Material allocation scheme generation method and system

Abstract

The invention provides a material allocation scheme generation method and system, and the method comprises the steps: obtaining electric power material data corresponding to a material business characteristic tag from a data platform based on the material business characteristic tag inputted by a user; based on a pre-constructed reservoir age analysis model, calculating the reservoir age of each type of electric power materials in the electric power material data to obtain the reservoir age of each type of electric power materials and a risk probability value of becoming an overstocked material; and based on the electric power material data, the storage age of each type of electric power materials and the risk probability value of becoming the overstocked materials, solving to obtain an optimal allocation scheme of each type of electric power materials by taking the lowest storage logistics cost and the shortest emergency response time as targets. Through the constructed library age analysis model, the optimal allocation scheme of each type of electric power materials can be reliably and quickly generated, the preferential use of the materials in the library is facilitated, and the waste caused by repeated purchase is avoided.

Description

Technical Field

[0001] This invention belongs to the technical field of dynamic resource allocation schemes, specifically relating to a method and system for generating resource allocation schemes. Background Technology

[0002] With the power grid company's strategic goal of "building a world-class energy internet enterprise with outstanding global competitiveness," the company, as a leader in energy industry supply chain management, is applying new technologies to drive the transformation of traditional supply chains into modern (intelligent) supply chains. Inventory management in power grid companies is directly related to the operation of corporate assets and funds. Inventory backlog inevitably ties up a large amount of working capital, serving as a barometer of the company's working capital and being crucial to the company's management and decision-making.

[0003] Materials in stock should be used first to avoid duplicate purchases and waste, but currently, accurate information on the age of materials in stock cannot be obtained reliably and quickly. Summary of the Invention

[0004] To overcome the shortcomings of the prior art, in a first aspect, this invention application proposes a method for generating a material allocation plan, comprising: Based on the material business characteristic tags input by the user, the power material data corresponding to the material business characteristic tags is obtained from the data platform; Based on the pre-built inventory age analysis model, the inventory age of each type of power material in the power material data is calculated to obtain the inventory age of each type of power material and the risk probability value of becoming stockpiled material. Based on the power material data, the inventory age of each type of power material, and the risk probability value of becoming stockpiled materials, the optimal allocation scheme for each type of power material is obtained with the objectives of minimizing warehousing and logistics costs and minimizing emergency response time. The pre-construction process of the inventory aging analysis model includes: determining a material business characteristic label system for each type of power material based on historical power material data; constructing an inventory aging management classification tree for the material business characteristic labels corresponding to the material business characteristic label system for each type of power material; determining the influencing factors for each type of power material in batch management, inventory transfer, and inventory consumption based on the inventory aging management classification tree; constructing the inventory aging analysis model based on the material business characteristic labels for each type of power material, the influencing factors, and the inventory turnover optimization model; and the inventory turnover optimization model is pre-constructed based on the historical power material data and the inventory aging management strategy.

[0005] Preferably, the calculation of the inventory age of each type of power material in the power material data based on the pre-built inventory age analysis model, to obtain the inventory age of each type of power material and the probability value of becoming stockpiled material, includes: Based on a pre-built inventory age analysis model, the inbound time series, outbound frequency characteristics, and inventory dynamic change patterns of each type of power material are extracted. Based on the material business characteristic tags, the inventory age of each type of power material in the power material data is calculated to obtain the inventory age of each type of power material. Based on the inbound time series, outbound frequency characteristics, and inventory dynamic change patterns of each type of power material, as well as the preset rules for stockpiled materials, the risk probability value of each type of power material becoming stockpiled material is determined.

[0006] Preferably, the step of determining the material business characteristic labeling system for each type of power material based on historical power material data includes: Based on historical power materials data, we extract warehouse geographic coordinates, material technical parameter descriptions, and inventory dynamic status information for each type of power materials. By using a preset tag weight configuration algorithm, influence weights are assigned to tags of different material business characteristics. Based on the influence weight of each type of power material, the warehouse geographical coordinates, material technical parameter descriptions, and inventory dynamic status information of each type of power material are integrated in multiple dimensions to generate a material business characteristic tag system with a hierarchical structure.

[0007] Preferably, the process of constructing the inventory turnover optimization model includes: Based on historical power material data, the basic material attribute characteristics, business flow characteristics, and spatiotemporal distribution characteristics of each type of power material are extracted to construct a multi-dimensional feature vector. The basic classification of historical power material business is obtained and integrated into the multi-dimensional feature vector according to the correlation between power material category and business scenario to form a corresponding relationship matrix; Based on the aforementioned correspondence matrix, with the goal of maximizing inventory turnover rate and using the aforementioned inventory aging management strategy as a constraint, an inventory turnover optimization model is constructed.

[0008] Preferably, the pre-construction process of the inventory aging management strategy includes: Based on historical power material data, the random forest algorithm was used to identify core material data affecting inventory turnover through feature importance ranking and Gini coefficient analysis. An adaptive Gaussian mixture model is used to extract representative typical material data based on the distribution characteristics of historical power material data. Based on the inventory age management requirements for various types of power materials, a core early warning threshold is set for each type of power material to form a material management strategy. Through strategy iterative optimization algorithms, the material management strategy is collaboratively optimized with the core material data and the typical material data to construct an inventory age management strategy.

[0009] Preferably, the step of setting core early warning thresholds for each type of power material based on the inventory age management requirements of various power materials to form a material management strategy includes: For the historical power material data, a multimodal anomaly detection method is used, combined with box plot statistical analysis and / or scatter plot distribution clustering, to identify abnormal material data and normal material data; Root cause analysis was performed on the aforementioned abnormal material data to determine the cause of the abnormality. If the cause of the anomaly is an input error, the abnormal material data will be corrected to normal material data; if the cause of the anomaly is an anomaly in power material business, business analysis results will be generated by combining business graph analysis technology. Based on the inventory age management requirements of the various types of power materials, the business analysis results, and the normal data of the materials, a core early warning threshold is set for each type of power material as a material management strategy.

[0010] Preferably, if the cause of the anomaly is an anomaly in power material services, then by combining business graph analysis technology, a business analysis result is formed, including: If the cause of the anomaly is an anomaly in the power materials business, then a multi-dimensional analysis network containing business process nodes, resource relationships, and impact propagation paths will be constructed using business graph analysis technology. Based on the aforementioned multidimensional analysis network, the impact of power material business anomalies is assessed and root cause localization is performed, generating business analysis results.

[0011] Preferably, the step of identifying core material data affecting inventory turnover based on historical power material data, using a random forest algorithm, and through feature importance ranking and Gini coefficient analysis, includes: Based on the multi-dimensional characteristics of the historical targets of power material business, the random forest algorithm is used to rank the features by the Gini coefficient to obtain the ranking results. Based on the sorting results, a subset of key features affecting inventory turnover is selected from historical power material data; Based on the historical power material data and the key feature subset, the core material data affecting inventory turnover is identified.

