Resource scheduling method and device, computer device, storage medium and program product

Abstract

The application relates to a resource scheduling method and device, computer equipment, a storage medium and a program product. The method comprises the following steps: obtaining a resource scheduling request initiated by a user, a user satisfaction index and related data of each supply provider; inputting the resource scheduling request and the related data into a resource demand prediction model to obtain a target resource procurement quantity and a target resource procurement time; then inputting the target resource procurement quantity, the target resource procurement time, the related data and the user satisfaction index into a strategy initialization module included in a resource scheduling strategy construction model to obtain an initial resource scheduling strategy; finally inputting the initial resource scheduling strategy into a strategy optimization module included in the resource scheduling strategy construction model to obtain a target resource scheduling strategy, and performing resource scheduling according to the target resource scheduling strategy. Through the method, reasonable allocation of resources is realized, scheduling accuracy of resource scheduling is improved, and invalid scheduling of resources is avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of resource scheduling, and in particular to a resource scheduling method and device, computer equipment, computer readable storage medium and computer program product. BACKGROUND

[0002] In terms of resource scheduling, the current main basis for arranging resource distribution and allocation is production plan and inventory level, and warehouse managers extract resources from inventory and ship them according to the demand notification of the production department.

[0003] However, the current resource scheduling method has the problem of low scheduling accuracy. SUMMARY

[0004] Therefore, it is necessary to provide a resource scheduling method, device, computer equipment, computer readable storage medium and computer program product capable of improving scheduling accuracy.

[0005] In a first aspect, the present application provides a resource scheduling method, comprising:

[0006] obtaining a resource scheduling request initiated by a user, a user satisfaction index, supply capacity data of each supply provider, production plan data, real-time inventory data and transportation cost data;

[0007] inputting the resource scheduling request, the supply capacity data and the production plan data into a pre-constructed resource demand prediction model to obtain a target resource procurement quantity and a target resource procurement time;

[0008] inputting the target resource procurement quantity, the target resource procurement time, the production plan data, the real-time inventory data, the transportation cost data and the user satisfaction index into a strategy initialization module included in a pre-constructed resource scheduling strategy construction model to obtain an initial resource scheduling strategy through the strategy initialization module; the initial resource scheduling strategy includes a plurality of resource scheduling groups, and each resource scheduling group is used to represent the allocation quantity of any resource at a current production site;

[0009] inputting the initial resource scheduling strategy into a strategy optimization module included in the resource scheduling strategy construction model to obtain a target resource scheduling strategy through the strategy optimization module, and scheduling resources according to the target resource scheduling strategy.

[0010] In one embodiment, the strategy optimization module includes a group selection submodule and a group change submodule.

[0011] inputting the initial resource scheduling strategy into a strategy optimization module included in the resource scheduling strategy construction model to obtain a target resource scheduling strategy through the strategy optimization module, including:

[0012] The initial resource scheduling strategy is input into a group selection submodule, and a target resource scheduling group is obtained through the group selection submodule;

[0013] The target resource scheduling group is input into a group change submodule, and a changed target resource scheduling group is obtained through a crossover operation and a mutation operation on the target resource scheduling group;

[0014] A current resource scheduling strategy is constructed according to the changed target resource scheduling group and other resource scheduling groups in the initial resource scheduling strategy except the target resource scheduling group;

[0015] In a case where a termination judgment result of the current resource scheduling strategy meets a preset condition, the current resource scheduling strategy is taken as a target resource scheduling strategy.

[0016] In one of the embodiments, taking the current resource scheduling strategy as the target resource scheduling strategy comprises:

[0017] Based on the current resource scheduling strategy and a preset fitness function, a fitness corresponding to the current resource scheduling strategy is obtained;

[0018] In a case where a group change number of the group change submodule reaches a preset iteration number or the fitness reaches a preset fitness threshold, the current resource scheduling strategy is taken as the target resource scheduling strategy.

[0019] In one embodiment, the resource demand prediction model is trained through the following steps:

[0020] Historical resource demand data, historical resource scheduling requests, supply capacity data of each supply provider, and production plan data are obtained;

[0021] According to the historical resource demand data, the historical resource scheduling requests, the supply capacity data, and the production plan data, a feature data training set and a feature data test set are respectively constructed;

[0022] The model parameters of the resource demand prediction model to be trained are trained by using the feature data training set, and a target resource demand prediction model is obtained;

[0023] The feature data test set is input into the target resource demand prediction model, and a performance test is performed on the target resource demand prediction model. If a test result of the performance test indicates that the target resource demand prediction model passes the test, the target resource demand prediction model is taken as a trained resource demand prediction model.

[0024] In one exemplary embodiment, the model parameters of the resource demand prediction model to be trained are trained by using the feature data training set, and the target resource demand prediction model is obtained, comprising:

[0025] The feature data training set is input into the resource demand prediction model to be trained to obtain first predicted resource demand data corresponding to the historical resource scheduling request;

[0026] According to the loss function preset in the resource demand prediction model to be trained, the historical resource demand data contained in the feature data training set, and the first predicted resource demand data, a corresponding model loss value is obtained;

[0027] Based on the model loss value, the model parameters in the resource demand prediction model to be trained are adjusted by using the back propagation algorithm and the stochastic gradient descent method, and until the loss function converges, a target resource demand prediction model is obtained.

[0028] In one embodiment, the resource scheduling request, the supply capability data and the production plan data are input into the pre-constructed resource demand prediction model to obtain the target resource procurement quantity and the target resource procurement time, including:

[0029] The resource scheduling request, the supply capability data and the production plan data are input into the resource demand prediction model, and second predicted resource demand data is obtained through the resource demand prediction model;

[0030] The procurement cost data corresponding to the second predicted resource demand data, the evaluation information of each supply provider, the inventory cost data and the shortage cost data are obtained;

[0031] According to the second predicted resource demand data, the procurement cost data, the evaluation information, the inventory cost data and the shortage cost data, the target resource procurement quantity and the target resource procurement time are obtained.

