Resource scheduling optimization method and system based on artificial intelligence

Abstract

The present invention discloses a resource scheduling optimization method and system based on artificial intelligence, and relates to the field of data processing. The method comprises: obtaining pending order information; establishing an order-resource demand list; clustering all pending orders according to resource demand relationships, and constructing multi-resource scheduling related orders; extracting target resource information and scheduling dispute information, and constructing an evaluation function; merging the constraint requirements of the multi-resource scheduling related orders, constructing an optimization space, iteratively optimizing the scheduling relationship of the target resource information, and obtaining resource scheduling information. The method solves the technical problems in the prior art that the data quality is not high due to the redundancy of data, and the artificial intelligence algorithm is affected by the size of the data set and the complexity of the model, and it is difficult to meet the real-time requirements. By optimizing resource scheduling, a more reasonable allocation and utilization of resources and a dynamic adjustment of system resources are achieved, thereby improving the utilization efficiency of resources and reducing the technical effect of resource waste.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and specifically to a resource scheduling optimization method and system based on artificial intelligence. Background Art

[0002] With the rapid development of information technology, the scheduling and management of various resources have become increasingly complex. Traditional resource scheduling methods often rely on manual experience and fixed rules, which cannot meet the requirements of modern complex systems for high efficiency, high accuracy and high flexibility in resource scheduling. Therefore, artificial intelligence-based resource scheduling optimization technology has emerged.

[0003] Artificial intelligence technology, especially machine learning and deep learning, can automatically discover the rules and patterns of resource scheduling by analyzing and learning a large amount of historical data, thereby realizing intelligent scheduling and optimization of resources. This method can greatly improve the efficiency and accuracy of resource scheduling, reduce the cost of manual intervention, and meet the needs of modern complex systems for resource scheduling.

[0004] To sum up, the existing technology has technical problems such as low data quality due to data redundancy, and artificial intelligence algorithms are affected by the size of the data set and the complexity of the model, making it difficult to meet real-time requirements. Summary of the invention

[0005] Based on this, it is necessary to provide an artificial intelligence-based resource scheduling optimization method and system to address the above-mentioned technical problems. This can solve the technical problems in the prior art that the data quality is low due to data redundancy, and the artificial intelligence algorithm is affected by the size of the data set and the complexity of the model, making it difficult to meet the real-time requirements. By optimizing resource scheduling, a more reasonable allocation and utilization of resources and dynamic adjustment of system resources can be achieved, thereby improving resource utilization efficiency and reducing resource waste.

[0006] In a first aspect, a resource scheduling optimization method based on artificial intelligence is provided, and the method includes: connecting to an order management platform to obtain pending order information; performing resource element analysis on the pending order information to establish an order-resource requirement list; clustering all pending orders according to resource requirement relationships based on the order-resource requirement list to construct multi-resource scheduling associated orders; extracting target resource information and scheduling dispute information based on the multi-resource scheduling associated orders; constructing an evaluation function based on the target resource information and scheduling dispute information of the multi-resource scheduling associated orders; merging the constraint requirements of the multi-resource scheduling associated orders based on the evaluation function to construct an optimization space, iteratively optimizing the scheduling relationship of the target resource information through the optimization space until the optimization target is reached, and obtaining resource scheduling information of the multi-resource scheduling associated orders.

[0007] In a second aspect, a resource scheduling optimization system based on artificial intelligence is provided, and the system includes: a to-be-done order information acquisition module, the to-be-done order information acquisition module is used to connect to an order management platform to obtain to-be-done order information; a resource element parsing module, the resource element parsing module is used to perform resource element parsing on the to-be-done order information and establish an order-resource requirement list; a multi-resource scheduling associated order construction module, the multi-resource scheduling associated order construction module is used to cluster all to-be-done orders according to resource requirement relationships based on the order-resource requirement list, and construct multi-resource scheduling associated orders; an information extraction module, the information extraction module is used to extract target resource information and scheduling dispute information based on the multi-resource scheduling associated orders; an evaluation function construction module, the evaluation function construction module is used to construct an evaluation function based on the target resource information and scheduling dispute information of the multi-resource scheduling associated orders; an optimization space construction module, the optimization space construction module is used to merge the constraint requirements of the multi-resource scheduling associated orders based on the evaluation function, construct an optimization space, and iteratively optimize the scheduling relationship of the target resource information through the optimization space until the optimization target is reached, thereby obtaining resource scheduling information of the multi-resource scheduling associated orders.

[0008] The above-mentioned artificial intelligence-based resource scheduling optimization method and system solve the technical problems in the prior art that the data quality is low due to data redundancy, and the artificial intelligence algorithm is affected by the size of the data set and the complexity of the model, making it difficult to meet the real-time requirements. By optimizing resource scheduling, a more reasonable allocation and utilization of resources and dynamic adjustment of system resources are achieved, thereby improving resource utilization efficiency and reducing resource waste.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of a process of a resource scheduling optimization method based on artificial intelligence in one embodiment;

[0011] Figure 2 A schematic diagram of a process for obtaining scheduling dispute information in a resource scheduling optimization method based on artificial intelligence in one embodiment;

[0012] Figure 3 The structure block diagram of an artificial intelligence-based resource scheduling optimization system in one embodiment.