[0012] Secondly, this invention application also proposes a material allocation scheme generation system, comprising: The power material data acquisition module is used to acquire power material data corresponding to the material business characteristic tags input by the user from the data platform. The calculation module is used to calculate the inventory age of each type of power material in the power material data based on a pre-built inventory age analysis model, to obtain the inventory age of each type of power material and the risk probability value of becoming stockpiled material. The pre-construction process of the inventory age analysis model includes: determining a material business characteristic tag system for each type of power material based on historical power material data; constructing an inventory age management classification tree for the material business characteristic tags corresponding to the material business characteristic tag system for each type of power material; determining the influencing factors for each type of power material in batch management, inventory transfer, and inventory consumption processes based on the inventory age management classification tree; and constructing the inventory age analysis model based on the material business characteristic tags of each type of power material, the influencing factors, and the inventory turnover optimization model. The inventory turnover optimization model is pre-constructed based on the historical power material data and the inventory age management strategy. The optimal allocation scheme solution module is used to solve for the optimal allocation scheme of each type of power material based on the power material data, the inventory age of each type of power material and the risk probability value of becoming stockpiled materials, with the objectives of minimizing warehousing and logistics costs and minimizing emergency response time.

[0013] Furthermore, the computing module includes: The extraction submodule is used to extract the inbound time series, outbound frequency characteristics and inventory dynamic change patterns of each type of power material based on a pre-built inventory age analysis model. The inventory age calculation submodule is used to calculate the inventory age of each type of power material in the power material data according to the material business characteristic tags, so as to obtain the inventory age of each type of power material. The risk probability value calculation submodule is used to determine the risk probability value of each type of power material becoming stockpiled material based on the inbound time series, outbound frequency characteristics, and inventory dynamic change patterns of each type of power material, as well as the preset stockpiled material rules.

[0014] Furthermore, the system also includes a library age analysis model construction module, used for: Based on historical power materials data, we extract warehouse geographic coordinates, material technical parameter descriptions, and inventory dynamic status information for each type of power materials. By using a preset tag weight configuration algorithm, influence weights are assigned to tags of different material business characteristics. Based on the influence weight of each type of power material, the warehouse geographical coordinates, material technical parameter descriptions, and inventory dynamic status information of each type of power material are integrated in multiple dimensions to generate a material business characteristic tag system with a hierarchical structure.

[0015] Furthermore, the system also includes an inventory turnover optimization model construction module, used for: Based on historical power material data, the basic material attribute characteristics, business flow characteristics, and spatiotemporal distribution characteristics of each type of power material are extracted to construct a multi-dimensional feature vector. The basic classification of historical power material business is obtained and integrated into the multi-dimensional feature vector according to the correlation between power material category and business scenario to form a corresponding relationship matrix; Based on the aforementioned correspondence matrix, with the goal of maximizing inventory turnover rate and using the aforementioned inventory aging management strategy as a constraint, an inventory turnover optimization model is constructed.

[0016] Furthermore, the system also includes a warehouse aging management strategy construction module; the warehouse aging management strategy construction module includes: The core material data identification submodule is used to identify core material data affecting inventory turnover based on historical power material data, using the random forest algorithm, feature importance ranking and Gini coefficient analysis. The typical material data extraction submodule is used to extract representative typical material data based on the distribution characteristics of historical power material data using an adaptive Gaussian mixture model. The materials management strategy determination submodule is used to set core early warning thresholds for each type of power materials according to the inventory age management requirements of various power materials, and form a materials management strategy. The inventory age management strategy construction submodule is used to construct an inventory age management strategy by co-optimizing the materials management strategy with the core materials data and the typical materials data through a strategy iterative optimization algorithm.

[0017] Furthermore, the materials management strategy determination submodule includes: The data identification unit is used to identify abnormal and normal data of the historical power materials by using a multimodal anomaly detection method, combined with box plot statistical analysis and / or scatter plot distribution clustering. An anomaly cause analysis unit is used to perform root cause analysis based on the material anomaly data to obtain the anomaly cause; The business analysis result acquisition unit is used to correct the abnormal material data to normal material data if the cause of the abnormality is an input error; and to form a business analysis result by combining business graph analysis technology if the cause of the abnormality is an abnormality in power material business. The materials management strategy determination unit is used to set core early warning thresholds for each type of power materials based on the inventory age management requirements of the various types of power materials, the business analysis results, and the normal data of the materials, as a materials management strategy.

[0018] Furthermore, the business analysis result acquisition unit is specifically used for: If the cause of the anomaly is an anomaly in the power materials business, then a multi-dimensional analysis network containing business process nodes, resource relationships, and impact propagation paths will be constructed using business graph analysis technology. Based on the aforementioned multidimensional analysis network, the impact of power material business anomalies is assessed and root cause localization is performed, generating business analysis results.

[0019] Furthermore, the core material data identification submodule is specifically used for: Based on the multi-dimensional characteristics of the historical targets of power material business, the random forest algorithm is used to rank the features by the Gini coefficient to obtain the ranking results. Based on the sorting results, a subset of key features affecting inventory turnover is selected from historical power material data; Based on the historical power material data and the key feature subset, the core material data affecting inventory turnover is identified.

[0020] Thirdly, this application also proposes an electronic device, comprising: at least one processor and a memory; wherein the memory and the processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, the method for generating a material allocation plan is implemented.

[0021] Fourthly, this application also proposes a readable storage medium having an executable program stored thereon, wherein when the executable program is executed, it implements the aforementioned method for generating a material allocation scheme.

[0022] Compared with the closest prior art, the present invention application has the following beneficial effects: The present invention provides a method and system for generating material allocation schemes, comprising: obtaining power material data corresponding to the material business characteristic tags input by the user from a data platform; and calculating the inventory age of each type of power material in the power material data based on a pre-built inventory age analysis model to obtain the inventory age of each type of power material and the risk probability value of becoming stockpiled material. Based on the power material data, the inventory age of each type of power material, and the probability of becoming stockpiled materials, the optimal allocation scheme for each type of power material is obtained with the objectives of minimizing warehousing and logistics costs and shortening emergency response time. The pre-construction process of the inventory age analysis model includes: determining a material business characteristic tagging system for each type of power material based on historical power material data; constructing an inventory age management classification tree for the material business characteristic tags corresponding to the material business characteristic tagging system for each type of power material; determining the influencing factors for each type of power material in batch management, inventory transfer, and inventory consumption based on the inventory age management classification tree; and constructing the inventory age analysis model based on the material business characteristic tags of each type of power material, the influencing factors, and the inventory turnover optimization model. The inventory turnover optimization model is pre-constructed based on the historical power material data and the inventory age management strategy. Through the constructed inventory age analysis model, the optimal allocation scheme for each type of power material can be reliably and quickly generated, which is beneficial for prioritizing the use of in-stock materials and avoiding waste caused by duplicate purchases. Attached Figure Description

[0023] Figure 1 A flowchart of a method for generating a material allocation scheme provided in this invention application; Figure 2 An architecture diagram of a material allocation scheme generation system provided in this invention application; Figure 3 The deployment topology of the system on the State Grid Cloud Platform in Embodiment 2 provided in this application; Figure 4 A schematic diagram of data flow and functional module linkage of the system in a typical business scenario of the State Grid in Embodiment 2 provided for this invention application; Figure 5 This is a schematic diagram of the operation of an electronic device provided in this invention application. Detailed Implementation

[0024] The specific embodiments of this invention will be further described in detail below with reference to the accompanying drawings.