[0032] In a second aspect, the application also provides a resource scheduling device, comprising:

[0033] The data acquisition module is configured to acquire a resource scheduling request initiated by a user, a user satisfaction index, supply capability data of each supply provider, production plan data, real-time inventory data and transportation cost data;

[0034] The data preparation module is configured to input the resource scheduling request, the supply capability data and the production plan data into a pre-constructed resource demand prediction model to obtain a target resource procurement quantity and a target resource procurement time;

[0035] The data input module is configured to input the target resource procurement quantity, the target resource procurement time, the production plan data, the real-time inventory data, the transportation cost data and the user satisfaction index into a strategy initialization module contained in a pre-constructed resource scheduling strategy construction model, and obtain an initial resource scheduling strategy through the strategy initialization module; the initial resource scheduling strategy comprises a plurality of resource scheduling groups, and each resource scheduling group is used to represent the allocation quantity of any resource at a current production site.

[0036] The resource scheduling module is configured to input the initial resource scheduling strategy into a policy optimization module included in the resource scheduling strategy construction model, obtain a target resource scheduling strategy through the policy optimization module, and perform resource scheduling according to the target resource scheduling strategy.

[0037] In a third aspect, the present application further provides a computer device, comprising a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program:

[0038] obtaining a resource scheduling request initiated by a user, a user satisfaction index, supply capacity data of each supply provider, production plan data, real-time inventory data, and transportation cost data;

[0039] inputting the resource scheduling request, the supply capacity data, and the production plan data into a pre-constructed resource demand prediction model to obtain a target resource procurement quantity and a target resource procurement time;

[0040] inputting the target resource procurement quantity, the target resource procurement time, the production plan data, the real-time inventory data, the transportation cost data, and the user satisfaction index into a policy initialization module included in a pre-constructed resource scheduling strategy construction model to obtain an initial resource scheduling strategy through the policy initialization module; the initial resource scheduling strategy comprises a plurality of resource scheduling groups, and each resource scheduling group is used to represent an allocation quantity of any resource at a current production site;

[0041] inputting the initial resource scheduling strategy into a policy optimization module included in the resource scheduling strategy construction model, obtaining a target resource scheduling strategy through the policy optimization module, and performing resource scheduling according to the target resource scheduling strategy.

[0042] In a fourth aspect, the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0043] obtaining a resource scheduling request initiated by a user, a user satisfaction index, supply capacity data of each supply provider, production plan data, real-time inventory data, and transportation cost data;

[0044] inputting the resource scheduling request, the supply capacity data, and the production plan data into a pre-constructed resource demand prediction model to obtain a target resource procurement quantity and a target resource procurement time;

[0045] The target resource procurement quantity, the target resource procurement time, production plan data, real-time inventory data, transportation cost data, and user satisfaction indicators are input into a strategy initialization module included in a pre-constructed resource scheduling strategy construction model, and an initial resource scheduling strategy is obtained through the strategy initialization module; the initial resource scheduling strategy includes a plurality of resource scheduling groups, and the resource scheduling groups are used to represent the allocation quantity of any resource at the current production site;

[0046] The initial resource scheduling strategy is input into a strategy optimization module included in the resource scheduling strategy construction model, a target resource scheduling strategy is obtained through the strategy optimization module, and resource scheduling is performed according to the target resource scheduling strategy.

[0047] In a fifth aspect, the present application also provides a computer program product, comprising a computer program which, when executed by a processor, implements the following steps:

[0048] A resource scheduling request initiated by a user, user satisfaction indicators, supply capacity data of each supply provider, production plan data, real-time inventory data, and transportation cost data are obtained;

[0049] The resource scheduling request, the supply capacity data, and the production plan data are input into a pre-constructed resource demand prediction model to obtain a target resource procurement quantity and a target resource procurement time;

[0050] The target resource procurement quantity, the target resource procurement time, production plan data, real-time inventory data, transportation cost data, and user satisfaction indicators are input into a strategy initialization module included in a pre-constructed resource scheduling strategy construction model, and an initial resource scheduling strategy is obtained through the strategy initialization module; the initial resource scheduling strategy includes a plurality of resource scheduling groups, and the resource scheduling groups are used to represent the allocation quantity of any resource at the current production site;

[0051] The initial resource scheduling strategy is input into a strategy optimization module included in the resource scheduling strategy construction model, a target resource scheduling strategy is obtained through the strategy optimization module, and resource scheduling is performed according to the target resource scheduling strategy.

[0052] The resource scheduling method, device, computer device, computer readable storage medium and computer program product obtain a resource scheduling request initiated by a user, a user satisfaction index, supply capacity data of each supply provider, production plan data, real-time inventory data and transportation cost data, input the resource scheduling request, the supply capacity data and the production plan data into a pre-constructed resource demand prediction model, obtain a target resource procurement quantity and a target resource procurement time, then input the target resource procurement quantity, the target resource procurement time, the production plan data, the real-time inventory data, the transportation cost data and the user satisfaction index into a strategy initialization module included in a pre-constructed resource scheduling strategy construction model, obtain an initial resource scheduling strategy including a plurality of resource scheduling groups through the strategy initialization module, wherein the resource scheduling group is used to represent the allocation quantity of any kind of resource at a current production site, finally input the initial resource scheduling strategy into a strategy optimization module included in the resource scheduling strategy construction model, obtain a target resource scheduling strategy through the strategy optimization module, and perform resource scheduling according to the target resource scheduling strategy. First, the resource demand prediction model is used to determine the current resource procurement quantity and resource procurement time of the user, the strategy is initialized according to the determined resource procurement quantity and time, and the related data of each supply provider and the user satisfaction index, and the strategy is optimized in real time, the rationality and effectiveness of the resource allocation of the finally determined resource scheduling strategy are improved, and the scheduling accuracy of the resource scheduling is improved, and invalid resource scheduling is avoided. BRIEF DESCRIPTION OF DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other related drawings can be obtained by those skilled in the art without creative labor.