[0013] Explanation of the reference numerals: pending order information acquisition module 11, resource element parsing module 12, multi-resource scheduling associated order construction module 13, information extraction module 14, evaluation function construction module 15, optimization space construction module 16. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0015] like Figure 1 As shown, the present application provides a resource scheduling optimization method based on artificial intelligence, the method comprising:

[0016] Connect to the order management platform to obtain pending order information.

[0017] Artificial intelligence, referred to as AI, is a branch of computer science that aims to simulate, extend and expand human intelligence, and to understand and realize human intelligence by developing and applying various theories, methods, technologies and application systems. Resource scheduling optimization refers to the process of rationally utilizing and configuring various resources under specific environments and conditions to improve the efficiency and performance of the system. It involves the allocation, scheduling and management of resources, aiming to maximize the utilization of resources and the optimal performance of the system. The present application provides a resource scheduling optimization method based on artificial intelligence, which solves the technical problems in the prior art that resource planning deviations are large, and resource availability, performance and latency may change. The efficiency and accuracy of resource scheduling can be improved, thereby better meeting the needs of various applications.

[0018] An order management platform is a software system used to manage the entire life cycle of an enterprise's sales orders. It usually includes functions such as order creation, order review, order tracking, order modification, order cancellation, and integration with other enterprise systems such as inventory systems, logistics systems, and financial systems. To-do order information refers to order data that has not been completed or needs further processing. In the context of an order management platform, to-do orders may include orders in statuses such as waiting for review, waiting for payment, waiting for shipment, and waiting for customer confirmation. Information usually includes order number, order details, customer information, product list, payment status, logistics status, etc. By connecting to the order management platform and obtaining to-do order information, enterprises can grasp the order status in real time, improve order processing efficiency, optimize inventory management, and provide customers with a better service experience.

[0019] The pending order information is parsed for resource elements, and an order-resource requirement list is established.

[0020] Clarify the resource elements involved in the order processing process, which in this application refers to the feature extraction of the agency order information. For example, human resources include workers, technicians, machine equipment resources, material resources, transportation resources, etc., and perform preset dimension analysis based on local resource conditions. For example, in locomotive manufacturing, consider workers, production equipment, materials and transportation. Associate each order with the corresponding resource requirements, and establish an order-resource requirement list, including order numbers and basic information, specific requirements for each resource element, including quantity, type, time, etc., and the priority or urgency of resource requirements. Through the above method, the specific resource requirements of each pending order can be more clearly understood, so that resource scheduling and management can be carried out more accurately, and order processing efficiency and enterprise operation efficiency can be improved.

[0021] According to the order-resource requirement list, all pending orders are clustered according to resource requirement relationships, and multi-resource scheduling associated orders are constructed.

[0022] Ensure that the order-resource requirement list has been correctly established and contains the specific requirement information of each order for various resources, such as manpower, material resources, equipment, etc. Analyze the correlation and similarity between resource requirements. For example, some orders may require the same raw materials, production equipment or delivery routes. These orders may have similarities in resource scheduling. Formulate a clustering strategy to define the clustering standards and rules based on the similarity and correlation of resource requirements. According to the characteristics of the data and the clustering strategy, select a suitable clustering algorithm, take the order-resource requirement list as input data, and apply the selected clustering algorithm for clustering analysis. During the clustering process, the orders are divided into different clusters according to the similarity and correlation of the resource requirements of the orders. According to the clustering results, the orders belonging to the same cluster are regarded as multi-resource scheduling associated orders. These orders have similarities in resource scheduling and can be uniformly managed and optimized. Develop a resource scheduling plan for each associated order group to ensure the reasonable allocation and efficient use of resources. Through the above method, cluster analysis can be performed based on the resource requirement relationship of the order to construct multi-resource scheduling associated orders. This not only helps to improve the efficiency and accuracy of resource scheduling, but also optimizes the order processing process and improves customer satisfaction.

[0023] Based on the order-resource requirement list, a timing chain is constructed with the order completion time period, and the orders are added to the timing chain according to the time correspondence to obtain the time alignment relationship of each order; based on the time alignment relationship of each order, the resource overlapping relationship of the orders of the same period is calculated according to the resource requirements of each order; clustering is performed based on the overlapping number of the resource overlapping relationship to obtain an overlapping resource order clustering set; local resource constraint information is obtained, and the local resource constraint information includes resource information, resource constraint amount, and constraint time period; the overlapping resource order clustering set is screened using the local resource constraint information to obtain the multi-resource scheduling associated order, and the multi-resource scheduling associated order is an order in which the demand for overlapping resources exceeds the local resource constraint information and there is a mutual constraint relationship between resource allocation.

