Team data includes team output, number of personnel involved, working hours, number of certified personnel, and total number of team members;

[0102] Calculate vehicle utilization rate and team productivity by analyzing vehicle working data and team data;

[0103] The calculation process for vehicle utilization rate is as follows: the ratio of total transport volume to transport trips is recorded as the first ratio, and the first ratio is the average load per trip.

[0104] The ratio of average single-trip loading to vehicle rated capacity is recorded as the second ratio, the ratio of effective vehicle operating time to planned working time is recorded as the third ratio, and the product of the second ratio and the third ratio is recorded as the first product, which is the vehicle utilization rate.

[0105] The specific process for calculating the productivity of a work group is as follows: the product of the number of people involved and the working time is recorded as the second product, which is the total working hours; the ratio of the work group output to the second product is recorded as the fourth ratio; the ratio of the fourth ratio to the total number of people in the work group is recorded as the fifth ratio; and the fifth ratio is the productivity of the work group.

[0106] In one embodiment, step S1 aims to provide precise quantitative input for resource scheduling. First, construction data is collected, specifically including vehicle work data (total transport volume, number of transport trips, vehicle rated capacity, effective vehicle operating time, planned working time) and work team data (work team output, number of personnel involved, working time, number of certified personnel, total number of personnel in the work team). The vehicle rated capacity refers to the maximum safe loading volume per trip, and the effective vehicle operating time is determined according to the definition of "pure working time of machinery" in the "Standardized Quota for Construction Machinery in Water Conservancy Projects" (2025 edition). It is a supporting quota for the Ministry of Water Resources' "Regulations on the Compilation of Design Estimates (Estimations) for Water Conservancy Projects" (Water Resources Department

[2024] No. 323), which will be implemented from April 1, 2025. It is the current effective standard. Based on vehicle work data and team data, the vehicle utilization rate is calculated. First, the average load per vehicle trip is calculated, and the ratio of this load to the vehicle's rated capacity is recorded as the second ratio, which reflects the loading efficiency. Then, the ratio of the vehicle's effective running time to the planned working time is recorded as the third ratio, which reflects the time utilization efficiency. Finally, the two second ratios and the third ratio are multiplied together to obtain the vehicle utilization rate, which comprehensively reflects the vehicle loading and time utilization level. The calculation of team productivity involves first calculating the total working hours. The ratio of team output to total working hours is recorded as the fourth ratio. The ratio of the fourth ratio to the total number of team members is recorded as the fifth ratio. The fifth ratio is the team productivity. This scheme transforms complex construction performance into two core quantitative indicators: vehicle utilization rate and team productivity. This provides a reliable data foundation for constructing the objective function and achieving precise resource synergy optimization. In addition, as a management prerequisite for implementing this method, before using the above team data for calculation and scheduling, the team must meet the preset qualification compliance rate requirement. The qualification compliance rate requirement is that the ratio of the number of certified personnel in the team to the total number of team members is not less than the standard set according to the engineering safety specifications, such as 95%.

[0107] Step S2: Construct a three-segment code based on the construction task, and form a construction resource scheduling scheme based on the three-segment code;

[0108] The construction tasks include vehicle transportation tasks, work group operation sequence tasks, and task duration adaptation tasks.

[0109] Each construction task corresponds to a unique construction task ID. Vehicle transportation tasks, work group operation sequence tasks, and task duration adaptation tasks under the same construction task have the same ID.

[0110] The three-segment coding includes the vehicle transportation allocation segment, the team operation sequence segment, and the task duration adaptation segment.

[0111] The three-segment encoding is binary, in which the number of encoding gene bits is the same for the vehicle transportation allocation segment, the team operation sequence segment, and the task duration adaptation segment.

[0112] The vehicle transportation allocation section, the team operation sequence section, and the task duration adaptation section correspond to the vehicle transportation task, the team operation sequence task, and the task duration adaptation task, respectively.

[0113] Step S2 includes steps S201, S202 and S203;

[0114] The construction logic of step S201 is as follows: for the vehicle transportation allocation segment, each gene bit corresponds to one vehicle transportation task in ascending order of vehicle transportation task number, thus obtaining the vehicle transportation allocation coding sequence.

[0115] One vehicle transportation task corresponds to one vehicle number, and one vehicle number corresponds to one vehicle.

[0116] The construction logic of step S202 is as follows: for the work sequence segment of the work group, the gene position corresponds to one work sequence task in the order of the work sequence task number from smallest to largest.

[0117] Each gene locus is a unique work group task sequence code. Work group tasks are sorted according to the binary natural size order of the work group task sequence code. When multiple work group task sequences have the same work group task sequence code, the work group task sequence with the smaller work group task sequence number is selected for priority execution, thus obtaining the work group task sequence code sequence.

[0118] The construction logic of step S203 is as follows: for the task duration adaptation segment, the gene bit corresponds to one task duration adaptation task in the order of task duration adaptation task number, and the gene bit is a unique task duration code.

[0119] The preset duration range is divided into two continuous and non-overlapping duration intervals. The duration intervals are arranged in ascending order of their lower limits. The task duration codes are mapped to the arranged duration intervals in binary natural size order. Through the one-to-one correspondence between the task duration code sequence and the duration interval, the duration interval of the corresponding task duration is determined, and the task duration adaptation code sequence is obtained.

[0120] Based on the vehicle transportation allocation code sequence corresponding to the vehicle transportation allocation segment, the team operation sequence code sequence corresponding to the team operation sequence segment, and the task duration adaptation code sequence corresponding to the task duration adaptation segment, they are sorted in sequence to form a construction resource scheduling scheme.