[0054] Figure 1 An application environment diagram of a resource scheduling method in an embodiment;

[0055] Figure 2 A flowchart of a resource scheduling method in an embodiment;

[0056] Figure 3 A schematic diagram of a material procurement strategy optimization model in an embodiment;

[0057] Figure 4 A schematic diagram of a material scheduling strategy optimization model in another embodiment;

[0058] Figure 5 A structural block diagram of a resource scheduling device in an embodiment;

[0059] Figure 6 Figure 1 is a schematic diagram of the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0060] For the purpose of making the purpose, technical scheme and advantages of the present application clearer, the present application is further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0061] The resource scheduling method provided by the embodiments of the present application can be applied in an application environment as shown in Figure 1 The terminal 102 communicates with the server 104 through a network. The data storage system can store data required to be processed by the server 104. The data storage system can be integrated on the server 104, or placed on a cloud or other network server. The server 104 obtains a resource scheduling request initiated by a user from the terminal 102, obtains a user satisfaction index, supply capacity data of each supply provider, production plan data, real-time inventory data and transportation cost data from the data storage system, inputs the resource scheduling request, the supply capacity data and the production plan data into a pre-constructed resource demand prediction model to obtain a target resource procurement quantity and a target resource procurement time, and then inputs the target resource procurement quantity, the target procurement time, the production plan data, the real-time inventory data, the transportation cost data and the user satisfaction index into a strategy initialization module included in a resource scheduling strategy construction model constructed by Xie Na, obtains an initial resource scheduling strategy through the strategy initialization module, wherein the initial resource scheduling strategy includes a plurality of resource scheduling groups, and each resource scheduling group is used to represent the allocation quantity of any resource at a current production site, and finally inputs the initial resource scheduling strategy into a strategy optimization module included in the resource scheduling strategy construction model to obtain a target resource scheduling strategy through the strategy optimization module, and performs resource scheduling according to the target resource scheduling strategy. The terminal 102 can be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, Internet of Things devices and portable wearable devices. The Internet of Things device can be a smart speaker, a smart television, a smart air conditioner, a smart vehicle-mounted device, a projection device, etc. The portable wearable device can be a smart watch, a smart bracelet, a head-mounted device, etc. The head-mounted device can be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The server 104 can be a stand-alone physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0062] In one exemplary embodiment, as Figure 2As shown, a resource scheduling method is provided, and the method is applied to Figure 1 The server 104 in the method is taken as an example for illustration, and the method comprises the following steps S201 to S204.

[0063] In step S201, a user-initiated resource scheduling request, user satisfaction indicators, supply capacity data of each supply provider, production plan data, real-time inventory data and transportation cost data are obtained.

[0064] The supply capacity data can be understood as the resources or services that can be provided within a certain time, the production plan data can be understood as the resource data of the subsequent planned production of the supply provider, the real-time inventory data can be understood as the real-time stored resource data of the supply provider, and the transportation cost data can be understood as the cost data consumed by the transportation of resources from A to B.

[0065] For example, the server 104 obtains the user-initiated resource scheduling request through the terminal 102, and can obtain the pre-set user satisfaction indicators, supply capacity data of each supply provider, production plan data, real-time inventory data and transportation cost data from the data storage system. Obtaining multi-source data lays a solid data foundation for subsequent data preparation and improves the accuracy of the finally determined target resource scheduling strategy.

[0066] In step S202, the resource scheduling request, the supply capacity data and the production plan data are input into a pre-constructed resource demand prediction model to obtain a target resource procurement quantity and a target resource procurement time.

[0067] Optionally, the server 104 inputs the resource scheduling request, the supply capacity data and the production plan data into the pre-constructed resource demand prediction model, obtains the corresponding predicted resource demand data through the resource demand prediction model, calculates the procurement cost data corresponding to the predicted resource demand data, the evaluation information of each supply provider, the inventory cost data and the shortage cost data, and finally obtains the target resource procurement quantity and the target resource procurement time according to the predicted resource demand data, the procurement cost data, the evaluation information, the inventory cost data and the shortage cost data. The predicted resource demand data, the evaluation information of each supply provider and various cost data are used as the basis for determining the resource procurement information, which improves the determination accuracy of the procurement quantity and the procurement time, and further improves the determination accuracy of the resource scheduling strategy.

[0068] Step S203, input the target resource procurement quantity, target resource procurement time, production plan data, real-time inventory data, transportation cost data, and user satisfaction index into the strategy initialization module included in the pre-constructed resource scheduling strategy construction model, and obtain an initial resource scheduling strategy through the strategy initialization module. The initial resource scheduling strategy includes a plurality of resource scheduling groups, and each resource scheduling group is used to represent the allocation quantity of any resource at the current production site.

[0069] The resource scheduling group can be understood as the allocation quantity of any resource at the current production site.

[0070] For example, the server 104 inputs the target resource procurement quantity, target resource procurement time, production plan data, real-time inventory data, transportation cost data, and user satisfaction index into the strategy initialization module included in the pre-constructed resource scheduling strategy construction model, encodes and initializes the population: encodes the material scheduling scheme, and adopts binary encoding or integer encoding. Assuming that n kinds of materials need to be scheduled to m production sites, the allocation quantity of each material at each production site can be regarded as a gene (i.e., a resource scheduling group), all genes form a chromosome, and represent a resource scheduling scheme. An initial population is randomly generated, and the population size is determined according to the complexity of the problem and the computing resources, and is generally set to several tens to several hundred individuals. Through the encoding method, a resource scheduling scheme can be quickly formed, which lays a data foundation for subsequent construction of the target resource scheduling strategy.

[0071] Step S204, input the initial resource scheduling strategy into the strategy optimization module included in the resource scheduling strategy construction model, obtain a target resource scheduling strategy through the strategy optimization module, and perform resource scheduling according to the target resource scheduling strategy.

[0072] Optionally, the server 104 inputs the initial resource scheduling strategy into the strategy optimization module included in the resource scheduling strategy construction model, performs strategy optimization on the initial resource scheduling strategy through the strategy optimization module, performs cross and mutation operations on the resource scheduling groups included in the initial resource scheduling strategy, until a set iteration end condition is met, takes the optimized resource scheduling strategy as the target resource scheduling strategy, and finally performs resource scheduling according to the target resource scheduling strategy. Through the cross and mutation operations on the initial resource scheduling strategy, a better resource allocation scheme can be explored, the resource utilization efficiency is improved, and after multiple rounds of iteration optimization, the generated target resource scheduling strategy can better meet the production demand, reduce the transportation cost, and improve the user satisfaction.