[0024] According to the order completion time cycle in the order-resource requirement list, a time chain is constructed. This time chain arranges the orders in chronological order, and each order has a clear start time and end time. Through this time chain, the distribution and relationship of each order in the time dimension can be clearly seen. According to the time correspondence, the orders are added to the time chain. It involves comparing the start time and end time of the orders to ensure that their positions on the time chain are accurate. Through this process, the time alignment relationship between the orders can be obtained, that is, which orders are parallel in the same time period and which orders are serial. Based on the time alignment relationship of the orders, the resource requirements of each order are further analyzed. If two or more orders require the same or similar resources in the same time period, there is a resource overlap relationship between them. By calculating the number and type of resource overlap, the degree of this overlap relationship can be quantified. Clustering is performed based on the overlap number of resource overlap relationships as a condition. Specifically, orders with an overlap number exceeding a certain threshold can be classified into the same cluster. This threshold can be set according to the actual situation, such as the scarcity of resources, the importance of orders, and other factors. Through this process, a cluster set of overlapping resource orders can be obtained, that is, a set of orders with more overlapping resource requirements. Obtain local resource constraint information. This information includes the enterprise's currently available resource information, resource constraints, i.e., how many orders each resource can support at most in the same time period, and constraint time period, i.e., the time period during which resources are available, as the basis for formulating multi-resource scheduling strategies. Use local resource constraint information to filter the overlapping resource order clustering set to obtain multi-resource scheduling-related orders. These orders are those whose demand for overlapping resources exceeds the local resource constraint information and have a mutual constraint relationship in resource allocation. Specifically, if the total amount of resources required by the orders in an order cluster exceeds the constraint of the local resources, and these orders overlap in time, then these orders are regarded as multi-resource scheduling-related orders. Based on multi-resource scheduling-related orders and local resource constraint information, a multi-resource scheduling strategy can be formulated. This strategy needs to consider how to reasonably allocate and schedule resources to meet the needs of orders while avoiding resource waste and conflicts. This may involve a variety of measures such as adjusting the execution order of orders, optimizing the way resources are used, and seeking external resource support. Through the above method, orders and resources can be better managed, resource utilization efficiency can be improved, and operating costs can be reduced.

[0025] Based on the multi-resource scheduling associated orders, target resource information and scheduling dispute information are extracted.

[0026] Target resource information refers to the key resources involved in multi-resource scheduling and their specific requirements, which may be raw materials, production equipment, manpower, transportation vehicles, etc., depending on the specific requirements of the order. Scheduling dispute information refers to factors that may cause conflicts or disputes during the scheduling process. For multi-resource scheduling and associated orders, scheduling disputes are usually related to the allocation and priority setting of resources. Analyze resource conflicts between associated orders. This can be achieved by comparing the resource demand time windows and demand quantities of orders. If two or more orders require the same or similar resources in the same time window and the total amount of these resources exceeds the available amount, a resource conflict will occur. During the scheduling process, it may be necessary to set different priorities for orders. However, the setting of priorities often causes disputes because different orders may have different importance and urgency. Therefore, these disputes need to be recorded and analyzed to make reasonable decisions during the scheduling process. Evaluate different scheduling schemes to identify possible disputes. For example, a scheduling scheme may meet the needs of some orders, but may cause delays or increased costs for other orders. These potential disputes also need to be recorded. By integrating and presenting this information, decision makers can better understand the complexity and challenges of multi-resource scheduling and associated orders, and thus develop more effective scheduling strategies.

[0027] like Figure 2 As shown, based on the multi-resource scheduling associated orders, the constrained overlapping resources and multi-order constraint relationships are extracted, the constrained overlapping resources are resources whose demand exceeds the local resource constraint information, and the multi-order constraint relationship is the demand relationship of each order; the constrained overlapping resources are used as the target resource information, wherein the target resource information includes the constrained resources and the constraint time; the demand deviation is determined based on the multi-order constraint relationship and the local resource constraint information of the corresponding resources; the multi-order constraint relationship and the demand deviation are combined to obtain the scheduling dispute information.

[0028] Constraint overlapping resources refer to those resources whose demand exceeds the local resource constraint information in the multi-resource scheduling associated orders. They are bottlenecks in the scheduling process because they limit the execution of orders. The extraction process includes traversing each order in the multi-resource scheduling associated orders, checking whether the resources required for each order exceed the constraint amount by comparing with the local resource constraint information. If the total demand of a resource exceeds its constraint amount, the resource is marked as a constraint overlapping resource, and the specific information of each constraint overlapping resource is recorded, including resource type and constraint time, that is, the time period when the demand exceeds the constraint amount. Multi-order constraint relations refer to the mutual constraints between orders due to resource requirements. These relations determine how to reasonably arrange the execution order and priority of orders under limited resources. The analysis process includes analyzing the dependencies between them according to the time requirements and resource requirements of the orders; determining which orders will be affected by the execution of other orders, or which orders will affect other orders; and recording these relations for subsequent scheduling decisions. Demand deviation refers to the difference between the actual demand of each order and the local resource constraint information. This value reflects the degree of satisfaction of resource requirements and possible resource shortages or surpluses. For each order, the demand deviation is calculated according to the type and quantity of resources required and the local resource constraint information of the corresponding resources; if the demand of the order is less than or equal to the constraint, the demand deviation is zero; if the demand is greater than the constraint, the demand deviation is the difference between the demand and the constraint; the demand deviation of each order is recorded for subsequent scheduling optimization. The multi-order constraint relationship and the demand deviation are combined to form scheduling dispute information. This information describes how to balance the demand and priority of each order under limited resources, as well as possible resource conflicts and shortages. Scheduling dispute information can be presented in the form of tables, charts or reports so that decision makers can clearly understand the problems and challenges in the current scheduling process and formulate effective scheduling strategies accordingly. Based on the extracted overlapping resources, multi-order constraint relationships and scheduling dispute information, specific scheduling strategies are formulated, which may include adjusting the execution order of orders, optimizing the allocation and utilization of resources, and seeking external resource support. The formulation of scheduling strategies needs to take into account multiple factors, including the importance, urgency, availability and cost of orders. Through reasonable scheduling strategies, order demand can be met to the greatest extent while reducing the risk of resource conflicts and shortages. The above method provides a basis for subsequent optimization of resource scheduling.