[0121] In one embodiment, according to step S2, the present invention proposes to construct a three-segment code based on construction tasks, and to form a construction resource scheduling scheme based on this code. Construction tasks include vehicle transportation tasks, work group operation sequence tasks, and task duration adaptation tasks. Each construction task corresponds to a unique construction task ID. Vehicle transportation tasks, work group operation sequence tasks, and task duration adaptation tasks for the same construction task share the same ID, ensuring logical association between tasks. The three-segment code uses binary representation and consists of a vehicle transportation allocation segment, a work group operation sequence segment, and a task duration adaptation segment concatenated sequentially. Its core design lies in the fact that the three coding segments have the same number of gene positions, and the index order of the gene positions within each segment strictly corresponds to the ascending order of the construction task numbers. For any construction task ID, its corresponding vehicle allocation, operation sequence, and task duration information are encoded respectively on gene positions with the same index value in the three segments. This structure ensures that the resource and time information of all tasks in an individual scheduling scheme can be completely and correlatedly represented. The specific construction process includes: First, constructing a vehicle transportation allocation segment. In this segment, the gene bits correspond to vehicle transportation tasks with increasing index numbers. Each gene bit is assigned a binary code value, which directly represents the vehicle number assigned to the corresponding task. For example, the preset code 001 corresponds to transport vehicle number 1, thus generating a vehicle transportation allocation code sequence. Second, constructing a shift work sequence segment. Similarly, the gene bits in this segment correspond to shift work sequence tasks with increasing index numbers. Each gene bit is assigned a unique binary sequence code. The execution priority of all shift work sequence tasks is determined by globally ascending sorting the binary values ​​of their corresponding gene bit codes. If the values ​​are the same, the task with the smaller corresponding task number is executed first, thus generating a shift work sequence code sequence that defines the task sequence relationship. Finally, a task duration adaptation segment is constructed. The gene bits in this segment correspond to the task duration adaptation tasks with increasing index numbers. Each gene bit is assigned a binary duration code. The preset total task duration range is divided into 0 consecutive and non-overlapping intervals, such as [0, T1) and [T1, Tmax], and sorted according to the lower limit of the interval. Each duration code is statically mapped to a specific interval based on the size of its binary value. Thus, the duration adaptation code sequence actually assigns a planned duration range to each task. Concatenating the above three code sequences in a fixed order constitutes a binary entity representing a complete scheduling scheme. This coding structure, through a three-segment design of equal length and positional association, tightly couples vehicle resource allocation, work sequence timing, and task time estimation into a unified expression. This not only ensures the logical integrity of the scheduling scheme but also allows it to be directly processed by the evolutionary calculation method of the genetic algorithm, laying a crucial foundation for subsequent global automated optimization.

[0122] Step S3: Construct the conflict evaluation function and the objective function;

[0123] The expression for the conflict evaluation function is:

[0124] ;

[0125] ;

[0126] ;

[0127] ;

[0128] ;

[0129] in, This represents the conflict assessment function value, where 'a' represents the conflict value for repeated vehicle assignment. This indicates the conflict value in the work sequence of the work group. This indicates a conflict value for task duration adaptation. , , These correspond to the conflict value weights for vehicle duplicate allocation, work group operation sequence, and task duration adaptation, respectively.

[0130] Indicates the total number of vehicles. Indicates the vehicle number. Indicates the duration of vehicle overlap;

[0131] This represents the m-th task number in the work group's operational sequence. This indicates the total number of tasks in the work group's work sequence. This represents the end time of the m-th task in the order of work group operations. Indicates the first The start time of each work group's work sequence tasks;

[0132] This indicates that the task duration is adapted to the task number. This indicates that the task duration is adapted to the total number of tasks. This represents the theoretically shortest duration for a duration-adaptive task. This indicates that the task duration is adapted to the total task capacity, and s represents the maximum capacity of each vehicle. This indicates that the duration of the i-th task is adapted to the task plan's allocated duration;

[0133] The objective function expression is: ;

[0134] Where Y represents the scheduling efficiency value, B represents vehicle utilization rate, and A represents team productivity. The corresponding weight representing vehicle utilization rate. This represents the weight corresponding to the productivity of the work team, and and The sum is 1.

[0135] In one embodiment, according to step S3, a conflict evaluation function and an objective function are constructed for scheme evaluation and optimization. The conflict evaluation function acts as a penalty function in this application. The conflict evaluation function is used to quantify the resource and logical conflicts existing in the scheduling scheme, where a, b, and c represent the vehicle repetition allocation conflict value, the shift operation sequence conflict value, and the task duration adaptation conflict value, respectively. , , Each type of conflict is assigned a pre-set weighting coefficient to adjust its importance in the overall assessment.

[0136] The conflict value 'a' for repeated vehicle assignment is calculated as follows: ;

[0137] Where U represents the total number of vehicles, and u is the vehicle number. This represents the duration of the overlapping time intervals between different tasks assigned to the same vehicle u in the decoded scheduling scheme. (Function) ,when If the value is greater than 0, meaning there is time overlap, then the positive values ​​are accumulated; if... If the value is less than 0, it means there is no overlap or the vehicles are exactly connected, so the cumulative value is 0. Therefore, the actual effect of this formula is to accumulate the total conflict duration of all vehicles with overlapping time.

[0138] The formula for calculating the conflict value b of the work sequence in a work group is: ;

[0139] Among them, the summation symbol This indicates traversing all adjacent task pairs in the work order of each work group, i.e., from task 1 to task 2. There are tasks, each m corresponds to the m-th task and the ... There are 1 task pairs, and the summation term is... ; This represents the end time of the m-th task. Indicates the first The formula calculates the start time of each task by comparing it with the end times of preceding tasks. and the start time of subsequent tasks ,like Then, the difference is accumulated, thereby penalizing the logical conflict of reversed process times.