[0073] In the resource scheduling method, the resource scheduling request initiated by the user, the user satisfaction index, the supply capacity data of each supply provider, the production plan data, the real-time inventory data and the transportation cost data are obtained, the resource scheduling request, the supply capacity data and the production plan data are input into a pre-constructed resource demand prediction model to obtain a target resource procurement quantity and a target resource procurement time, and then the target resource procurement quantity, the target resource procurement time, the production plan data, the real-time inventory data, the transportation cost data and the user satisfaction index are input into a strategy initialization module included in a pre-constructed resource scheduling strategy construction model to obtain an initial resource scheduling strategy including a plurality of resource scheduling groups through the strategy initialization module, wherein the resource scheduling group is used to represent the allocation quantity of any kind of resource at the current production site, and finally the initial resource scheduling strategy is input into a strategy optimization module included in the resource scheduling strategy construction model to obtain a target resource scheduling strategy through the strategy optimization module, and resource scheduling is performed according to the target resource scheduling strategy. First, the resource demand prediction model is used to determine the current resource procurement quantity and resource procurement time of the user, and then the determined resource procurement quantity and time, the related data of each supply provider and the user satisfaction index are used for strategy initialization, and the strategy is optimized in real time, so as to improve the rationality and effectiveness of the resource allocation of the finally determined resource scheduling strategy, thereby improving the scheduling accuracy of the resource scheduling and avoiding the invalid scheduling of the resource.

[0074] In one embodiment, the strategy optimization module includes a group selection submodule and a group change submodule.

[0075] The initial resource scheduling strategy is input into the strategy optimization module included in the resource scheduling strategy construction model, and the target resource scheduling strategy is obtained through the strategy optimization module, including: the initial resource scheduling strategy is input into the group selection submodule, and the target resource scheduling group is obtained through the group selection submodule; the target resource scheduling group is input into the group change submodule, and the target resource scheduling group is subjected to cross operation and mutation operation to obtain a changed target resource scheduling group; according to the changed target resource scheduling group and other resource scheduling groups in the initial resource scheduling strategy except the target resource scheduling group, a current resource scheduling strategy is constructed; and in a case where a termination judgment result of the current resource scheduling strategy meets a preset condition, the current resource scheduling strategy is taken as the target resource scheduling strategy.

[0076] Optionally, the server 104 inputs the initial resource scheduling strategy into a group selection submodule included in the strategy optimization module, selects the resource scheduling groups included in the initial resource scheduling strategy through the group selection submodule by using methods such as roulette selection and tournament selection, determines the target resource scheduling group to be changed, then inputs the target resource scheduling group into a group change submodule included in the strategy optimization module, performs cross operation and mutation operation on the target resource scheduling group through the group change submodule, obtains the changed target resource scheduling group, and finally constructs the current resource scheduling strategy according to the changed target resource scheduling group and other resource scheduling groups except the target resource scheduling group in the initial resource scheduling strategy. In a case where a termination judgment result of the current resource scheduling strategy meets a preset condition, the current resource scheduling strategy is taken as the target resource scheduling strategy.

[0077] The above method can achieve the following technical effects:

[0078] 1. By using methods such as roulette selection and tournament selection to select the resource scheduling groups, a balance between diversity and adaptability can be ensured, so that the target resource scheduling group to be changed can be effectively locked.

[0079] 2. Through cross operation, the excellent characteristics of different resource scheduling groups can be integrated, so as to generate a new scheduling combination, which can bring a more optimal resource allocation mode. The mutation operation introduces randomness, which can increase the diversity of solutions, avoid the algorithm from falling into local optimum, and help to find potential better solutions.

[0080] 3. After optimization in the above manner, the target resource scheduling strategy obtained can reduce unnecessary resource waste, improve production efficiency, and thus reduce overall operating costs.

[0081] In one of the embodiments, taking the current resource scheduling strategy as the target resource scheduling strategy includes: based on the current resource scheduling strategy and a pre-set fitness function, obtaining the fitness corresponding to the current resource scheduling strategy; in a case where the number of group changes of the group change submodule reaches a pre-set iteration number, or the fitness reaches a pre-set fitness threshold, taking the current resource scheduling strategy as the target resource scheduling strategy.

[0082] The fitness function can be understood as being designed to minimize transportation costs, maximize customer satisfaction, and ensure timely resource supply. The transportation cost is taken as the main target, and factors such as the timeliness of resource supply and the satisfaction cost of production demand are also considered. The higher the value of the fitness function, the better the scheduling scheme.

[0083] Exemplarily, the server 104 obtains the fitness corresponding to the current resource scheduling strategy based on the current resource scheduling strategy and the pre-set fitness function, and if the number of group changes of the group change sub-module reaches the pre-set iteration number or the fitness reaches the pre-set fitness threshold, the continuous optimization of the current resource scheduling strategy is terminated, and the current resource scheduling strategy is taken as the target resource scheduling strategy. Based on the fitness function, the advantages and disadvantages of the current resource scheduling strategy can be quantified, and a clear basis is provided for the subsequent optimization decision, so that the evaluation of the strategy is more scientific and objective; by setting the iteration number and the fitness threshold, it is ensured that the optimization process is a dynamic and continuous improvement process. When the condition is met, the optimization can be terminated in time to avoid unnecessary waste of resources.

[0084] In an exemplary embodiment, the resource demand prediction model is trained by the following steps:

[0085] The historical resource demand data, the historical resource scheduling request, the supply capacity data of each supply provider and the production plan data are obtained; the feature data training set and the feature data test set are constructed according to the historical resource demand data, the historical resource scheduling request, the supply capacity data and the production plan data; the model parameters of the resource demand prediction model to be trained are trained by using the feature data training set, and the target resource demand prediction model is obtained; the feature data test set is input into the target resource demand prediction model, and the performance of the target resource demand prediction model is tested, and if the test result of the performance test represents that the target resource demand prediction model passes the test, the target resource demand prediction model is taken as the trained resource demand prediction model.