[0025] Example 1: like Figure 1 As shown, this invention application proposes a method for generating a material allocation plan, which may include: Step 1: Based on the material business characteristic tags input by the user, obtain the power material data corresponding to the material business characteristic tags from the data platform; Step 2: Based on the pre-built inventory aging analysis model, calculate the inventory aging of each type of power material in the power material data to obtain the inventory aging of each type of power material and the risk probability value of becoming stockpiled material; wherein, the pre-construction process of the inventory aging analysis model includes: determining the material business characteristic tag system for each type of power material based on historical power material data; constructing an inventory aging management classification tree for the material business characteristic tags corresponding to the material business characteristic tag system for each type of power material; determining the influencing factors of each type of power material in batch management, inventory transfer, and inventory consumption based on the inventory aging management classification tree; constructing the inventory aging analysis model based on the material business characteristic tags of each type of power material, the influencing factors, and the inventory turnover optimization model; the inventory turnover optimization model is pre-constructed based on the historical power material data and inventory aging management strategy; Step 3: Based on the power material data, the inventory age of each type of power material, and the risk probability value of becoming stockpiled materials, with the objectives of minimizing warehousing and logistics costs and minimizing emergency response time, the optimal allocation scheme for each type of power material is obtained.

[0026] In steps 1-3 above, users can select or input material business characteristic tags (such as "surplus returned materials" or "project temporary storage materials") through the interface to filter out the corresponding power material data as the target power material data. Subsequently, a pre-built inventory aging analysis model is invoked. The construction of this inventory aging analysis model is the core of the method of this invention: First, based on historical power material data, a material business characteristic tag system for each type of material is determined (e.g., combining the basic classification of four elements: "grid province, factory, material, batch," and material business characteristic tags such as "inventory status" and "inventory aging"); then, a clear inventory aging management classification tree is constructed; next, by analyzing the business flow process, the influencing factors of materials in batch management, inventory transfer (e.g., distinguishing between valid and invalid transfers), and inventory consumption processes are determined; finally, these influencing factors are combined with an inventory turnover optimization model built based on historical power material data, aiming to improve turnover rate, to form the final inventory aging analysis model. After the inventory aging analysis model runs, it outputs the precise inventory aging of each type of material and the probability value of becoming stockpiled material. Ultimately, with the goals of minimizing warehousing and logistics costs and minimizing emergency response time, optimization algorithms are used to generate the optimal allocation plan for each type of power supply.

[0027] In step 2 above, when pre-constructing the inventory age analysis model, based on historical power material data, the material business characteristic tagging system for each type of power material is determined, which may include: Step 2a.1: Based on historical power material data, extract the warehouse geographic coordinates, material technical parameter descriptions, and inventory dynamic status information for each type of power material; Step 2a.2: Assign influence weights to different material business characteristic tags using a preset tag weight configuration algorithm; based on the influence weights of each type of power material, integrate the warehouse geographic coordinates, material technical parameter descriptions, and inventory dynamic status information of each type of power material in multiple dimensions to generate a material business characteristic tag system with a hierarchical structure.

[0028] The specific process of constructing the material business characteristic tagging system in steps 2a.1-2a.2 above is as follows: Based on historical power material data obtained from the data platform, the warehouse geographical coordinates (from the warehouse master data definition table), material technical parameter descriptions (from the material mapping table), and inventory dynamic status information (such as "in transit," "in stock," "frozen," etc., from the batch inventory table) of each type of material are extracted. Subsequently, using a preset tag weight configuration algorithm, such as the AHP (Analytic Hierarchy Process) based on business expert experience, influence weights are assigned to different material business characteristic tags such as "storage location," "material type," and "inventory status." Finally, based on these weights, the above multi-source information is multi-dimensionally integrated to generate a standardized material business characteristic tagging system with a two-layer structure of "basic classification - characteristic tag." For example, a two-layer material business characteristic tagging system is as follows: Assuming there is a batch of 10kV oil-immersed transformers, such a material business characteristic tagging system can be constructed for them:

[0029] The aforementioned construction of the inventory aging management classification tree can integrate the material attribute characteristics of each province and network, adhering to the principle of clear category distinction and classification design. It can integrate material business characteristic tags from relevant information such as batch inventory tables, organizational structure mapping tables, assessment scope tables, professional warehouse inventory tables, warehouse master data definition tables, material mapping tables, goods movement tables, and material voucher tables, and standardize material classification based on key dimensions such as storage mode, storage location, and material type, thereby forming the inventory aging management classification tree. Subsequently, based on the inventory aging management classification tree, an analysis of the influencing factors of inventory aging management is carried out. This process comprehensively considers the influencing factors of various materials in batch management, inventory transfer, and inventory consumption processes through the analysis of key material business flow processes.

[0030] Furthermore, in step 2, the process of constructing the inventory turnover optimization model may include: Step 2b.1: Based on historical power material data, extract the basic material attribute features, business flow features, and spatiotemporal distribution features of each type of power material to construct a multi-dimensional feature vector; Step 2b.2: The basic classification of historical power material business is obtained and integrated into the multi-dimensional feature vector according to the correlation between power material category and business scenario to form a corresponding relationship matrix; Step 2b.3: Based on the correspondence matrix, with the goal of maximizing inventory turnover rate and with the inventory aging management strategy as a constraint, build an inventory turnover optimization model.

[0031] The above-mentioned inventory turnover optimization model is constructed as follows: First, based on historical power material data, the basic material attribute features (such as material code, type), business flow features (such as inbound / outbound frequency, average inventory age), and spatiotemporal distribution features (such as warehouse level, region) of each type of material are extracted from master data and business flow data to jointly construct a multi-dimensional feature vector. Next, the correlation between the classification elements defined in the basic classification system (such as "network province, factory, material, batch") and specific business scenarios (such as "project materials", "emergency spare parts") is integrated into the above multi-dimensional feature vector to construct a structured correspondence matrix. Finally, based on this correspondence matrix, with maximizing inventory turnover rate as the core optimization objective, and introducing inventory age management strategies, such as "First-In-First-Out (FIFO)" or "red, yellow, green warning thresholds," as constraints, a quantifiable inventory turnover optimization model is built.