[0029] An evaluation function is constructed according to the target resource information and scheduling dispute information of the multi-resource scheduling associated orders.

[0030] The goal of the evaluation function needs to be clarified. The evaluation function should be able to reflect the performance of the scheduling scheme in terms of resource utilization, order completion time, cost, customer satisfaction, etc. For each key indicator, it needs to be quantified into a specific value or ratio so that it can be calculated in the evaluation function. Based on the key indicators and quantification methods, an evaluation function is constructed. Scheduling dispute information usually reflects problems such as resource conflicts and order priority conflicts. When constructing the evaluation function, this information needs to be taken into account as constraints or penalty items. For example, a penalty factor can be set for each disputed order or resource and included in the evaluation function. After the evaluation function is constructed, it is necessary to verify and adjust it using actual data or simulated data. By comparing the evaluation results of different scheduling schemes, it can be checked whether the evaluation function can accurately reflect the performance of the scheduling scheme. If it is found that there is a large deviation between the evaluation results and the actual situation, it is necessary to adjust the parameters or structure of the evaluation function. The constructed evaluation function is applied to the actual multi-resource scheduling problem. By calculating the evaluation scores of different scheduling schemes, the optimal scheduling scheme can be selected or decision support can be provided.

[0031] Establish a process tracking module for the multi-resource scheduling associated orders, perform progress analysis according to the process flow of the order, and locate the progress constraint demand of the target resource information; fit the evaluation function relationship of the corresponding order target resource information according to the progress constraint demand; set the order evaluation weight according to the influence of the multi-resource scheduling associated orders; and construct the evaluation function by combining the evaluation function relationship of all order target resource information and the corresponding order evaluation weights.

[0032] Establish a process tracking module, clarify the process flow of each order, including the required resources, steps, time, etc., design a module to track the real-time progress of the order, such as using project management software, etc., to collect key data of the order in the process flow, such as start time, end time, resource usage, etc. According to the process flow of the order, analyze the progress of each step, determine the completed and unfinished parts, locate the target resources required for the current stage according to the progress analysis results, and calculate the progress constraint demand of the target resources based on the current progress of the order and the remaining process flow. For example, there are other processes that will have an impact on the allocation of target resource information. For example, the previous process has been arranged, and the subsequent process must keep up with this progress. There may be stagnation in the subsequent process, or the previous process technology has not been completed, or is being completed one after another. There are too many arrangements for this process, and it is useless if it is not completed in the previous one, so according to the previous arrangements and plans, the current resources are specifically scheduled. According to the progress constraint demand of the target resource information, the relevant evaluation indicators are determined, such as resource utilization, order delay rate, etc. A mathematical model of the evaluation function is established for the target resource information of each order. The model should be able to reflect the relationship between resource demand and evaluation indicators. The parameters of the evaluation function are fitted using historical data or simulation data to ensure the accuracy of the model. The influence of each order is analyzed according to factors such as the importance, urgency, and customer requirements of the order. An evaluation weight is set for each order, which should be able to reflect the influence and priority of the order. The evaluation weight of the order is dynamically adjusted according to the actual situation during the order execution process. The evaluation function relationship of the target resource information of all orders is combined to form a unified evaluation function framework. According to the evaluation weight of the order, each indicator in the evaluation function is weighted to ensure fairness between different orders. The evaluation function is verified using actual data or simulation data, and optimized and adjusted according to the verification results. Through the above method, a process tracking module based on multi-resource scheduling associated orders is established, and the evaluation function is constructed according to the process flow and progress of the order, providing strong support for scheduling decisions.

[0033] The expression of the evaluation function is: ,in, The progress constraint demand for the first multi-resource scheduling associated orders, Get the value of the first target resource information, The progress constraint demand for the n-th resource scheduling associated order, The value of the nth target resource information, n is the total number of multi-resource scheduling related orders, n is a positive integer, , is the local resource constraint maximum value of the target resource information, , , are the order evaluation weights of the first, second, and up to the nth resource scheduling associated orders, and, .

[0034] Construct an evaluation function expression that takes into account the progress constraint requirements of each multi-resource scheduling associated order, the value of the target resource information, and the evaluation weight of the order. The progress constraint demand for the first multi-resource scheduling associated orders, Get the value of the first target resource information, The progress constraint demand for the n-th resource scheduling associated order, The value of the nth target resource information, n is the total number of multi-resource scheduling related orders, n is a positive integer, , is the local resource constraint maximum value of the target resource information, , , are the order evaluation weights of the first, second, and up to the nth resource scheduling associated orders, and, This expression assumes that resource requirements are measured in the same units and can be directly compared and calculated. In actual applications, this expression needs to be adjusted or optimized according to specific circumstances to adapt to different business scenarios and requirements.