[0086] Optionally, multi-source data is collected through Internet of Things technology and enterprise internal information systems, including historical resource demand data, historical resource scheduling requests, supply capacity data of supply providers, production plan data, etc. The collected data is cleaned, sorted and analyzed, and key features are extracted. A demand prediction model is constructed using a Long Short-Term Memory (LSTM) algorithm. The historical material usage data is normalized to map the data to the [0, 1] interval to facilitate model training. An LSTM network is constructed including an input layer, a hidden layer and an output layer. The number of nodes in the input layer is determined according to the number of input features, and the historical length of time series data can be used as an input feature, i.e. the number of input layer nodes is equal to the historical length. The number of nodes in the hidden layer can be determined according to experience or experiment, and is generally set to A, B, etc. The number of nodes in the output layer is 1, indicating the predicted material demand. The preprocessed data is divided into a training set and a test set. The training set is used to train the LSTM model, and the test set is used to evaluate the trained model. Mean Squared Error (MSE), Mean Absolute Error (MAE) and other indicators are calculated to measure the prediction accuracy of the model. If the model performance does not meet the requirements, adjust the model parameters or increase the amount of training data and retrain.

[0087] In the above manner, the following technical effects are achieved:

[0088] 1. The LSTM algorithm is good at processing time series data and can capture long-term dependencies in the data, thereby improving the accuracy of material demand prediction.

[0089] 2. Cleaning and sorting the collected data can remove noise and irrelevant information, extract key features, and ensure the quality of data for model training.

[0090] 3. Using Mean Squared Error (MSE) and Mean Absolute Error (MAE) and other indicators to evaluate model performance can provide a clear direction for model improvement. When the model performance does not meet the requirements, by adjusting the parameters or increasing the amount of training data, iterative optimization can be performed to gradually improve the prediction ability of the model.

[0091] In one embodiment, the model parameters of the to-be-trained resource demand prediction model are trained by using the feature data training set to obtain a target resource demand prediction model, including: inputting the feature data training set into the to-be-trained resource demand prediction model to obtain first predicted resource demand data corresponding to the historical resource scheduling requests; obtaining corresponding model loss values according to a loss function pre-set in the to-be-trained resource demand prediction model, historical resource demand data contained in the feature data training set, and the first predicted resource demand data; based on the model loss values, corresponding adjustment is made to the model parameters in the to-be-trained resource demand prediction model by using a backpropagation algorithm and a stochastic gradient descent method, and until the loss function converges, the target resource demand prediction model is obtained.

[0092] The backpropagation algorithm can be understood as an important algorithm in the neural network training process, which is mainly used to calculate the gradient of each parameter in the neural network, so as to update these parameters by gradient descent and other optimization methods; the stochastic gradient descent method (SGD) can be understood as an algorithm for optimizing machine learning models, which is used to iteratively update the parameters of the model.

[0093] Exemplarily, the server 104 inputs the feature data training set into the to-be-trained resource demand prediction model to obtain first predicted resource demand data corresponding to the historical resource scheduling requests, and obtains corresponding model loss values according to a loss function pre-set in the to-be-trained resource demand prediction model, historical resource demand data contained in the feature data training set, and the first predicted resource demand data. In the case that the model loss values do not satisfy the loss value threshold of the model training result, the model parameters in the to-be-trained resource demand prediction model are adjusted by using the backpropagation algorithm and the stochastic gradient descent method, and until the loss function converges, the model parameters are no longer updated, and the current resource demand prediction model is taken as the target resource demand prediction model.

[0094] Through the above steps, the following technical effects are achieved:

[0095] 1. By continuously adjusting the model parameters, the feedback mechanism based on historical data improves the accuracy of resource demand prediction and reduces the prediction error.

[0096] 2. The application of backpropagation and stochastic gradient descent method makes the model training process highly automated, can quickly adapt to changes in data, and improves the training efficiency.

[0097] 3. The threshold of loss function convergence is set to effectively control the quality of model training, ensure that the model stops training after reaching the predetermined performance standard, and avoid overfitting.

[0098] In one of the embodiments, the resource scheduling request, the supply capacity data and the production plan data are input into a pre-constructed resource demand prediction model to obtain the target resource procurement quantity and the target resource procurement time, including: inputting the resource scheduling request, the supply capacity data and the production plan data into the resource demand prediction model, and obtaining second predicted resource demand data through the resource demand prediction model; obtaining procurement cost data corresponding to the second predicted resource demand data, evaluation information of each supply provider, inventory cost data and shortage cost data; and obtaining the target resource procurement quantity and the target resource procurement time according to the second predicted resource demand data, the procurement cost data, the evaluation information, the inventory cost data and the shortage cost data.

[0099] Optionally, a material procurement plan is formulated in combination with the demand prediction data and the supply provider evaluation information. The optimal procurement quantity and procurement time are determined by considering factors such as procurement cost, inventory cost and shortage cost.

[0100] Cost analysis: analyze procurement cost, including the unit price of materials, procurement quantity, transportation cost, procurement procedure cost, etc. Negotiate with the supply provider to obtain more favorable prices and procurement terms. Consider storage cost, capital occupation cost, material loss, etc. inventory cost. According to the inventory cost model, determine a reasonable inventory level. Evaluate the shortage cost, the loss caused by production interruption and delayed delivery due to material shortage, etc.

[0101] Determine the procurement quantity: based on the demand prediction result, determine the economic order quantity (EOQ) in combination with the safety inventory level. The EOQ formula is , where D is the annual demand, S is the cost of each order, and H is the unit inventory holding cost.

[0102] Consider market supply situation and price fluctuation factors to adjust the procurement quantity appropriately. Increase the procurement quantity when the price is low to reduce the procurement cost. Through comprehensive cost analysis, the enterprise can more effectively control the procurement cost and inventory cost, avoid unnecessary expenditure, thereby improving the overall economic benefit, and using the EOQ model to determine a reasonable procurement quantity helps to reduce the inventory holding cost, while ensuring the stability of the material supply and reducing the risk of shortage.