[0032] Furthermore, in step 2, the pre-construction process of the inventory aging management strategy may include: Step 2C.1: Based on historical power material data, the random forest algorithm is used to identify core material data affecting inventory turnover through feature importance ranking and Gini coefficient analysis; Step 2C.2: Using an adaptive Gaussian mixture model, representative typical material data are extracted based on the distribution characteristics of historical power material data; Step 2C.3: Based on the inventory age management requirements of various types of power materials, set core early warning thresholds for each type of power material to form a material management strategy; Step 2C.4: Through strategy iterative optimization algorithm, coordinate the material management strategy with the core material data and the typical material data to construct an inventory age management strategy.

[0033] The construction process of the inventory aging management strategy in steps 2C.1-2C.4 above is as follows: First, based on historical power material data, random forest is used to perform feature importance ranking and Gini coefficient analysis to identify core material data that has a key impact on inventory turnover from numerous fields such as "inventory location," "material description," and "amount." Simultaneously, an adaptive Gaussian mixture model is used to extract typical material data that represents most normal situations based on the distribution characteristics of historical power material data. Then, combining the management requirements of various power materials (such as the different inventory aging standards for "project temporary storage materials" and "surplus returned materials"), a multimodal anomaly detection method is used to set core early warning thresholds, thus initially forming a material management strategy. Finally, through a strategy iterative optimization algorithm, the above material management strategy, core material data, and typical material data are collaboratively simulated and dynamically adjusted to construct a practically executable inventory aging management strategy.

[0034] The core and typical material data processing flow described above is as follows: Step C.1: Identification and Clarification of Core Material Data. In conjunction with the business objectives of inventory turnover analysis, clarify and define the scope of core material data (e.g., material description, inventory location, value, etc.). Subsequently, collaborate with business departments and data providers to clarify data definitions, statistical standards, and calculation logic, ensuring consistent understanding. Simultaneously, check data completeness, determining if there are missing fields or incomplete records. For missing data, assess its impact on business analysis, deciding whether to supplement data collection or use reasonable estimation methods. Step C.2: Extraction and Integration of Typical Material Data. Based on the characteristics of typical material data, determine the screening dimensions (e.g., time period, business scenario). Using data analysis tools, select representative data according to certain time periods, business scenarios, or user groups. Collect the data to the data platform following unified standards. During this process, perform preliminary data verification to ensure correct format and reasonable values, avoiding deviations in subsequent analysis due to incorrect data format or abnormal values.

[0035] In step 2C.1 above, the step of identifying core material data affecting inventory turnover based on historical power material data using a random forest algorithm, through feature importance ranking and Gini coefficient analysis, may include: Step 2C.1.1: Based on the multi-dimensional characteristics of the historical targets of power material business, the random forest algorithm is used to rank the features by Gini coefficient to obtain the ranking results; Step 2C.1.2: Based on the sorting results, select a subset of key features affecting inventory turnover from the historical power materials data; Step 2C.1.3: Based on the historical power material data and the key feature subset, identify the core material data that affects inventory turnover.

[0036] In steps 2C.1.1-2C.1.3 above, firstly, based on the acquired historical power material data, multi-dimensional features including "provincial company," "inventory location," "material description," "quantity," "amount," "inventory status," and "inventory age" are extracted, and a random forest model is constructed using these features. This random forest model ranks the features by importance using the Gini coefficient, thus obtaining a quantified sequence of feature importance. Subsequently, based on this ranking result (for example, identifying "inventory age" and "amount" as the most important features), corresponding key feature subsets are selected from the historical power material data. Finally, by combining the complete historical power material data with the extracted key feature subsets, core material data that significantly impacts inventory turnover is accurately located, providing clear data support for the formulation of subsequent inventory age management strategies.

[0037] Further, in step 2C.3, the step of setting core early warning thresholds for each type of power material based on the inventory age management requirements of various power materials to form a material management strategy includes: Step 2C.3.1: For the historical power material data, a multimodal anomaly detection method is used, combined with box plot statistical analysis and / or scatter plot distribution clustering, to identify abnormal material data and normal material data; Step 2C.3.2: Perform root cause analysis based on the abnormal material data to obtain the cause of the abnormality; Step 2C.3.3: If the cause of the anomaly is an input error, then correct the abnormal material data to normal material data; if the cause of the anomaly is an anomaly in power material business, then combine business graph analysis technology to form business analysis results. Step 2C.3.4: Based on the inventory age management requirements of the various types of power materials, the business analysis results, and the normal data of the materials, set a core early warning threshold for each type of power material as a material management strategy. In steps 2C.3.1-2C.3.4 above, the process begins with meticulous preprocessing of historical power material data: a multimodal anomaly detection method is employed, combining box plot statistical analysis (used to identify outliers in numerical fields such as "amount" and "quantity") and scatter plot distribution clustering (used to discover abnormal combinations in business logic), thereby accurately distinguishing abnormal material data from normal material data. Subsequently, root cause analysis is performed on the identified abnormal material data. If it is an input error (such as incorrect quantity units), it is corrected; if it is indeed a power material business anomaly (such as a certain material in a specific warehouse having no entry or exit records for a long period of time), in-depth analysis is conducted using business graph analysis technology to form business analysis results. Finally, based on the inventory age management requirements of various types of power materials, the aforementioned business analysis results, and the cleaned normal material data, differentiated core early warning thresholds are set for each type of material (e.g., a three-level early warning mechanism of "red, yellow, and green"), thereby forming a practically executable material management strategy. The above-mentioned methods for identifying and processing abnormal material data include: using statistical methods and business experience to identify abnormal material data; specific methods include: setting a reasonable threshold range by calculating the mean and standard deviation, and considering data exceeding this range as abnormal material data; and using visualization tools such as box plots and scatter plots to visually identify outliers in the data as abnormal material data. For identified abnormal material data, it is necessary to analyze its causes in depth to determine whether it stems from data entry errors, system malfunctions, or genuine business anomalies. If it is a data error, timely communication and correction with relevant personnel is required; if it is a business anomaly, a detailed analysis based on the business context is necessary, and it should be retained and specifically explained in the analysis report if necessary to ensure the accuracy and reliability of the data processing results and provide effective support for subsequent strategy formulation.

[0038] The aforementioned materials management strategy is designed to adapt to the differences in system application models and implementation methods among different provincial network companies. Following the principle of "one item, one benchmark," a differentiated management strategy is developed: First, parameters are configured based on the differences in material characteristics, and batch modeling technology is used to set a unique benchmark inventory age for each type of material. Second, a "red, yellow, and green" graded early warning mechanism is established, with three warning thresholds corresponding to severe backlog, stagnation, and healthy inventory age states, enabling rapid identification and differentiation of material inventory age. Finally, a comprehensive warehouse management system is implemented, strengthening inventory age management and rigidly enforcing the "first-in, first-out" principle to form a graded and classified management strategy that includes differentiated inventory age clearing targets and corresponding disposal measures, comprehensively improving the quality and efficiency of inventory clearance efforts.

[0039] Further, in step 2C.3.3, if the cause of the anomaly is an anomaly in power material services, then the business analysis results formed by combining business graph analysis technology may include: Step 2C.3.3.1: If the cause of the anomaly is an anomaly in the power materials business, then use business graph analysis technology to construct a multi-dimensional analysis network that includes business process nodes, resource association relationships, and impact propagation paths; Step 2C.3.3.2: Based on the multidimensional analysis network, assess the impact and root cause of power material business anomalies, and generate business analysis results.