[0035] Based on the information of all pending orders, a process tracking module is constructed, and the process tracking module tracks and analyzes the order completion status by collecting the life cycle monitoring data of the order; the life cycle of the order is obtained, and the life cycle is located according to the target resource information to determine the positioning period of the target resource; the process flow resource quantity is analyzed using the life cycle monitoring data and the positioning period to determine the supply resource quantity; the process progress rate is determined according to the historical process progress data of the positioning period, and the constraint rate is obtained by integrating the demand rate of the order; the progress constraint demand quantity of the target resource information is obtained according to the constraint rate and the supply resource quantity.

[0036] Based on the information of all pending orders, a process tracking module is constructed to determine the method of collecting order life cycle monitoring data, including data collection points, data types, such as start time, end time, resource usage, etc., and collection frequency. A database is designed and established to store order information, life cycle monitoring data, and other related data. A process tracking system that can automatically collect data, store data, and conduct preliminary analysis is developed, namely, a process tracking module. The system automatically or manually collects data from each stage of order creation to completion, and analyzes the current completion status of the order based on the collected data, such as completed, in progress, pending, etc. The life cycle information of the order is obtained from the database, and the target resources that need to be tracked and analyzed are determined based on the demand and resource allocation of the order. The positioning cycle of the target resource is determined based on its usage in the order life cycle. Combined with the life cycle monitoring data of the order, the usage of the target resource at different stages is analyzed. According to the process flow and resource usage of the order, the amount of supply resources required in a specific positioning cycle is determined. The historical process progress data of the target resource in similar orders is obtained from the database, and the average or expected progress rate of the target resource in a specific positioning cycle is calculated based on the historical data. Consider the demand rate of the current order for the target resource and combine it with the process progress rate to determine the constraint rate of the order. According to the determined constraint rate and the amount of supply resources, calculate the progress constraint demand of the target resource within a specific positioning cycle. According to the actual demand situation, make necessary adjustments and optimizations to the calculated progress constraint demand. Through the above method, an effective process tracking module is constructed to conduct a comprehensive analysis of the order completion status, target resource positioning cycle, supply resource quantity, and progress constraint demand, providing strong support for resource scheduling and order management.

[0037] Based on the evaluation function, the constraint requirements of the multiple resource scheduling associated orders are combined to construct an optimization space, and the scheduling relationship of the target resource information is iteratively optimized through the optimization space until the optimization target is reached, thereby obtaining resource scheduling information of the multiple resource scheduling associated orders.

[0038] Clarify the optimization goals, such as minimizing order completion time, maximizing resource utilization, minimizing costs, etc. According to the evaluation function and the constraint requirements of the order, determine the constraints in the resource scheduling process, such as resource quantity restrictions, order priority, etc. Based on the constraints, construct a feasible scheduling solution set, namely the optimization space. Randomly select one or more initial scheduling solutions from the optimization space as the starting point, use the evaluation function to evaluate the initial solution, and obtain its score. According to the optimization strategy, generate new candidate scheduling solutions in the optimization space, use the evaluation function to evaluate the candidate solution, obtain its score, compare the score of the candidate solution with the current optimal solution, and select the solution with a higher score as the new current optimal solution. Check whether the iteration termination conditions are met, such as reaching the maximum number of iterations, the current optimal solution has not been significantly improved in multiple consecutive iterations, etc. If the conditions are met, stop the iteration; otherwise, continue the iterative optimization. After the iterative optimization is completed, output the current optimal solution as the resource scheduling information of the multi-resource scheduling associated order, parse the final solution, and obtain detailed information such as resource allocation, start time, and end time of each order. According to the scheduling information obtained by the analysis, formulate a specific resource scheduling plan and implement it. By constructing the evaluation function through the above method, we can comprehensively consider multiple dimensions such as the order life cycle, resource constraints, and progress requirements to ensure the comprehensive performance of the scheduling solution. The iterative optimization process can adaptively adjust the optimization strategy and parameters according to the actual situation to adapt to different orders and resource environments. Through iteration, we can gradually approach the optimal solution and improve the efficiency and quality of the scheduling solution.

[0039] Set optimization iteration rules, and construct an optimization space based on the evaluation function and the optimization iteration rules; obtain the constraint requirements of the multi-resource scheduling associated orders, including time constraint requirements, material constraint requirements, and process flow constraint requirements; perform correlation analysis based on the constraint requirements and the target resource information to determine the associated constraint requirements; and add the associated constraint requirements as constraint conditions to the optimization space.

[0040] Set the conditions for the termination of iteration, such as reaching the preset maximum number of iterations, the quality improvement of the solution after multiple consecutive iterations is lower than a certain threshold, etc., select the iteration strategy suitable for the characteristics of the problem, such as genetic algorithm, particle swarm optimization, simulated annealing, etc., and set relevant parameters, such as population size, iteration step size, etc. According to the characteristics of resource scheduling, define decision variables, such as the allocation of resources, the start and end time of the task, etc., set reasonable boundary values ​​and step sizes for the decision variables to ensure the effectiveness and rationality of the search space. In the defined search space, randomly generate a certain number of initial solutions as the starting point of the iteration. Obtain the time constraints such as the time window and delivery date of the order to ensure that the scheduling plan is completed within the specified time, understand the type, quantity, arrival time, etc. of the materials required for the order, ensure the continuity and sufficiency of the material supply during the scheduling process, analyze the process flow of the order, understand the dependencies and execution order between different tasks, and ensure that the scheduling plan meets the process requirements. Analyze the obtained constraint requirements, and combine the attributes, status and other information of the target resources to find out the correlation between them. Through correlation analysis, determine the constraint requirements that are closely related to the target resource information, which will directly affect the results of resource scheduling. According to the associated constraint requirements, the definition and boundaries of the decision variables are adjusted to ensure that the search space can cover all possible scheduling solutions that meet the constraints. In the updated search space, the initial solution set is regenerated or adjusted to ensure that they can meet the new constraints. According to the associated constraint requirements, the evaluation function is updated to ensure that it can accurately evaluate the performance of the scheduling solution under the constraints. Through the above method, an optimization space based on the evaluation function, optimization iteration rules and associated constraint requirements is constructed, and the optimal scheduling solution that meets the associated order constraint requirements of multi-resource scheduling is found through iterative optimization.