[0140] The task duration adaptation conflict value c is calculated as follows:

[0141] ;

[0142] Where i is the task number for task duration adaptation, and H is the total number of tasks. The time allocation for the i-th task, obtained through three-segment encoding and decoding, is determined by, for example, taking the median of the corresponding time interval. S represents the total transport capacity of the task, and s represents the maximum transport capacity per vehicle. This formula aims to detect the planned time. Is it shorter than the theoretical minimum required time? And cumulative penalties will be imposed on any shortfall, thereby avoiding overly tight schedules that would prevent the transportation requirements from being met;

[0143] The objective function Y is used to comprehensively evaluate the positive benefits of the scheduling scheme, and its expression is: ;

[0144] Where A and B are the vehicle utilization rate and shift productivity calculated in step S1, respectively. and The preset weighting coefficients are used, and they satisfy the following conditions: + =1, thus combining vehicle utilization and team productivity into a comprehensive benefit value.

[0145] In subsequent genetic algorithms, the fitness of individuals is usually determined by the objective function and the conflict evaluation function. This mechanism guides the optimization search direction, effectively suppressing vehicle resource conflicts, process sequence conflicts and unreasonable schedule settings while pursuing high vehicle utilization and high team productivity. This drives the algorithm to finally produce a construction resource scheduling scheme that achieves the optimal balance between efficiency and feasibility.

[0146] Step S3: Randomly generate an initial population based on the construction resource scheduling scheme, and perform selection, crossover and mutation operations on the initial population in sequence based on the genetic algorithm to generate the first generation population;

[0147] Step S3 includes steps S301 and S302;

[0148] Step S301: Based on the vehicle transportation allocation coding sequence, the shift operation sequence coding sequence, and the task duration adaptation coding sequence, an initial population is generated using a random initialization method.

[0149] Each individual in the initial population includes a vehicle transportation allocation coding sequence, a shift work sequence coding sequence, and a task duration adaptation coding sequence. The number of gene positions in the vehicle transportation allocation coding sequence, shift work sequence coding sequence, and task duration adaptation coding sequence are the same. During the random initialization process, each gene position is independently assigned a corresponding three-segment code.

[0150] The expression for calculating the fitness value of the initial population is: ;

[0151] in, Indicates the fitness value. This represents the penalty coefficient.

[0152] Step S302, the logic of performing crossover and mutation operations on the initial population based on the genetic algorithm is as follows: select N individuals from the initial population as the set of individuals to be operated on based on the genetic algorithm;

[0153] For the set of individuals to be operated on, crossover and mutation operations are performed based on the adaptive crossover and adaptive mutation probabilities of each individual in the set, according to the genetic algorithm.

[0154] The expression for the adaptive crossover probability is:

[0155] ;

[0156] The expression for the adaptive mutation probability is:

[0157] ;

[0158] in, Indicates the adaptive crossover probability. This represents the adaptive mutation probability. This represents the maximum fitness value of an individual in the current population. This represents the average fitness value of all individuals in the current population. This indicates that the parent individual with the higher fitness value in the crossover operation has a higher fitness value. This represents the fitness value of the individual after the mutation operation. , , and All are constants;

[0159] The individuals that have undergone crossover and mutation operations are replaced with the individuals in the initial population to obtain the first generation population.

[0160] In one embodiment, based on the three-segment encoding defined in step S2—namely, the vehicle transportation allocation segment, the shift operation sequence segment, and the task duration adaptation segment—an initial population is generated using a random initialization method. Each individual in the initial population is represented by a complete binary string formed by concatenating the first and last parts of the three encoding sequences. Each encoding segment has the same number of gene bits, and each gene bit is independently and randomly assigned a value, thus forming a batch of initial solutions representing different scheduling schemes. Based on the scheduling efficiency objective function Y and the conflict evaluation function Z, and according to the fitness calculation formula… Calculate the fitness value Q for each individual, where λ is a preset penalty coefficient. The penalty coefficient λ belongs to the engineering scenario adaptation item and needs to be determined through pre-experimentation based on the project scale, such as the number of vehicles and the complexity of the task. There is no need to fix a specific value. This formula combines the comprehensive benefits of the solution with the conflict penalty to quantify the merits of individuals.

[0161] After obtaining the initial population and its fitness, a selection operation based on roulette wheel selection is performed using a genetic algorithm. Several superior individuals are selected from the current population to form the parent set based on their fitness ratio. The core innovation lies in the introduction of an adaptive crossover and mutation probability mechanism. With the probability of mutation It is not a fixed constant, but dynamically adjusted based on the relative performance of the fitness of the individuals participating in the operation within the current population. Specifically, the crossover probability... The value depends on the higher fitness value of the two parent individuals to be crossbred. Its calculation is shown as a piecewise function; when F is greater than or equal to the population average fitness... hour, It decreases linearly to a base value as F increases. When F is less than In this case, a lower fixed probability is directly adopted. Similarly, mutation probability Then the fitness value of the individual to be mutated Related, when Greater than or equal to hour, Follow Increases and decreases linearly, when Less than In this case, a higher fixed probability is used. This adaptive mechanism ensures that the algorithm can dynamically balance exploration and mining during evolution. Based on the calculated adaptive probability, random single-point crossover operations are performed on the selected parent individuals across all chromaticities, exchanging some gene segments, and positional mutation operations are performed to randomly flip individual gene positions, generating new offspring individuals. An intergenerational replacement strategy is then used to replace individuals in the original population, thus completing one generation of evolution and obtaining the first generation population. This process will be iterative, driving the population to continuously evolve towards higher scheduling efficiency and fewer resource conflicts until the termination condition is met, ultimately outputting a near-optimal construction resource scheduling scheme. The constants P1, P2, P3, P4, and the penalty coefficient λ in the algorithm are specific to the engineering scenario and need to be determined through pre-experiments based on the project scale, such as the number of vehicles and task complexity; they do not require fixed values.