[0040] In steps 2C.3.3.1-2C.3.3.2 above, when the root cause of the anomaly is an anomaly in the power materials business, how to combine business graph analysis technology to form business analysis results? The specific process is as follows: First, a multi-dimensional analysis network is constructed using business graph analysis technology. This multi-dimensional analysis network uses entities such as materials, warehouses, projects, and suppliers as business process nodes, and inventory transfers and requisition relationships as resource associations, clearly depicting the impact propagation path of anomalies such as inventory backlog along these relationship chains. Based on this multi-dimensional analysis network, the identified business anomalies are quantitatively impacted (e.g., determining the number and scope of downstream projects affected by the anomaly) and precisely rooted (e.g., tracing back to material retention caused by changes in the requirements of a specific project). Finally, business analysis results containing quantitative impact assessment and qualitative cause analysis are generated, providing core decision-making basis for subsequently formulating precise and differentiated materials management strategies.

[0041] Furthermore, in step 2 above, the calculation of the inventory age of each type of power material in the power material data based on the pre-built inventory age analysis model, to obtain the inventory age of each type of power material and the probability value of it becoming stockpiled material, may include: Step 2.1: Based on the pre-built inventory aging analysis model, extract the inbound time series, outbound frequency characteristics and inventory dynamic change patterns of each type of power material; Step 2.2: Calculate the inventory age of each type of power material in the power material data according to the material business characteristic tags to obtain the inventory age of each type of power material; Step 2.3: Based on the inbound time series, outbound frequency characteristics, and inventory dynamic change patterns of each type of power material, as well as the preset rules for stockpiled materials, determine the risk probability value of each type of power material becoming stockpiled material.

[0042] In steps 2.1-2.3 above, the specific calculation process of the inventory aging analysis model is as follows: First, the inventory aging analysis model extracts the inbound time series of each type of power material from the goods movement table and material voucher table, analyzes the outbound frequency characteristics from the outbound records, and identifies dynamic inventory change patterns from continuous inventory snapshots. Then, according to the constructed material business characteristic labeling system (for example, for materials with the material business characteristic label "surplus returned"), its inventory age from inbound to present is calculated. Finally, combined with preset rules for stockpiled materials, such as "project temporary storage materials that have been in stock for more than 12 months" or "surplus returned materials that have been stockpiled for more than 3 years," the model comprehensively considers the inbound time series, outbound frequency characteristics, and dynamic inventory change patterns, and uses a probabilistic statistical model to quantify the risk probability value of this type of material becoming stockpiled material.

[0043] When implementing the method of this invention, a series of data preparation tasks can be carried out in advance for key business fields (including provincial company, inventory location, warehouse location, material description, quantity, amount, inventory status, inventory age, etc.) in the batch inventory table, organization mapping table, evaluation scope table, professional warehouse inventory table, warehouse master data definition table, material mapping table, goods movement table, and material voucher table. These tasks include data selection and clarification, extraction and collection of typical material data, and identification and processing of abnormal material data. This process constructs a thematic database suitable for inventory age management business analysis. Subsequently, when using these tables and data, they can be directly extracted and used from the thematic database.

[0044] The purpose of this invention is to enable precise calculation of inventory age by setting personalized parameters based on the attributes of each type of power material through the inventory age analysis model and supporting methods provided by this invention. This assists enterprises in establishing an information sharing mechanism for usable inventory materials, thereby strengthening the unified allocation capability of material resources. While ensuring the supply of materials quickly and efficiently, this method also helps to tap potential and increase efficiency, improving overall management effectiveness. Addressing the need for prioritizing the use of inventory materials and avoiding duplicate purchases, the inventory turnover optimization model constructed in this invention provides effective information support for implementing the principle of "utilizing inventory first, then purchasing," assisting enterprises in rationally formulating material reserve quotas and achieving scientific inventory reduction. Based on the standardization of benchmark inventory age, a further inventory age analysis model is constructed to accurately reflect the inventory health index, achieving real-time monitoring of inventory operations and transparent management of inventory resources.

[0045] The method of this invention, through an inventory aging analysis model, enables enterprises to accurately grasp the inventory duration of each type of material, meeting the management requirements of internal control and external audit, and providing information support for implementing green, modern, and intelligent supply chains to improve quality and efficiency and achieve routine "refined physical management." Through an inventory turnover optimization model, it enhances the initiative of business departments in implementing the principle of "utilizing inventory first, then purchasing," assisting enterprises in achieving the management goals of "making good use of existing stock, standardizing and controlling new stock, reducing holding costs, and improving efficiency and effectiveness." By constructing a warehousing and logistics performance indicator system and a visual dashboard, a monitoring and early warning mechanism is established to achieve intelligent evaluation and multi-dimensional visual comparison of the effectiveness of inventory aging management.

[0046] The beneficial effects of this invention are: This invention addresses the differentiated needs of various provincial network companies in terms of inventory aging management (primarily reflected in information system application models, batch management methods, and material movement types). By constructing an inventory aging management classification tree, integrating the material attribute characteristics of each network province, and standardizing material classification based on key dimensions such as storage mode, storage location, and material type, this invention further analyzes the influencing factors of inventory aging management. Through a comprehensive review of the key business flow processes of materials, it considers the key influencing factors of various materials in batch management, inventory transfer, and inventory consumption, providing a basis for developing differentiated and refined inventory aging management strategies.

[0047] This invention constructs an inventory turnover optimization model based on inventory and historical inbound / outbound information. Its design fully considers the differentiated needs of different provincial network companies in terms of inventory age control. These differences are specifically reflected in: inconsistencies or variability in the application modes and implementation methods of existing information in various units; the coexistence of batch management, non-batch management, and hybrid management modes; and the complexity and diversity of material movement scenarios, including transfers between inventory locations, between factories, between companies, and between physical and virtual warehouses. This invention supports integrated research on business operations and data, specifically including: establishing a standardized material business characteristic labeling system; creating a correspondence matrix of "basic classification + characteristic labels" in inventory turnover optimization management to provide a business framework for constructing an inventory aging analysis model; promoting data cleaning and quality analysis during the construction of the inventory aging analysis model, covering the selection and clarification of key business fields, the extraction and collection of typical material data, and the identification and processing of abnormal material data; and constructing an inventory turnover optimization model through deep integration of business rules and data characteristics, forming a model system covering material inventory aging calculation and inventory utilization optimization, providing algorithmic support for achieving dynamic and precise physical management. Ultimately, this creates a virtuous cycle of synergistic resonance and mutual promotion among multiple elements of management systems, business processes, and basic data.