[0103] In one exemplary embodiment, a specific implementation of a resource scheduling method is provided, specifically as follows:

[0104] Collect multi-source data through Internet of Things technology and enterprise internal information system, including historical material (i.e. the aforementioned resource) usage data, market demand data, supplier (i.e. the aforementioned supply provider) supply capacity data, production plan data, etc. Clean, organize and analyze the collected data to extract key features.

[0105] 1. Demand forecasting model (i.e., the aforementioned resource demand forecasting model):

[0106] A long short-term memory network (LSTM) algorithm is used to construct the demand forecasting model. The historical material usage data is normalized to map the data to the [0, 1] interval to facilitate model training. An LSTM network is constructed with an input layer, a hidden layer, and an output layer. The number of nodes in the input layer is determined based on the number of input features, and the historical length of the time series data can be used as an input feature, i.e., the number of input layer nodes equals the historical length. The number of nodes in the hidden layer can be determined based on experience or experiment, and is generally set to 64, 128, etc. The number of nodes in the output layer is 1, representing the predicted material demand. The preprocessed data is divided into a training set and a test set. The training set is used to train the LSTM model, and the backpropagation algorithm and stochastic gradient descent (SGD) are used to optimize the model parameters. During training, appropriate learning rate, iteration times, and other parameters are set to ensure model convergence. The learning rate can be set to 0.001, and the iteration times are 1000. The test set is used to evaluate the trained model, and the mean squared error (MSE), mean absolute error (MAE), and other indicators are calculated to measure the prediction accuracy of the model. If the model performance does not meet the requirements, adjust the model parameters or increase the amount of training data and retrain.

[0107] 2. Material procurement optimization strategy:

[0108] As shown in Figure 3 , combined with the demand forecasting results and supplier evaluation, a material procurement plan is developed. Considering procurement cost, inventory cost, and shortage cost, etc., the optimal procurement quantity and procurement time are determined.

[0109] Cost analysis: Analyze procurement costs, including material unit price, procurement quantity, transportation costs, and procurement fees. Negotiate with suppliers for more favorable prices and procurement terms. Consider storage costs, capital occupation costs, and material loss, etc. inventory costs. According to the inventory cost model, determine a reasonable inventory level. Evaluate the cost of shortages, such as production interruptions and delayed deliveries caused by material shortages.

[0110] Determine the procurement quantity: Based on the demand forecasting results, combined with the safety stock level, determine the economic order quantity (EOQ). The EOQ formula is , where D is the annual demand, S is the cost of each order, and H is the unit inventory holding cost.

[0111] Consider market supply conditions and price fluctuations, and adjust the procurement quantity appropriately. Increase procurement quantity when prices are low to reduce procurement costs.

[0112] 3. Material scheduling optimization strategy:

[0113] As shown in Figure 4 , according to the purchase time, purchase quantity, production plan, real-time inventory level, intelligent algorithm is used for material scheduling optimization. Dynamically adjust the distribution route and allocation scheme of materials to ensure that materials can be timely and accurately supplied to the production site.

[0114] Encoding and initialization of population: encode the material scheduling scheme, using binary encoding or integer encoding method. Assuming that n kinds of materials need to be dispatched to m production locations, the allocation quantity of each material at each production location can be regarded as a gene, and all genes form a chromosome, representing a material scheduling scheme. Randomly generate an initial population, and the population size is determined according to the complexity of the problem and the computing resources, generally set to several tens to several hundred individuals.

[0115] Fitness function design: design the fitness function with the goal of minimizing transportation cost, maximizing customer satisfaction, and ensuring timely supply of materials. Transportation cost is the main target, while considering the timeliness of material supply and the satisfaction of production demand. The higher the value of the fitness function, the better the scheduling scheme.

[0116] Selection operation: use appropriate selection methods such as roulette wheel selection, tournament selection, etc. to select excellent individuals from the current population into the next generation population. The probability of selection is proportional to the fitness value of the individual, and the higher the fitness value of the individual, the greater the probability of being selected.

[0117] Crossover operation: crossover operation is performed on the selected individuals to generate new individuals. Common crossover methods include single-point crossover, multi-point crossover, etc. For example, in single-point crossover, a crossover point is randomly selected, and the parts after the crossover point of two individuals are exchanged to generate two new individuals.

[0118] Mutation operation: mutation operation is performed on the individuals after crossover to increase the diversity of the population. Mutation operation can randomly change the value of some genes in the individual. With a certain probability, the value of a gene is changed from 0 to 1 or from 1 to 0.

[0119] Termination condition judgment: set termination conditions such as reaching the maximum number of iterations, the average fitness value of the population reaching a certain threshold, etc. When the termination condition is met, the algorithm stops running and outputs the optimal material scheduling scheme.

[0120] Result output and application: select the individual with the highest fitness value from the final population as the optimal solution, i.e. the optimal material scheduling scheme. Apply the optimal scheduling scheme to actual material scheduling to guide the distribution and allocation of materials.

[0121] Compared with the prior art, the application has the following technical advantages:

[0122] 1. By combining resource prediction data, supply provider evaluation information, procurement cost, inventory cost and stockout cost, a more realistic procurement plan is formulated, the plan including the time and quantity of specific procurement and the resource procurement from which supply provider, helping the user to quickly procure materials.

[0123] 2. Under the constraints of procurement time, procurement quantity, production plan and real-time inventory level, genetic algorithm is used for initialization of resource scheduling strategy and further optimization of the strategy, improving the accuracy of resource allocation and resource scheduling strategy, and avoiding invalid calling of resources, thereby saving resource scheduling cost.

[0124] It should be understood that, although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0125] Based on the same inventive concept, the embodiments of the application also provide a resource scheduling device for implementing the above-mentioned resource scheduling method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more resource scheduling device embodiments provided below can refer to the limitations of the resource scheduling method described above, which will not be repeated here.

[0126] In one exemplary embodiment, as shown in Figure 5 a resource scheduling device is provided, comprising a data acquisition module 501, a data preparation module 502, a data input module 503 and a resource scheduling module 504, wherein:

[0127] The data acquisition module 501 is configured to acquire a resource scheduling request initiated by a user, a user satisfaction index, supply capacity data of each supply provider, production plan data, real-time inventory data and transportation cost data.