[0162] Step S4: Select the optimal fitness value from the individual fitness values ​​calculated by the conflict evaluation function and objective function in the first generation population, and preset the iteration conditions. If the iteration conditions are met, the iteration operation is executed. If the iteration conditions are not met, the individual with the optimal fitness value in the first generation population is decoded and the conflict is resolved to obtain the optimal construction resource scheduling scheme.

[0163] The selection logic for the optimal fitness value is as follows: compare the fitness values ​​of each individual in the first generation population, and select the maximum fitness value as the optimal fitness value.

[0164] Step S4 includes steps S401, S402 and S403;

[0165] Step S401: After each generation T population is generated, calculate the fitness value of individuals in the generation T population, and obtain the optimal fitness value of individuals in the generation T population according to the selection logic of the optimal fitness value.

[0166] The preset iteration condition is to compare the optimal fitness value with the optimal fitness value of an individual in the (T-1)th generation population.

[0167] If the optimal fitness value of an individual in the T-th generation population is greater than the optimal fitness value of an individual in the (T-1)-th generation population, then update the optimal fitness value of the individual in the (T-1)-th generation population to the optimal fitness value of the individual in the T-th generation population and reset the stall generation counter to zero; otherwise, increment the stall generation counter by one.

[0168] If the stall generation counter does not reach the preset threshold, the Tth generation population is used as the parent population for the next iteration, T=T+1, and the Tth generation population is used as the parent population. Then, the logic for the crossover and mutation operations in step S302 is returned.

[0169] If the stall generation counter reaches a preset threshold, then the individual with the best fitness value in the Tth generation population is decoded and conflict resolution is performed.

[0170] Step S402: Decode the individuals with the best fitness value in the Tth generation population to obtain the corresponding construction resource scheduling scheme;

[0171] The decoding process specifically includes decoding the vehicle transportation allocation coding sequence, the shift operation sequence coding sequence, and the task duration adaptation coding sequence in the three-segment coding.

[0172] Decode the vehicle transportation allocation code sequence, read the vehicle transportation allocation code sequence of the vehicle transportation allocation segment, convert the vehicle transportation allocation code sequence into vehicle number, use the vehicle number as matrix element, and construct the vehicle transportation task-vehicle number matrix, denoted as matrix L;

[0173] Each row of matrix L corresponds to a vehicle transportation task, and each column corresponds to a vehicle number;

[0174] Decode the work order coding sequence of the work group, read the work order coding sequence of the work group sequence segment, convert the work order coding sequence into work order task number, use the work order task number as matrix element, construct the work group task-work order task coding matrix, denoted as matrix S;

[0175] Each row of matrix S corresponds to a work group's work sequence task, and each column corresponds to a work group's work sequence task code;

[0176] Decode the task duration adaptation encoding sequence, read the task duration adaptation encoding sequence of the task duration adaptation segment, convert the task duration adaptation encoding sequence into a duration interval according to the one-to-one correspondence between the task duration adaptation encoding sequence and the duration interval, and use the duration interval as matrix elements to construct the task duration adaptation task-duration interval matrix, denoted as matrix X;

[0177] Each row of matrix X corresponds to a task duration matching task, and each column corresponds to a duration range.

[0178] Based on the IDs of vehicle transportation tasks, work group operation sequence tasks, and task duration adaptation tasks, matrices L, S, and X are associated to obtain the vehicle number, work group operation sequence code, and duration range corresponding to the construction resource scheduling scheme.

[0179] In one embodiment, according to step S4, the present invention performs a termination determination on the iterative optimization process of the genetic algorithm, and decodes and resolves conflicts for the optimal individual to output the final executable construction resource scheduling scheme. First, starting from the first generation of the population, after the evolution of each generation of the population, denoted as the Tth generation, the fitness value of all individuals in the Tth generation population is calculated, and the individual with the highest fitness value is selected and its fitness value is recorded as the optimal fitness value of the current generation. The preset iteration termination condition is implemented based on a stall generation counter. The optimal fitness value of the T-th generation population is compared with the optimal fitness values ​​of the previous generation and the (T-1)-th generation population. If the fitness value of the T-th generation population is better, the historical best value is updated and the stall counter is reset to zero. If it is not improved, the stall counter is incremented. If the value of the stall counter has not reached a preset threshold, such as 10 generations, the algorithm continues, using the T-th generation population as the parent, and returns to the crossover and mutation operations in step S302 to generate the next generation population. If the stall counter reaches the threshold, the algorithm is determined to have converged or entered a plateau, and the iteration stops. The algorithm then enters the decoding and conflict resolution stage for the optimal individual of the T-th generation population. The decoding process is performed on the selected optimal individual, i.e., the binary individual with the highest fitness value. This individual is formed by concatenating the three-segment encoding sequence defined in step S2. The decoding process unfolds segment by segment, strictly following the reverse operation of the encoding rules.

[0180] The vehicle transportation allocation segment is decoded, its binary sequence is read, and the encoded value of each gene bit is converted into a specific vehicle number through a preset mapping rule. The vehicle transportation task-vehicle number association matrix L is constructed with the vehicle transportation task as the row and the vehicle number as the column.

[0181] The work sequence segment of the work group is decoded, and its binary sequence is read. According to the rules defined in step S2, the global execution priority order of all tasks is determined based on the size of the binary value of the work sequence code of each gene position. The task with the smaller value is decoded first, and a sorted list of task numbers is obtained. In this way, a work group task-execution order matrix S with a clear process sequence is constructed.