[0048] When the method of the present invention is used for inventory management in power grid enterprises, the following effects can also be achieved: 1. Improve economic efficiency (1) Achieve accurate analysis of inventory age: The inventory management of power grid enterprises is directly related to the operation of enterprise assets and funds. Inventory backlog will inevitably occupy a large amount of working capital, which is a barometer of the enterprise's working capital and is crucial to the management and decision-making of the enterprise. Through the inventory age analysis model provided by this invention, personalized parameters can be set based on the attributes of each type of material to achieve accurate calculation of inventory age, help enterprises establish information sharing of usable inventory materials, thereby strengthening the unified allocation of material resources, ensuring material supply quickly and efficiently, and tapping potential to increase efficiency and improve management efficiency. (2) Support "utilize inventory first, then purchase" and achieve inventory reduction: Inventory materials should be used first to avoid repeated purchases and waste, but at present, material information and inventory age information have not been integrated into all warehouse levels, and it is impossible to reliably and quickly obtain accurate material information. By applying the inventory turnover optimization model of this invention, it provides strong information support for "utilize inventory first, then purchase", which can objectively help enterprises reasonably formulate material reserve quotas and achieve inventory reduction.

[0049] 2. Improve management level: (1) Construct relevant inventory assessment indicators to accurately reflect the warehouse management level: Based on the standardization of benchmark inventory age, construct an inventory age analysis model to accurately reflect the inventory age health index, realize real-time monitoring of inventory business, and make inventory resources transparent and visible. (2) Improve decision support capabilities: Based on the accurate and timely data analysis results generated by the inventory age analysis model and inventory turnover optimization model, provide management with a more scientific basis for decision-making, and effectively improve the strategic planning and execution capabilities of enterprises. (3) Clarify the goals of warehouse business and management personnel at all levels: Through the implementation of the inventory age analysis model and inventory turnover optimization model, a clear assessment target system has been established, which has enhanced the execution motivation of warehouse business and management personnel at all levels in terms of inventory reduction, and comprehensively improved the level of material support and guarantee.

[0050] 3. Enhance social benefits: (1) Optimize resource allocation and emergency response mechanism: The inventory aging analysis model and inventory turnover optimization model can strengthen the foundation of energy security and social operation by optimizing resource allocation and emergency response mechanism. It can improve the efficiency of material allocation in the power grid to deal with emergencies such as extreme weather and equipment failure, shorten the repair cycle, reduce power outage losses, and ensure the stability of residents' lives and key areas of electricity use. At the same time, through cross-regional and cross-project material coordination and scheduling, it breaks the decentralized barriers of traditional inventory management, realizes the efficient circulation and rational utilization of idle resources, fundamentally reduces the waste of material accumulation, and provides resilient support for the stable operation of the power system. (2) Serve green, low-carbon and rural revitalization strategy: The inventory aging analysis model and inventory turnover optimization model can be deeply integrated into the green, low-carbon and rural revitalization strategy and become an important engine for promoting sustainable social development. Under the "dual carbon" goal, it can help the energy structure to transform towards clean energy by promoting the consumption of new energy, optimizing energy storage configuration and promoting the recycling of waste materials, and accelerate the construction of a new type of low-consumption and high-efficiency power system. In rural development, a reservoir age analysis model can be set up to prioritize the transformation of rural power grids and the electricity demand of industries, shorten the construction cycle of power grids in remote areas, improve the reliability of power supply, provide power security for rural industrial upgrading and infrastructure improvement, effectively narrow the urban-rural development gap, and consolidate the energy foundation for rural revitalization. (3) Promote industry technological progress and policy coordination: This invention relies on digital and intelligent technologies to promote the intelligent transformation of the power supply chain and cultivate new resource management talents. Its deep integration with energy policies and regional planning has formed a cross-departmental coordination mechanism, providing a practical example for institutional innovation in the energy field and realizing the coordinated development of technological innovation and comprehensive social benefits.

[0051] Example 2: like Figure 2 As shown, the present invention also provides a material allocation scheme generation system, comprising: The power material data acquisition module is used to acquire power material data corresponding to the material business characteristic tags input by the user from the data platform. The calculation module is used to calculate the inventory age of each type of power material in the power material data based on a pre-built inventory age analysis model, to obtain the inventory age of each type of power material and the risk probability value of becoming stockpiled material. The pre-construction process of the inventory age analysis model includes: determining a material business characteristic tag system for each type of power material based on historical power material data; constructing an inventory age management classification tree for the material business characteristic tags corresponding to the material business characteristic tag system for each type of power material; determining the influencing factors for each type of power material in batch management, inventory transfer, and inventory consumption processes based on the inventory age management classification tree; and constructing the inventory age analysis model based on the material business characteristic tags of each type of power material, the influencing factors, and the inventory turnover optimization model. The inventory turnover optimization model is pre-constructed based on the historical power material data and the inventory age management strategy. The optimal allocation scheme solution module is used to solve for the optimal allocation scheme of each type of power material based on the power material data, the inventory age of each type of power material and the risk probability value of becoming stockpiled materials, with the objectives of minimizing warehousing and logistics costs and minimizing emergency response time.

[0052] Furthermore, the computing module includes: The extraction submodule is used to extract the inbound time series, outbound frequency characteristics and inventory dynamic change patterns of each type of power material based on a pre-built inventory age analysis model. The inventory age calculation submodule is used to calculate the inventory age of each type of power material in the power material data according to the material business characteristic tags, so as to obtain the inventory age of each type of power material. The risk probability value calculation submodule is used to determine the risk probability value of each type of power material becoming stockpiled material based on the inbound time series, outbound frequency characteristics, and inventory dynamic change patterns of each type of power material, as well as the preset stockpiled material rules.

[0053] Furthermore, the system also includes a library age analysis model construction module, used for: Based on historical power materials data, we extract warehouse geographic coordinates, material technical parameter descriptions, and inventory dynamic status information for each type of power materials. By using a preset tag weight configuration algorithm, influence weights are assigned to tags of different material business characteristics. Based on the influence weight of each type of power material, the warehouse geographical coordinates, material technical parameter descriptions, and inventory dynamic status information of each type of power material are integrated in multiple dimensions to generate a material business characteristic tag system with a hierarchical structure.

[0054] Furthermore, the system also includes an inventory turnover optimization model construction module, used for: Based on historical power material data, the basic material attribute characteristics, business flow characteristics, and spatiotemporal distribution characteristics of each type of power material are extracted to construct a multi-dimensional feature vector. The basic classification of historical power material business is obtained and integrated into the multi-dimensional feature vector according to the correlation between power material category and business scenario to form a corresponding relationship matrix; Based on the aforementioned correspondence matrix, with the goal of maximizing inventory turnover rate and using the aforementioned inventory aging management strategy as a constraint, an inventory turnover optimization model is constructed.