[0041] like Figure 3 As shown, the embodiment of the present application includes a resource scheduling optimization system based on artificial intelligence, and the system includes:

[0042] A pending order information acquisition module 11, wherein the pending order information acquisition module 11 is used to connect to the order management platform to acquire pending order information;

[0043] A resource element parsing module 12, the resource element parsing module 12 is used to parse the pending order information for resource elements and establish an order-resource requirement list;

[0044] A multi-resource scheduling associated order construction module 13, the multi-resource scheduling associated order construction module 13 is used to cluster all pending orders according to resource demand relationships according to the order-resource demand list, and construct multi-resource scheduling associated orders;

[0045] An information extraction module 14, the information extraction module 14 is used to extract target resource information and scheduling dispute information based on the multi-resource scheduling associated order;

[0046] An evaluation function construction module 15, wherein the evaluation function construction module 15 is used to construct an evaluation function according to target resource information and scheduling dispute information of the multi-resource scheduling associated order;

[0047] The optimization space construction module 16 is used to merge the constraint requirements of the multi-resource scheduling associated orders based on the evaluation function, construct an optimization space, iteratively optimize the scheduling relationship of the target resource information through the optimization space until the optimization target is reached, and obtain the resource scheduling information of the multi-resource scheduling associated orders.

[0048] Furthermore, the embodiment of the present application also includes:

[0049] A time alignment relationship acquisition module, which is used to build a time sequence chain based on the order-resource requirement list with the order completion time period, and add the order to the time sequence chain according to the time correspondence relationship to obtain the time alignment relationship of each order;

[0050] A resource overlap relationship calculation module, the resource overlap relationship calculation module is used to calculate the resource overlap relationship of the orders of the same period according to the resource requirements of each order based on the time alignment relationship of each order;

[0051] An overlapping resource order clustering set obtaining module, the overlapping resource order clustering set obtaining module is used to perform clustering based on the overlapping quantity of the resource overlapping relationship as a condition to obtain an overlapping resource order clustering set;

[0052] A local resource constraint information acquisition module, the local resource constraint information acquisition module is used to obtain local resource constraint information, the local resource constraint information includes resource information, resource constraint amount, and constraint time period;

[0053] An overlapping resource order cluster set screening module is used to use the local resource constraint information to screen the overlapping resource order cluster set to obtain the multi-resource scheduling related orders. The multi-resource scheduling related orders are orders in which the demand for overlapping resources exceeds the local resource constraint information and there is a mutual constraint relationship between resource allocation.

[0054] Furthermore, the embodiment of the present application also includes:

[0055] A multi-order constraint relationship module, the multi-order constraint relationship module is used to extract the constraint overlapping resources and multi-order constraint relationships based on the multi-resource scheduling associated orders, the constraint overlapping resources are resources whose demand exceeds the local resource constraint information, and the multi-order constraint relationship is the demand relationship of each order;

[0056] A target resource information acquisition module, the target resource information acquisition module is used to use the constraint overlapping resources as the target resource information, wherein the target resource information includes constraint resources and constraint time;

[0057] A demand deviation amount determination module, the demand deviation amount determination module is used to determine the demand deviation amount according to the multi-order constraint relationship and the local resource constraint information of the corresponding resources;

[0058] A scheduling dispute information acquisition module is used to combine the multi-order constraint relationship and the demand deviation amount to obtain the scheduling dispute information.

[0059] Furthermore, the embodiment of the present application also includes:

[0060] A progress analysis module, which is used to establish a process tracking module for the multi-resource scheduling associated order, perform progress analysis according to the process flow of the order, and locate the progress constraint demand of the target resource information;

[0061] An evaluation function relationship fitting module, wherein the evaluation function relationship fitting module is used to fit the evaluation function relationship of the corresponding order target resource information according to the progress constraint demand;

[0062] An order evaluation weight setting module, the order evaluation weight setting module is used to set the order evaluation weight according to the influence of the multi-resource scheduling related orders;

[0063] The function relationship combination module is used to combine the evaluation function relationship of all order target resource information and the corresponding order evaluation weights to construct the evaluation function.

[0064] Furthermore, the embodiment of the present application also includes:

[0065] An evaluation function expression expansion module, wherein the evaluation function expression expansion module is used for the expression of the evaluation function: ,in, The progress constraint demand for the first multi-resource scheduling associated orders, Get the value of the first target resource information, The progress constraint demand for the n-th resource scheduling associated order, The value of the nth target resource information, n is the total number of multi-resource scheduling related orders, n is a positive integer, , is the local resource constraint maximum value of the target resource information, , , are the order evaluation weights of the first, second, and up to the nth resource scheduling associated orders, and, .