[0182] Decode the task duration adaptation segment, read its binary sequence, and convert the code of each gene position into a specific planned duration interval, such as [2 hours, 4 hours], according to the preset encoding-duration interval mapping table. Then, construct the task-duration interval matrix X with the task duration adaptation task as the row and the duration interval as the column.

[0183] Since the same construction task ID resides at the same index position in the three encoded segments, corresponding rows in matrices L, S, and X can be associated through the task ID. This allows for the complete parsing of the specific vehicles allocated to each construction task, its execution order within the global process, and the planned duration interval, forming a structured draft scheduling scheme. Based on this decoded scheduling scheme, conflict resolution processing is performed. For example, a simulated annealing local search algorithm is used to fine-tune resource allocation or timing to eliminate any minor conflicts that may remain in the decoded scheme. Ultimately, a construction resource scheduling scheme that achieves an optimal balance in terms of efficiency, feasibility, and robustness is output. This process ensures that the optimization algorithm converges within an acceptable computational time and produces a high-quality scheduling plan that can be directly applied to engineering practice.

[0184] Step 403: Based on the construction resource scheduling scheme obtained from decoding, perform conflict resolution, specifically including extracting the initial decision variable vector of the construction resource scheduling scheme;

[0185] The initial decision variable vector is a set of key parameters extracted from the construction resource scheduling scheme, specifically including the vehicle number corresponding to each vehicle transportation task, the work sequence code corresponding to each shift work sequence task, and the duration interval corresponding to each task duration adaptation task.

[0186] Using the construction task ID as an index, the vehicle number, work sequence code, and duration range corresponding to the same construction task ID are used as sub-vectors;

[0187] Using the initial decision variable vector as the initial search point of the simulated annealing algorithm, local optimization is performed to obtain the local optimum solution;

[0188] The local optimal solution is re-executed in sequence through the construction logic of step S201, step S202, and step S203, and the local optimal solution is converted into a three-segment encoding format, which is denoted as the local optimal individual code.

[0189] Then replace any ordinary individual in the current T-generation population with a locally optimal individual code, except for the individual with the optimal fitness value;

[0190] When the preset iteration termination condition is met, the individual with the best fitness value in the Tth generation population is output, and the individual is decoded to obtain the optimal construction resource scheduling scheme.

[0191] The specific termination condition for the iteration is that the stall generation counter reaches a preset threshold, and the individual with the best fitness value in the Tth generation population has performed conflict resolution.

[0192] In one embodiment, according to step S403, after obtaining the draft structured scheduling scheme through decoding, the present invention performs a key conflict resolution step, which aims to refine the local optimization of the global approximate optimal solution obtained by the genetic algorithm, so as to further improve the feasibility and quality of the scheme. Specifically, firstly, an initial decision variable vector is extracted from the scheme obtained by decoding. This vector is an integrated set of key parameters, whose elements are organized according to the construction task ID. For each construction task ID, its corresponding vehicle number (from the vehicle transportation allocation segment), operation sequence code (from the team operation sequence segment), and planned duration interval (from the task duration adaptation segment) are combined to form a sub-vector. The sub-vectors of all tasks are arranged in order to form a complete initial decision variable vector. This vector transforms the core decision information of the scheduling scheme into a numerical representation suitable for processing by the local search algorithm.

[0193] Using the initial decision variable vector as the initial search point of the simulated annealing algorithm, this local optimization process directly applies to the specific construction scheme manifestation, that is, it adjusts and optimizes the actual execution parameters of vehicle ID and operation time corresponding to each construction task, with the optimization objective being to minimize the conflict evaluation function Z value defined in step S3. Heuristic search is performed in the local neighborhood of the solution space. By introducing the inherent probabilistic jump strategy of the simulated annealing algorithm, it effectively avoids getting trapped in local minima during the process of gradually reducing the temperature parameter, thereby finding a local optimum with fewer conflicts and higher quality than the initial solution, that is, the optimized decision variable vector.

[0194] After obtaining a local optimum, to reintegrate it into the evolutionary framework of the genetic algorithm, the optimization result at the phenotypic level needs to be re-encoded into a genotype format. Here, the specific construction plan (including vehicle ID and time) belongs to the phenotypic, while the three-segment code corresponds to the genotype. Therefore, following the encoding construction logic defined in steps S201, S202, and S203, the vehicle allocation, work sequence, and time interval information contained in the optimized decision variable vector are reverse-converted into standard binary encoding sequences for vehicle transportation allocation, shift work sequence, and task time adaptation, respectively. These sequences are then concatenated into a new individual, denoted as the local optimum individual code. Using an elite preservation and population update strategy, this local optimum individual code replaces the current T-generation population. Except for the historical best individual (the individual with the highest fitness value), any ordinary individual is used. This approach preserves the globally optimal solution found so far and injects high-quality new solutions obtained from local optimization into the population, enriching the population's diversity and providing better genetic material for possible subsequent evolution. When the preset iteration termination condition is met, that is, the stall generation counter has reached the preset threshold, and the historical best individual in the current generation has undergone the above conflict resolution process, the genetic algorithm finally stops. At this time, the historical best individual is output, and its final decoding yields the optimal construction resource scheduling scheme that combines high global efficiency and strong local feasibility. This conflict resolution mechanism serves as an important bridge between global optimization and final application, significantly improving the practicality and engineering reliability of the output scheme.

[0195] Example 2, refer to Figure 2 In another embodiment of the present invention, which differs from the first embodiment, a water conservancy and hydropower construction resource scheduling system is provided, including a calculation module, a construction module, a generation module and a scheduling module;

[0196] The calculation module calculates vehicle utilization and work team productivity based on construction data.