[0055] Furthermore, the system also includes a warehouse aging management strategy construction module; the warehouse aging management strategy construction module includes: The core material data identification submodule is used to identify core material data affecting inventory turnover based on historical power material data, using the random forest algorithm, feature importance ranking and Gini coefficient analysis. The typical material data extraction submodule is used to extract representative typical material data based on the distribution characteristics of historical power material data using an adaptive Gaussian mixture model. The materials management strategy determination submodule is used to set core early warning thresholds for each type of power materials according to the inventory age management requirements of various power materials, and form a materials management strategy. The inventory age management strategy construction submodule is used to construct an inventory age management strategy by co-optimizing the materials management strategy with the core materials data and the typical materials data through a strategy iterative optimization algorithm.

[0056] Furthermore, the materials management strategy determination submodule includes: The data identification unit is used to identify abnormal and normal data of the historical power materials by using a multimodal anomaly detection method, combined with box plot statistical analysis and / or scatter plot distribution clustering. An anomaly cause analysis unit is used to perform root cause analysis based on the material anomaly data to obtain the anomaly cause; The business analysis result acquisition unit is used to correct the abnormal material data to normal material data if the cause of the abnormality is an input error; and to form a business analysis result by combining business graph analysis technology if the cause of the abnormality is an abnormality in power material business. The materials management strategy determination unit is used to set core early warning thresholds for each type of power materials based on the inventory age management requirements of the various types of power materials, the business analysis results, and the normal data of the materials, as a materials management strategy.

[0057] Furthermore, the business analysis result acquisition unit is specifically used for: If the cause of the anomaly is an anomaly in the power materials business, then a multi-dimensional analysis network containing business process nodes, resource relationships, and impact propagation paths will be constructed using business graph analysis technology. Based on the aforementioned multidimensional analysis network, the impact of power material business anomalies is assessed and root cause localization is performed, generating business analysis results.

[0058] Furthermore, the core material data identification submodule is specifically used for: Based on the multi-dimensional characteristics of the historical targets of power material business, the random forest algorithm is used to rank the features by the Gini coefficient to obtain the ranking results. Based on the sorting results, a subset of key features affecting inventory turnover is selected from historical power material data; Based on the historical power material data and the key feature subset, the core material data affecting inventory turnover is identified.

[0059] To illustrate Embodiment 2 of the present invention: An example of a typical implementation method is given below: The system of this invention is deployed in the management information area of ​​the intranet, such as... Figure 3The diagram illustrates the system's deployment topology on the State Grid Cloud Platform, showcasing a single-level deployment architecture with cloud-based clusters and component collaboration. The system employs a single-level deployment architecture, deployed entirely on the State Grid Cloud Platform, with the management information zone serving as the core deployment area, secured by a firewall. The left side of the system architecture integrates multiple foundational platforms, including a data platform, business platform, and technology platform. These platforms interact with the core system and invoke functions through standardized APIs (Application Programming Interfaces). The system utilizes an ECS (Elastic Compute Service) cluster of EDAS (Enterprise Distributed Application Service) for application deployment. This cluster includes core components such as a Kubernetes application server cluster, database servers, and an SLB (Server Load Balancer), all centrally scheduled and managed through cloud platform management nodes. The database servers include an RDS-MySQL (Relational Database Service for MySQL) cloud database server and a Redis (RemoteDictionary Server) remote server. The system architecture includes a data exchange area to enable data interaction with external systems, and supports secure remote access via a VPN (Virtual Private Network Gateway) gateway. For storage, the system employs distributed cloud storage to ensure high availability and reliability of data. The system also features optimized inventory turnover calculation logic. The system supports a multi-level access control mechanism: headquarters users can query data from all units, while provincial / municipal company users can only query data from their own unit.

[0060] Secondly, the system's functions are geared towards users of the State Grid Materials Department and the Materials Company's warehousing and distribution departments, using a PC (Personal Computer) as the application platform. Its core functional logic is as follows: The system obtains relevant data from the data platform, including batch inventory tables, organizational structure mapping tables, assessment scope tables, specialized warehouse inventory tables, warehouse master data definition tables, material mapping tables, cargo movement tables, and material voucher tables. By associating inventory materials with material receipts, and comparing receipt times with the current time, it identifies three categories of untapped materials: project temporary storage materials that have been in storage for over 12 months, surplus materials returned for over 3 years, and supplier-deposited materials that have been in storage for over one year. Based on the four-element information (grid province, factory, material, batch) of batch materials in the previous month's inventory age details, it extracts the current month's material outbound details (excluding invalid transfer vouchers), and associates the inventory age details table with the current month's outbound details table according to the four elements. Following the FIFO principle, it derives the material voucher, outbound quantity, and inventory type corresponding to each inventory age detail. The system uses the following calculation formula for inventory utilization statistics: Based on the inventory age details of the last day of the month, the remaining inventory amount to be utilized this month is calculated as follows: Remaining inventory amount to be utilized this month = Inventory amount utilized this month + Remaining inventory amount to be utilized this month; Remaining inventory amount to be utilized this month = Inventory amount utilized this month / Inventory amount to be utilized this month. The system's PC interface displays, in modules, fields such as provincial company, inventory location, warehouse location, material description, quantity, amount, inventory status, and inventory age for project temporary storage materials that have been in storage for over 12 months, surplus materials returned to storage for over 3 years, and supplier-deposited materials that have been in storage for over one year. It also displays key inventory utilization performance indicators such as the inventory utilization amount and inventory count of project temporary storage materials with an inventory age of over 12 months, surplus materials returned to storage for projects with an inventory age of 3-5 years, and surplus materials returned to storage for projects with an inventory age of 5 years or more. Figure 4 The diagram illustrates the data flow and functional module linkage of the system in a typical business scenario, demonstrating the closed-loop management of the entire process from data acquisition and inventory aging analysis to inventory utilization statistics. Through the implementation of this embodiment, the system achieves accurate identification of the inventory age of power materials, dynamic statistics of inventory utilization progress, and multi-dimensional visualization, providing power grid companies with efficient and transparent inventory management and allocation decision support.

[0061] Example 3: like Figure 5 As shown, the present invention also provides an electronic device, which may be a computer device, a microcontroller device, a smart mobile device, etc. The electronic device in this embodiment may include a processor, a memory, a transceiver component, etc. The memory, processor, and transceiver component are connected via a bus; the memory can be used to store executable programs, and an exemplary executable program may include instructions; the processor is used to execute the instructions stored in the memory. The memory can also be used to store data, which can be accessed and / or modified when instructions are executed.

[0062] The processor may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and it is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions in the storage medium to realize the corresponding method flow or corresponding function, so as to realize the steps of the material allocation scheme generation method in the above embodiments.

[0063] Example 4: Based on the same inventive concept, this invention also provides a readable storage medium, specifically an electronic device readable storage medium (Memory). This readable storage medium is a memory device within an electronic device used to store programs and data. It is understood that the storage medium here can include both built-in storage media within the electronic device and extended storage media supported by the electronic device. The storage medium provides storage space, which stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more executable programs (including program code). It should be noted that the storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. Loading and executing one or more instructions stored in the storage medium by the processor can implement the steps of the material allocation scheme generation method in the above embodiments.