[0128] The data preparation module 502 is configured to input the resource scheduling request, the supply capacity data and the production plan data into a pre-constructed resource demand prediction model to obtain a target resource procurement quantity and a target resource procurement time.

[0129] The data input module 503 is configured to input the target resource procurement quantity, the target resource procurement time, the production plan data, real-time inventory data, transportation cost data and a user satisfaction index into a strategy initialization module included in a pre-constructed resource scheduling strategy construction model to obtain an initial resource scheduling strategy through the strategy initialization module. The initial resource scheduling strategy includes a plurality of resource scheduling groups, and each resource scheduling group is used to represent an allocation quantity of any resource at a current production site.

[0130] The resource scheduling module 504 is configured to input the initial resource scheduling strategy into a strategy optimization module included in the resource scheduling strategy construction model to obtain a target resource scheduling strategy through the strategy optimization module, and perform resource scheduling according to the target resource scheduling strategy.

[0131] In one embodiment, the strategy optimization module includes a group selection sub-module and a group change sub-module. The resource scheduling module 504 further includes:

[0132] The target resource scheduling group determination sub-module is configured to input the initial resource scheduling strategy into the group selection sub-module to obtain a target resource scheduling group through the group selection sub-module.

[0133] The change sub-module is configured to input the target resource scheduling group into the group change sub-module, and perform a crossover operation and a mutation operation on the target resource scheduling group to obtain a changed target resource scheduling group.

[0134] The current resource scheduling strategy construction sub-module is configured to construct a current resource scheduling strategy according to the changed target resource scheduling group and other resource scheduling groups in the initial resource scheduling strategy except the target resource scheduling group.

[0135] The target resource scheduling strategy determination sub-module is configured to, in a case where a termination judgment result of the current resource scheduling strategy meets a preset condition, take the current resource scheduling strategy as the target resource scheduling strategy.

[0136] In one embodiment, the target resource scheduling strategy determination sub-module is further configured to obtain an adaptability corresponding to the current resource scheduling strategy based on the current resource scheduling strategy and a pre-set adaptability function. In a case where a group change number of the group change sub-module reaches a pre-set iteration number or the adaptability reaches a pre-set adaptability threshold, the current resource scheduling strategy is taken as the target resource scheduling strategy.

[0137] In an example embodiment, the resource scheduling apparatus further comprises a model training module configured to obtain historical resource demand data, historical resource scheduling requests, supply capacity data of each supply provider, and production plan data; construct a feature data training set and a feature data test set according to the historical resource demand data, the historical resource scheduling requests, the supply capacity data, and the production plan data; train model parameters of a to-be-trained resource demand prediction model using the feature data training set to obtain a target resource demand prediction model; input the feature data test set into the target resource demand prediction model to perform performance testing on the target resource demand prediction model, and if a test result of the performance testing indicates that the target resource demand prediction model passes the testing, take the target resource demand prediction model as a trained resource demand prediction model.

[0138] In an example embodiment, the model training module is further configured to input the feature data training set into the to-be-trained resource demand prediction model to obtain first predicted resource demand data corresponding to the historical resource scheduling requests; obtain a corresponding model loss value according to a pre-set loss function in the to-be-trained resource demand prediction model, historical resource demand data contained in the feature data training set, and the first predicted resource demand data; adjust the model parameters in the to-be-trained resource demand prediction model based on the model loss value using a back propagation algorithm and a stochastic gradient descent method, and obtain the target resource demand prediction model when the loss function converges.

[0139] In an example embodiment, the data preparation module is further configured to input the resource scheduling request, the supply capacity data, and the production plan data into the resource demand prediction model to obtain second predicted resource demand data through the resource demand prediction model; obtain procurement cost data corresponding to the second predicted resource demand data, evaluation information of each supply provider, inventory cost data, and shortage cost data; and obtain a target resource procurement quantity and a target resource procurement time according to the second predicted resource demand data, the procurement cost data, the evaluation information, the inventory cost data, and the shortage cost data.

[0140] The above modules of the resource scheduling apparatus can be realized in whole or in part by software, hardware, and a combination thereof. The above modules can be embedded in or independent of a processor in a computer device in a hardware form, or stored in a memory in a computer device in a software form, so as to be called and executed by a processor to perform operations corresponding to the above modules.

[0141] In an example embodiment, a computer device is provided, which can be a server, and an internal structure diagram of the computer device can be as shown in FIG. 1. Figure 6As shown in the figure. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store resource scheduling requests, user satisfaction indicators, supply capacity data, production plan data, real-time inventory data, transportation cost data, target resource procurement quantity and target resource procurement time. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor to implement a resource scheduling method.

[0142] Those skilled in the art can understand that, Figure 6 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0143] In one exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the resource scheduling method in the above embodiments.

[0144] In one embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by the processor to implement the resource scheduling method in the above embodiments.

[0145] In one embodiment, a computer program product is provided, including a computer program, and the computer program is executed by the processor to implement the resource scheduling method in the above embodiments.

[0146] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0147] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0148] The technical features of the above embodiments can be combined in any manner. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present application.

[0149] The above-described embodiments are merely illustrative of several embodiments of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A resource scheduling method, characterized in that, The method comprises: obtaining a user-initiated resource scheduling request, historical resource scheduling requests, historical resource demand data, user satisfaction indicators, supply capacity data of each supply provider, production plan data, real-time inventory data, and transportation cost data; According to the historical resource demand data, historical resource scheduling requests, supply capacity data, and production plan data, a feature data training set and a feature data test set are respectively constructed; using the feature data training set, the model parameters of the to-be-trained resource demand prediction model are trained to obtain a target resource demand prediction model; the feature data test set is input into the target resource demand prediction model to test the performance of the target resource demand prediction model, and if the test result of the performance test indicates that the target resource demand prediction model passes the test, the target resource demand prediction model is used as the trained resource demand prediction model; The resource scheduling request, supply capacity data, and production plan data are input into the resource demand prediction model to obtain second predicted resource demand data through the resource demand prediction model; the procurement cost data corresponding to the second predicted resource demand data, the evaluation information of each supply provider, the inventory cost data, and the stockout cost data are obtained; according to the second predicted resource demand data, the procurement cost data, the evaluation information, the inventory cost data, and the stockout cost data, the target resource procurement quantity and the target resource procurement time are obtained; The target resource procurement quantity, the target resource procurement time, the production plan data, the real-time inventory data, the transportation cost data, and the user satisfaction indicators are input into a strategy initialization module included in a pre-constructed resource scheduling strategy construction model to obtain an initial resource scheduling strategy through the strategy initialization module; the initial resource scheduling strategy comprises a plurality of resource scheduling groups, and each resource scheduling group is used to represent the allocation quantity of any resource at a current production site; The initial resource scheduling strategy is input into a strategy optimization module included in the resource scheduling strategy construction model to obtain a target resource scheduling strategy through the strategy optimization module, and resource scheduling is performed according to the target resource scheduling strategy.