[0197] The module constructs a three-segment code based on the preset construction tasks, and forms a construction resource scheduling scheme based on the three-segment code;

[0198] Generate modules to construct conflict evaluation functions and objective functions;

[0199] An initial population is randomly generated based on the construction resource scheduling scheme. The initial population is then subjected to selection, crossover and mutation operations in sequence using a genetic algorithm to obtain the first generation population.

[0200] The scheduling module selects the optimal fitness value from the individual fitness values ​​calculated by the conflict evaluation function and objective function in the first generation population, and presets the iteration conditions. When the iteration conditions are met, the iteration operation is executed. When the iteration conditions are not met, the individual with the optimal fitness value in the first generation population is decoded and the conflict is resolved to obtain the optimal construction resource scheduling scheme.

[0201] In one embodiment, construction data is collected, specifically including vehicle work data (total transport volume, number of transport trips, vehicle rated capacity, vehicle effective operating time, planned working time) and work team data (work team output, number of personnel involved, working time, number of certified personnel, total number of personnel in the work team). The vehicle rated capacity refers to the maximum safe loading volume per trip, and the vehicle effective operating time is determined according to the definition of "pure working time of machinery" in the "Standardized Quota for Construction Machinery in Water Conservancy Projects" (2025 edition). This standard is based on the Ministry of Water Resources' "Water Conservancy Project Design Estimates (Estimated)". The supporting quotas of the "Regulations on the Compilation of Vehicle Utilization Rates" (Water Resources Department

[2024] No. 323) will be implemented from April 1, 2025, and are currently effective standards. Based on vehicle work data and team data, the vehicle utilization rate is calculated. First, the average load per vehicle trip is calculated, and its ratio to the vehicle's rated capacity is recorded as the second ratio, which reflects loading efficiency. Then, the ratio of the vehicle's effective running time to the planned working time is recorded as the third ratio, which reflects time utilization efficiency. Finally, the two second ratios and the third ratio are multiplied together to obtain the vehicle utilization rate, which comprehensively reflects the vehicle's loading and time utilization level. The calculation of team productivity involves first calculating the total working hours. The ratio of team output to total working hours is recorded as the fourth ratio. The ratio of the fourth ratio to the total number of team members is recorded as the fifth ratio. The fifth ratio is the team productivity. This scheme transforms complex construction performance into two core quantitative indicators: vehicle utilization rate and team productivity. This provides a reliable data foundation for constructing the objective function and achieving precise resource synergy optimization. In addition, as a management prerequisite for implementing this method, before using the above team data for calculation and scheduling, the team must meet the preset qualification compliance rate requirement. The qualification compliance rate requirement is that the ratio of the number of certified personnel in the team to the total number of team members is not less than the standard set according to the engineering safety specifications, such as 95%.

[0202] This invention fundamentally solves the problem of coordination and coupling representation of vehicle and work team resources in scheduling schemes by designing a three-segment unified coding structure associated with task IDs. This coding integrates and associates vehicle allocation, work sequence, and task duration information for the same construction task, ensuring the logical integrity and direct executability of the scheduling scheme. It overcomes the inherent defects of traditional methods, such as resource planning disconnect and frequent timing conflicts caused by separate coding or modeling. Based on this coding structure, an adaptive genetic algorithm is used for global optimization. Its crossover and mutation probabilities can be dynamically adjusted according to individual fitness, effectively balancing the breadth and depth of the search and avoiding the problems of fixed parameters and premature convergence of traditional algorithms. This allows for more efficient searching of high-quality scheduling schemes that comprehensively improve vehicle utilization and work team productivity. In addition, a simulated annealing algorithm is introduced to resolve conflicts of local optima that occur during the optimization process, further ensuring the high feasibility of the final output scheme. This series of innovative technologies, with the three-segment coding as the core, together achieve a reliable and intelligent solution to the multi-resource collaborative scheduling problem in water conservancy and hydropower projects, significantly improving the overall efficiency and practical value of the scheduling scheme.

[0203] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0204] 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 technical solutions of the present invention, and all such modifications or substitutions should be covered within the protection scope of the present invention.