[0066] Furthermore, the embodiment of the present application also includes:

[0067] A process tracking module, which is used to build a process tracking module based on all pending order information, and the process tracking module tracks and analyzes the order completion status by collecting order life cycle monitoring data;

[0068] A life cycle positioning module, the life cycle positioning module is used to obtain the life cycle of the order, perform life cycle positioning according to the target resource information, and determine the positioning period of the target resource;

[0069] A supply resource quantity determination module, the supply resource quantity determination module is used to use the life cycle monitoring data and the positioning cycle to perform process resource quantity analysis and determine the supply resource quantity;

[0070] A constraint rate acquisition module, the constraint rate acquisition module is used to determine the process progress rate according to the historical process progress data of the positioning cycle, and to obtain the constraint rate by integrating the demand rate of the order;

[0071] A constraint demand quantity acquisition module is used to obtain the progress constraint demand quantity of the target resource information according to the constraint rate and the supply resource quantity.

[0072] Furthermore, the embodiment of the present application also includes:

[0073] An optimization iterative rule setting module, wherein the optimization iterative rule setting module is used to set an optimization iterative rule and construct an optimization space based on the evaluation function and the optimization iterative rule;

[0074] A constraint demand acquisition module, which is used to acquire the constraint demands of the multi-resource scheduling associated orders, including time constraint demands, material constraint demands, and process flow constraint demands;

[0075] An associated constraint requirement determination module, the associated constraint requirement determination module is used to perform a correlation analysis based on the constraint requirement and the target resource information to determine the associated constraint requirement;

[0076] A constraint condition adding module is used to add the associated constraint requirements as constraint conditions to the optimization space.

[0077] For the specific implementation of the resource scheduling optimization system based on artificial intelligence, please refer to the implementation of the resource scheduling optimization method based on artificial intelligence above, which will not be repeated here. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0078] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0079] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A resource scheduling optimization method based on artificial intelligence, characterized in that: The method comprises: Connect to the order management platform to obtain pending order information; Analyze the resource elements of the pending order information and establish an order-resource requirement list; According to the order-resource requirement list, all pending orders are clustered according to resource requirement relationships to construct multi-resource scheduling associated orders; Based on the multi-resource scheduling associated orders, extracting target resource information and scheduling dispute information; Constructing an evaluation function according to target resource information and scheduling dispute information of the multi-resource scheduling associated orders; Based on the evaluation function, the constraint requirements of the multiple resource scheduling associated orders are combined to construct an optimization space, and the scheduling relationship of the target resource information is iteratively optimized through the optimization space until the optimization target is reached, thereby obtaining resource scheduling information of the multiple resource scheduling associated orders; According to the order-resource requirement list, all pending orders are clustered according to resource requirement relationships to construct multi-resource scheduling associated orders, including: Based on the order-resource requirement list, a time sequence chain is constructed with the order completion time period, and the orders are added to the time sequence chain according to the time correspondence relationship to obtain the time alignment relationship of each order; Based on the time alignment relationship of the orders, the resource overlap relationship of the orders in the same period is calculated according to the resource requirements of the orders; Clustering is performed based on the overlapping quantity of the resource overlapping relationship as a condition to obtain an overlapping resource order clustering set; Obtaining local resource constraint information, wherein the local resource constraint information includes resource information, resource constraint amount, and constraint time period; The overlapping resource order cluster set is screened by using the local resource constraint information to obtain the multi-resource scheduling associated order, where the multi-resource scheduling associated order is an order in which the demand for overlapping resources exceeds the local resource constraint information and there is a mutual constraint relationship between resource allocation; Based on the multi-resource scheduling associated order, target resource information and scheduling dispute information are extracted, including: Based on the multi-resource scheduling associated orders, extracting the constrained overlapping resources and the multi-order constraint relationship, wherein the constrained overlapping resources are resources whose demand exceeds the local resource constraint information, and the multi-order constraint relationship is the demand relationship of each order; Using the constraint overlapping resources as the target resource information, wherein the target resource information includes constraint resources and constraint time; Determining a demand deviation amount according to the multi-order constraint relationship and local resource constraint information of corresponding resources; Combining the multi-order constraint relationship and the demand deviation amount to obtain the scheduling dispute information; According to the target resource information and scheduling dispute information of the multi-resource scheduling associated order, an evaluation function is constructed, including: Establishing a process tracking module for the multi-resource scheduling associated orders, performing progress analysis according to the process flow of the orders, and locating the progress constraint demand of the target resource information; According to the progress constraint demand, fitting the evaluation function relationship of the corresponding order target resource information; According to the influence of the multi-resource scheduling related orders, an order evaluation weight is set; The evaluation function is constructed by combining the evaluation function relationship of all order target resource information and the corresponding order evaluation weights.