[0064] Those skilled in the art will understand that embodiments of this invention can be provided as methods, systems, or computer program products. Therefore, this invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0065] This invention application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0066] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0067] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit its protection scope. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading the present invention, they can still make various changes, modifications or equivalent substitutions to the specific implementation of the application, but these changes, modifications or equivalent substitutions are all within the protection scope of the claims pending approval.

Claims

1. A method for generating a material allocation plan, characterized in that, include: Based on the material business characteristic tags input by the user, the power material data corresponding to the material business characteristic tags is obtained from the data platform; Based on the pre-built inventory age analysis model, the inventory age of each type of power material in the power material data is calculated to obtain the inventory age of each type of power material and the risk probability value of becoming stockpiled material. Based on the power material data, the inventory age of each type of power material, and the risk probability value of becoming stockpiled materials, the optimal allocation scheme for each type of power material is obtained with the objectives of minimizing warehousing and logistics costs and minimizing emergency response time. The pre-construction process of the inventory aging analysis model includes: determining a material business characteristic label system for each type of power material based on historical power material data; constructing an inventory aging management classification tree for the material business characteristic labels corresponding to the material business characteristic label system for each type of power material; determining the influencing factors for each type of power material in batch management, inventory transfer, and inventory consumption based on the inventory aging management classification tree; constructing the inventory aging analysis model based on the material business characteristic labels for each type of power material, the influencing factors, and the inventory turnover optimization model; and the inventory turnover optimization model is pre-constructed based on the historical power material data and the inventory aging management strategy.

2. The method according to claim 1, characterized in that, The pre-built inventory age analysis model calculates the inventory age of each type of power material in the power material data, obtaining the inventory age of each type of power material and the probability value of it becoming stockpiled material, including: Based on a pre-built inventory age analysis model, the inbound time series, outbound frequency characteristics, and inventory dynamic change patterns of each type of power material are extracted. Based on the material business characteristic tags, the inventory age of each type of power material in the power material data is calculated to obtain the inventory age of each type of power material. Based on the inbound time series, outbound frequency characteristics, and inventory dynamic change patterns of each type of power material, as well as the preset rules for stockpiled materials, the risk probability value of each type of power material becoming stockpiled material is determined.

3. The method according to claim 1, characterized in that, The aforementioned system for determining the business characteristic labeling system for each type of power material based on historical power material data includes: Based on historical power materials data, we extract warehouse geographic coordinates, material technical parameter descriptions, and inventory dynamic status information for each type of power materials. By using a preset tag weight configuration algorithm, influence weights are assigned to tags of different material business characteristics. Based on the influence weight of each type of power material, the warehouse geographical coordinates, material technical parameter descriptions, and inventory dynamic status information of each type of power material are integrated in multiple dimensions to generate a material business characteristic tag system with a hierarchical structure.

4. The method according to claim 1, characterized in that, The process of constructing the inventory turnover optimization model includes: Based on historical power material data, the basic material attribute characteristics, business flow characteristics, and spatiotemporal distribution characteristics of each type of power material are extracted to construct a multi-dimensional feature vector. The basic classification of historical power material business is obtained and integrated into the multi-dimensional feature vector according to the correlation between power material category and business scenario to form a corresponding relationship matrix; Based on the aforementioned correspondence matrix, with the goal of maximizing inventory turnover rate and using the aforementioned inventory aging management strategy as a constraint, an inventory turnover optimization model is constructed.

5. The method according to claim 4, characterized in that, The pre-construction process of the inventory aging management strategy includes: Based on historical power material data, the random forest algorithm was used to identify core material data affecting inventory turnover through feature importance ranking and Gini coefficient analysis. An adaptive Gaussian mixture model is used to extract representative typical material data based on the distribution characteristics of historical power material data. Based on the inventory age management requirements for various types of power materials, a core early warning threshold is set for each type of power material to form a material management strategy. Through strategy iterative optimization algorithms, the material management strategy is collaboratively optimized with the core material data and the typical material data to construct an inventory age management strategy.

6. The method according to claim 5, characterized in that, Based on the inventory age management requirements for various types of power materials, a core early warning threshold is set for each type of power material to form a material management strategy, including: For the historical power material data, a multimodal anomaly detection method is used, combined with box plot statistical analysis and / or scatter plot distribution clustering, to identify abnormal material data and normal material data; Root cause analysis was performed on the aforementioned abnormal material data to determine the cause of the abnormality. If the cause of the anomaly is an input error, the abnormal material data will be corrected to normal material data; if the cause of the anomaly is an anomaly in power material business, business analysis results will be generated by combining business graph analysis technology. Based on the inventory age management requirements of the various types of power materials, the business analysis results, and the normal data of the materials, a core early warning threshold is set for each type of power material as a material management strategy.

7. The method according to claim 6, characterized in that, If the cause of the anomaly is an anomaly in power material services, then, in conjunction with business graph analysis technology, a business analysis result is generated, including: If the cause of the anomaly is an anomaly in the power materials business, then a multi-dimensional analysis network containing business process nodes, resource relationships, and impact propagation paths will be constructed using business graph analysis technology. Based on the aforementioned multidimensional analysis network, the impact of power material business anomalies is assessed and root cause localization is performed, generating business analysis results.

8. The method according to claim 5, characterized in that, Based on historical power material data, a random forest algorithm is used to identify core material data affecting inventory turnover through feature importance ranking and Gini coefficient analysis, including: Based on the multi-dimensional characteristics of the historical targets of power material business, the random forest algorithm is used to rank the features by the Gini coefficient to obtain the ranking results. Based on the sorting results, a subset of key features affecting inventory turnover is selected from historical power material data; Based on the historical power material data and the key feature subset, the core material data affecting inventory turnover is identified.

9. A material allocation plan generation system, characterized in that, include: The power material data acquisition module is used to acquire power material data corresponding to the material business characteristic tags input by the user from the data platform. The calculation module is used to calculate the inventory age of each type of power material in the power material data based on a pre-built inventory age analysis model, to obtain the inventory age of each type of power material and the risk probability value of becoming stockpiled material. The pre-construction process of the inventory age analysis model includes: determining a material business characteristic tag system for each type of power material based on historical power material data; constructing an inventory age management classification tree for the material business characteristic tags corresponding to the material business characteristic tag system for each type of power material; determining the influencing factors for each type of power material in batch management, inventory transfer, and inventory consumption processes based on the inventory age management classification tree; and constructing the inventory age analysis model based on the material business characteristic tags of each type of power material, the influencing factors, and the inventory turnover optimization model. The inventory turnover optimization model is pre-constructed based on the historical power material data and the inventory age management strategy. The optimal allocation scheme solution module is used to solve for the optimal allocation scheme of each type of power material based on the power material data, the inventory age of each type of power material and the risk probability value of becoming stockpiled materials, with the objectives of minimizing warehousing and logistics costs and minimizing emergency response time.

10. An electronic device, characterized in that, include: At least one processor and memory; The memory and processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, the material allocation scheme generation method as described in any one of claims 1-8 is implemented.

Images