Claims

1. A method for scheduling water conservancy and hydropower construction resources, characterized in that, The steps include: Step S1, calculating vehicle utilization and team productivity based on construction data; Step S2: Construct a three-segment code based on the preset construction tasks, and form a construction resource scheduling scheme based on the three-segment code; Step S3, construct the conflict evaluation function and objective function, including: The expression for the conflict evaluation function is: ； ； ； ； ； in, This represents the conflict assessment function value, where 'a' represents the conflict value for repeated vehicle assignment. This indicates the conflict value in the work sequence of the work group. This indicates a conflict value for task duration adaptation. , , These correspond to the conflict value weights for vehicle duplicate allocation, work group operation sequence, and task duration adaptation, respectively. Indicates the total number of vehicles. Indicates the vehicle number. Indicates the duration of vehicle overlap; This represents the m-th task number in the work group's operational sequence. This indicates the total number of tasks in the work group's work sequence. This represents the end time of the sequential task for the m-th work group. Indicates the first The start time of each work group's work sequence tasks; This indicates that the task duration is adapted to the task number. This indicates that the task duration is adapted to the total number of tasks. This represents the theoretically shortest duration for a duration-adaptive task. This indicates that the task duration is adapted to the total task capacity, and s represents the maximum capacity of each vehicle. This indicates the planned allocation duration for the i-th task to adapt to the task's duration. The objective function expression is: ; Where Y represents the scheduling efficiency value, A represents the vehicle utilization rate, and B represents the team productivity. The corresponding weight representing vehicle utilization rate. This represents the weight corresponding to the productivity of the work team, and and The sum is 1; An initial population is randomly generated based on the construction resource scheduling scheme. A genetic algorithm is then used to sequentially perform selection, crossover, and mutation operations on the initial population to obtain the first generation population, which includes: Step S301: Based on the vehicle transportation allocation coding sequence, the shift operation sequence coding sequence, and the task duration adaptation coding sequence, the initial population is generated using a random initialization method. Each individual in the initial population includes the vehicle transportation allocation coding sequence, the shift operation sequence coding sequence, and the task duration adaptation coding sequence. The number of gene positions in the vehicle transportation allocation coding sequence, the shift operation sequence coding sequence, and the task duration adaptation coding sequence are the same. During the random initialization process, each gene position is independently assigned a corresponding three-segment code. The expression for calculating the fitness value of the initial population is: ; in, Indicates the fitness value. Indicates the penalty coefficient; Step S302, the logic of performing crossover and mutation operations on the initial population based on the genetic algorithm is as follows: select N individuals from the initial population as the set of individuals to be operated on based on the genetic algorithm; For the set of individuals to be operated on, crossover and mutation operations are performed based on the adaptive crossover and adaptive mutation probabilities of each individual in the set, according to the genetic algorithm. The expression for the adaptive crossover probability is: ； The expression for the adaptive mutation probability is: ； in, Indicates the adaptive crossover probability. This represents the adaptive mutation probability. This represents the maximum fitness value of an individual in the current population. This represents the average fitness value of all individuals in the current population. This indicates that the parent individual with the higher fitness value in the crossover operation has a higher fitness value. This represents the fitness value of the individual after the mutation operation. , , and All are constants; The individuals that have undergone crossover and mutation operations are replaced with the individuals in the initial population to obtain the first generation population; Step S4: Select the optimal fitness value from the individual fitness values ​​calculated by the conflict evaluation function and objective function in the first generation population, and preset the iteration conditions. If the iteration conditions are met, the iteration operation is executed. If the iteration conditions are not met, the individual with the optimal fitness value in the first generation population is decoded and the conflict is resolved to obtain the optimal construction resource scheduling scheme.

2. The water conservancy and hydropower construction resource scheduling method as described in claim 1, characterized in that, Step S1 also includes collecting construction data, which includes vehicle working data and work team data; Vehicle operational data includes total transport volume, number of transport trips, vehicle rated capacity, effective vehicle operating time, and planned working time. Team data includes team output, number of personnel involved, working hours, number of certified workers, and total number of team members; Calculate vehicle utilization rate and team productivity by analyzing vehicle working data and team data; The calculation process of the vehicle utilization rate is as follows: the ratio of the total transportation volume to the number of transportation trips is recorded as the first ratio, and the first ratio is the average loading volume per trip. The ratio of average single-trip loading to vehicle rated capacity is recorded as the second ratio, the ratio of effective vehicle operating time to planned working time is recorded as the third ratio, and the product of the second ratio and the third ratio is recorded as the first product. The first product is the vehicle utilization rate. The specific calculation process for the team productivity is as follows: the product of the number of people involved and the working time is recorded as the second product, the second product is the total working hours, the ratio of the team output to the second product is recorded as the fourth ratio, and the ratio of the fourth ratio to the total number of people in the team is recorded as the fifth ratio, the fifth ratio is the team productivity.

3. The water conservancy and hydropower construction resource scheduling method as described in claim 2, characterized in that, The construction tasks in step S2 include vehicle transportation tasks, work group operation sequence tasks, and task duration adaptation tasks. Each construction task corresponds to a unique construction task ID. Vehicle transportation tasks, work group operation sequence tasks, and task duration adaptation tasks under the same construction task have the same ID. The three-segment coding includes the vehicle transportation allocation segment, the team operation sequence segment, and the task duration adaptation segment. The three-segment encoding is binary, in which the number of encoding gene bits is the same for the vehicle transportation allocation segment, the team operation sequence segment, and the task duration adaptation segment. The vehicle transportation allocation section, the team operation sequence section, and the task duration adaptation section correspond to the vehicle transportation task, the team operation sequence task, and the task duration adaptation task, respectively. Step S2 includes steps S201, S202 and S203; The construction logic of step S201 is as follows: for the vehicle transportation allocation segment, each gene bit corresponds to one vehicle transportation task in ascending order of vehicle transportation task number, thus obtaining the vehicle transportation allocation coding sequence. One vehicle transportation task corresponds to one vehicle number, and one vehicle number corresponds to one vehicle. The construction logic of step S202 is as follows: for the work sequence segment of the work group, the gene position corresponds to one work sequence task in the order of the work sequence task number from smallest to largest. Each gene locus is a unique work group task sequence code. Work group tasks are sorted according to the binary natural size order of the work group task sequence code. When multiple work group task sequences have the same work group task sequence code, the work group task sequence with the smaller work group task sequence number is selected for priority execution, thus obtaining the work group task sequence code sequence. The construction logic of step S203 is as follows: for the task duration adaptation segment, the gene bit corresponds to one task duration adaptation task in the order of task duration adaptation task number, and the gene bit is a unique task duration code. The preset duration range is divided into two continuous and non-overlapping duration intervals. The duration intervals are arranged in ascending order of their lower limits. The task duration codes are mapped to the arranged duration intervals in binary natural size order. Through the one-to-one correspondence between the task duration code sequence and the duration interval, the duration interval of the corresponding task duration is determined, and the task duration adaptation code sequence is obtained. Based on the vehicle transportation allocation code sequence corresponding to the vehicle transportation allocation segment, the team operation sequence code sequence corresponding to the team operation sequence segment, and the task duration adaptation code sequence corresponding to the task duration adaptation segment, they are sorted in sequence to form a construction resource scheduling scheme.