2. The method of claim 1, wherein, The strategy optimization module comprises a group selection submodule and a group change submodule; The initial resource scheduling strategy is input into the strategy optimization module included in the resource scheduling strategy construction model to obtain the target resource scheduling strategy through the strategy optimization module, which comprises: The initial resource scheduling strategy is input into the group selection submodule to obtain a target resource scheduling group through the group selection submodule; The target resource scheduling group is input into the group change submodule to perform cross-operation and mutation operation on the target resource scheduling group to obtain a changed target resource scheduling group; According to the changed target resource scheduling group and other resource scheduling groups in the initial resource scheduling strategy except the target resource scheduling group, a current resource scheduling strategy is constructed; In a case where a termination judgment result of the current resource scheduling strategy meets a preset condition, the current resource scheduling strategy is used as the target resource scheduling strategy.

3. The method of claim 2, wherein, The current resource scheduling strategy is taken as the target resource scheduling strategy, including: Based on the current resource scheduling strategy and the pre-set fitness function, the fitness corresponding to the current resource scheduling strategy is obtained; In the case that the group change number of the group change submodule reaches the pre-set iteration number, or the fitness reaches the pre-set fitness threshold, the current resource scheduling strategy is taken as the target resource scheduling strategy.

4. The method of claim 1, wherein, The model parameters of the to-be-trained resource demand prediction model are trained by using the feature data training set, and a target resource demand prediction model is obtained, including: The feature data training set is input into the to-be-trained resource demand prediction model, and first predicted resource demand data corresponding to the historical resource scheduling request is obtained; According to the pre-set loss function in the to-be-trained resource demand prediction model, the historical resource demand data contained in the feature data training set, and the first predicted resource demand data, a corresponding model loss value is obtained; Based on the model loss value, the model parameters in the to-be-trained resource demand prediction model are adjusted by using the back propagation algorithm and the stochastic gradient descent method, until the loss function converges, and the target resource demand prediction model is obtained.

5. A resource scheduling apparatus, characterized by comprising: The device comprises: The data acquisition module is configured to acquire a resource scheduling request initiated by a user, historical resource scheduling requests, historical resource demand data, user satisfaction indicators, supply capacity data of each supply provider, production plan data, real-time inventory data, and transportation cost data; The data preparation module is configured to construct a feature data training set and a feature data test set based on the historical resource demand data, historical resource scheduling requests, supply capacity data, and production plan data; train model parameters of a to-be-trained resource demand prediction model by using the feature data training set to obtain a target resource demand prediction model; input the feature data test set into the target resource demand prediction model to test the performance of the target resource demand prediction model; if the test result of the performance test indicates that the target resource demand prediction model passes the test, the target resource demand prediction model is taken as a trained resource demand prediction model; input the resource scheduling request, supply capacity data, and production plan data into the resource demand prediction model to obtain second predicted resource demand data by using the resource demand prediction model; acquire procurement cost data corresponding to the second predicted resource demand data, evaluation information of each supply provider, inventory cost data, and shortage cost data; and acquire target resource procurement quantities and target resource procurement times based on the second predicted resource demand data, procurement cost data, evaluation information, inventory cost data, and shortage cost data. The data input module is configured to input the target resource procurement quantity, the target resource procurement time, production plan data, real-time inventory data, transportation cost data, and a user satisfaction index into a strategy initialization module included in a pre-constructed resource scheduling strategy construction model, and obtain an initial resource scheduling strategy through the strategy initialization module; the initial resource scheduling strategy includes a plurality of resource scheduling groups, and each resource scheduling group is used to represent an allocation quantity of any resource at a current production site; The resource scheduling module is configured to input the initial resource scheduling strategy into a strategy optimization module included in the resource scheduling strategy construction model, obtain a target resource scheduling strategy through the strategy optimization module, and perform resource scheduling according to the target resource scheduling strategy.

6. The apparatus of claim 5, wherein, The strategy optimization module includes a group selection submodule and a group change submodule; The resource scheduling module further includes: A target resource scheduling group determination submodule is configured to input the initial resource scheduling strategy into the group selection submodule, and obtain a target resource scheduling group through the group selection submodule; A change submodule is configured to input the target resource scheduling group into the group change submodule, perform crossover operation and mutation operation on the target resource scheduling group, and obtain a changed target resource scheduling group; A current resource scheduling strategy construction submodule is configured to construct a current resource scheduling strategy according to the changed target resource scheduling group and other resource scheduling groups in the initial resource scheduling strategy except the target resource scheduling group; A target resource scheduling strategy determination submodule is configured to, in a case where a termination judgment result of the current resource scheduling strategy meets a preset condition, take the current resource scheduling strategy as the target resource scheduling strategy.

7. The apparatus of claim 6, wherein, The resource scheduling module further includes: The target resource scheduling strategy determination submodule is further configured to, based on the current resource scheduling strategy and a pre-set fitness function, obtain a fitness corresponding to the current resource scheduling strategy; in a case where a group change number of the group change submodule reaches a pre-set iteration number or the fitness reaches a pre-set fitness threshold, take the current resource scheduling strategy as the target resource scheduling strategy.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to implement the steps of the method in any one of claims 1 to 4.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 4.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 4.

Images