2. The method according to claim 1, characterized in that The evaluation function is constructed by combining the evaluation function relationship of all order target resource information and the corresponding order evaluation weights, including: The expression of the evaluation function is: ,in, The progress constraint demand for the first multi-resource scheduling associated orders, Get the value of the first target resource information, The progress constraint demand for the n-th resource scheduling associated order, The value of the nth target resource information, n is the total number of multi-resource scheduling related orders, n is a positive integer, , is the local resource constraint maximum value of the target resource information, , , are the order evaluation weights of the first, second, and up to the nth resource scheduling associated orders, and, .

3. The method according to claim 1, characterized in that Establishing a process tracking module for the multi-resource scheduling associated order, performing progress analysis according to the process flow of the order, and locating the progress constraint demand of the target resource information, including: Based on all pending order information, a process tracking module is constructed, and the process tracking module tracks and analyzes the order completion status by collecting order life cycle monitoring data; Acquire the life cycle of the order, perform life cycle positioning according to the target resource information, and determine the positioning period of the target resource; Utilizing the life cycle monitoring data and positioning cycle to analyze the process resource quantity and determine the supply resource quantity; Determine the process progress rate based on the historical process progress data of the positioning cycle, integrate the demand rate of the order, and obtain the constraint rate; The progress-constrained demand quantity of the target resource information is obtained according to the constraint rate and the supplied resource quantity.

4. The method according to claim 1, characterized in that Based on the evaluation function, the constraint requirements of the multi-resource scheduling associated orders are combined to construct an optimization space, including: Setting optimization iteration rules, and constructing an optimization space based on the evaluation function and the optimization iteration rules; Obtaining the constraint requirements of the multi-resource scheduling associated orders, including time constraint requirements, material constraint requirements, and process flow constraint requirements; Performing a correlation analysis based on the constraint requirements and the target resource information to determine the associated constraint requirements; The associated constraint requirements are added as constraint conditions to the optimization space.

5. The resource scheduling optimization system based on artificial intelligence is characterized by: The system comprises: A pending order information acquisition module, which is used to connect to the order management platform to obtain pending order information; A resource element parsing module, the resource element parsing module is used to parse the pending order information for resource elements and establish an order-resource requirement list; A multi-resource scheduling associated order construction module, the multi-resource scheduling associated order construction module is used to cluster all pending orders according to resource demand relationships based on the order-resource demand list to construct multi-resource scheduling associated orders; An information extraction module, the information extraction module is used to extract target resource information and scheduling dispute information based on the multi-resource scheduling associated order; An evaluation function construction module, wherein the evaluation function construction module is used to construct an evaluation function according to target resource information and scheduling dispute information of the multi-resource scheduling associated order; An optimization space construction module, the optimization space construction module is used to combine the constraint requirements of the multiple resource scheduling associated orders based on the evaluation function, construct an optimization space, iteratively optimize the scheduling relationship of the target resource information through the optimization space until the optimization target is reached, and obtain resource scheduling information of the multiple resource scheduling associated orders; The system further comprises: A time alignment relationship acquisition module, which is used to build a time sequence chain based on the order-resource requirement list with the order completion time period, and add the order to the time sequence chain according to the time correspondence relationship to obtain the time alignment relationship of each order; A resource overlap relationship calculation module, the resource overlap relationship calculation module is used to calculate the resource overlap relationship of the orders of the same period according to the resource requirements of each order based on the time alignment relationship of each order; An overlapping resource order clustering set obtaining module, the overlapping resource order clustering set obtaining module is used to perform clustering based on the overlapping quantity of the resource overlapping relationship as a condition to obtain an overlapping resource order clustering set; A local resource constraint information acquisition module, the local resource constraint information acquisition module is used to obtain local resource constraint information, the local resource constraint information includes resource information, resource constraint amount, and constraint time period; An overlapping resource order cluster set screening module, the overlapping resource order cluster set screening module is used to screen the overlapping resource order cluster set by using the local resource constraint information to obtain the multi-resource scheduling associated order, the multi-resource scheduling associated order is an order in which the demand for overlapping resources exceeds the local resource constraint information and there is a mutual constraint relationship between resource allocation; A multi-order constraint relationship module, the multi-order constraint relationship module is used to extract the constraint overlapping resources and multi-order constraint relationships based on the multi-resource scheduling associated orders, the constraint overlapping resources are resources whose demand exceeds the local resource constraint information, and the multi-order constraint relationship is the demand relationship of each order; A target resource information acquisition module, the target resource information acquisition module is used to use the constraint overlapping resources as the target resource information, wherein the target resource information includes constraint resources and constraint time; A demand deviation amount determination module, the demand deviation amount determination module is used to determine the demand deviation amount according to the multi-order constraint relationship and the local resource constraint information of the corresponding resources; A scheduling dispute information acquisition module, the scheduling dispute information acquisition module is used to combine the multi-order constraint relationship and the demand deviation amount to obtain the scheduling dispute information; A progress analysis module, which is used to establish a process tracking module for the multi-resource scheduling associated order, perform progress analysis according to the process flow of the order, and locate the progress constraint demand of the target resource information; An evaluation function relationship fitting module, wherein the evaluation function relationship fitting module is used to fit the evaluation function relationship of the corresponding order target resource information according to the progress constraint demand; An order evaluation weight setting module, the order evaluation weight setting module is used to set the order evaluation weight according to the influence of the multi-resource scheduling related orders; The function relationship combination module is used to combine the evaluation function relationship of all order target resource information and the corresponding order evaluation weights to construct the evaluation function.

Images