4. The water conservancy and hydropower construction resource scheduling method as described in claim 3, characterized in that, The selection logic for the optimal fitness value in step S4 is as follows: compare the fitness values ​​of each individual in the first generation population, and select the maximum value among the fitness values ​​as the optimal fitness value; Step S4 includes steps S401, S402 and S403; Step S401: After each generation T population is generated, calculate the fitness value of individuals in the generation T population, and obtain the optimal fitness value of individuals in the generation T population according to the selection logic of the optimal fitness value. The preset iteration condition is specifically to compare the optimal fitness value with the optimal fitness value of an individual in the (T-1)th generation population; If the optimal fitness value of an individual in the T-th generation population is greater than the optimal fitness value of an individual in the (T-1)-th generation population, then the optimal fitness value of the individual in the (T-1)-th generation population is updated to the optimal fitness value of the individual in the T-th generation population, and the stall generation counter is reset to zero; otherwise, the stall generation counter is incremented by one. If the stall generation counter does not reach the preset threshold, the Tth generation population is used as the parent population for the next iteration, T=T+1, and the Tth generation population is used as the parent population. Then, the logic for the crossover and mutation operations in step S302 is returned. If the stall generation counter reaches a preset threshold, then the individual with the best fitness value in the Tth generation population is decoded and conflict resolution is performed. Step S402: Decode the individuals with the best fitness value in the Tth generation population to obtain the corresponding construction resource scheduling scheme; Step S403: Based on the construction resource scheduling scheme obtained from decoding, perform conflict resolution to obtain the optimal construction resource scheduling scheme.

5. The water conservancy and hydropower construction resource scheduling method as described in claim 4, characterized in that, The decoding process in step S402 specifically includes: Decode the vehicle transportation allocation coding sequence, the team operation sequence coding sequence, and the task duration adaptation coding sequence in the three-segment coding; Decode the vehicle transportation allocation code sequence, read the vehicle transportation allocation code sequence of the vehicle transportation allocation segment, convert the vehicle transportation allocation code sequence into vehicle number, use the vehicle number as matrix element, and construct the vehicle transportation task-vehicle number matrix, denoted as matrix L; Each row of the matrix L corresponds to a vehicle transportation task, and each column corresponds to a vehicle number. Decode the work order encoding sequence of the work group, read the work order encoding sequence of the work group work order segment, convert the work order encoding sequence of the work group work order into work order task number, use the work order task number as matrix element, construct the work order task-work order task encoding matrix, denoted as matrix S; Each row of the matrix S corresponds to a work group's work sequence task, and each column corresponds to a work group's work sequence task code. Decode the task duration adaptation encoding sequence, read the task duration adaptation encoding sequence of the task duration adaptation segment, convert the task duration adaptation encoding sequence into a duration interval according to the one-to-one correspondence between the task duration adaptation encoding sequence and the duration interval, and use the duration interval as matrix elements to construct a task duration adaptation task-duration interval matrix, denoted as matrix X; Each row of matrix X corresponds to a task duration adaptation task, and each column corresponds to a duration interval. Based on the IDs of vehicle transportation tasks, work group operation sequence tasks, and task duration adaptation tasks, matrices L, S, and X are associated to obtain the vehicle number, work group operation sequence code, and duration range corresponding to the construction resource scheduling scheme.

6. The water conservancy and hydropower construction resource scheduling method as described in claim 5, characterized in that, Step S403 specifically includes: Extract the initial decision variable vector of the construction resource scheduling plan; The initial decision variable vector is a set of key parameters extracted from the construction resource scheduling scheme, specifically including the vehicle number corresponding to each vehicle transportation task, the work sequence code corresponding to each work group work sequence task, and the duration interval corresponding to each task duration adaptation task. Using the construction task ID as an index, the vehicle number, work group operation sequence code, and duration range corresponding to the same construction task ID are used as sub-vectors; Using the initial decision variable vector as the initial search point of the simulated annealing algorithm, local optimization is performed to obtain the local optimal solution; The local optimal solution is re-executed sequentially through the construction logic of step S201, step S202, and step S203 to convert the local optimal solution into a three-segment encoding format, denoted as the local optimal individual code. Then, replace any ordinary individual in the current T-generation population, except for the individual with the best fitness value, with the locally optimal individual code. When the preset iteration termination condition is met, the individual with the best fitness value in the Tth generation population is output, and the individual is decoded to obtain the optimal construction resource scheduling scheme. The specific iteration termination condition is that the stall generation counter reaches a preset threshold, and the individual with the best fitness value in the T-th generation population has performed conflict resolution.

7. A water conservancy and hydropower construction resource scheduling system, which is applied to a water conservancy and hydropower construction resource scheduling method as described in any one of claims 1-6, characterized in that, It includes a calculation module, a construction module, a generation module, and a scheduling module; The calculation module calculates vehicle utilization and work team productivity based on construction data. The module constructs a three-segment code based on the preset construction tasks, and forms a construction resource scheduling scheme based on the three-segment code; Generate modules to construct conflict evaluation functions and objective functions; An initial population is randomly generated based on the construction resource scheduling scheme. The initial population is then subjected to selection, crossover and mutation operations in sequence using a genetic algorithm to obtain the first generation population. The scheduling module selects the optimal fitness value from the individual fitness values ​​calculated by the conflict evaluation function and objective function in the first generation population, and presets the iteration conditions. When the iteration conditions are met, the iteration operation is executed. When the iteration conditions are not met, the individual with the optimal fitness value in the first generation population is decoded and the conflict is resolved to obtain the optimal construction resource scheduling scheme.

Images