Material demand analysis and prediction method, device and storage medium based on big data

Abstract

The present invention provides a material demand analysis and prediction method, device and storage medium based on big data. By obtaining a multi-source historical material data set of the supply chain link where the target enterprise is located, the multi-source historical material data set is integrated across links to obtain a cross-link integrated data set, and the cross-link integrated data set is subjected to demand feature extraction processing to obtain the dynamic characteristics of material demand and the material trend evolution characteristics of each material flow sequence. Based on a preset dynamic analysis model, the dynamic characteristics of material demand and the material trend evolution characteristics are subjected to joint trend matching processing to generate a predicted demand distribution result of the material flow sequence, generate a material supply optimization strategy based on the predicted demand distribution result, and feed the material supply optimization strategy back to the supply chain management system to trigger material allocation operations. The present invention can avoid the lag of manual decision-making and at the same time enhance the accuracy and anti-interference ability of strategy execution.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a material demand analysis and forecasting method, device and storage medium based on big data. Background Art

[0002] With the increasing complexity of supply chain management, material demand analysis and forecasting technology has become a critical component of enterprise resource planning. Existing technologies typically use historical data from supply chain links (such as warehouse inventory or purchase order records) for independent modeling and analysis, generating local demand forecasts through time series forecasting or regression algorithms. However, these forecasts fail to accurately reflect actual demand changes under the multi-link linkage effect due to ignoring data conflicts between different links (such as timestamp misalignment and inconsistent measurement units) and the transmission effects of material flow status. Furthermore, traditional methods, often based on methods such as periodic mean analysis, lack the ability to capture sudden demand fluctuations and the dynamic correlation between node loads in the supply chain, making it difficult for supply strategies to adapt to real-time supply and demand imbalances. Furthermore, the output of existing forecasting models often remains at the demand numerical level, lacking deep coupling with the multi-node resource allocation mechanism of the supply chain system. This creates a gap in the translation of forecast results into execution strategies, further amplifying the market response risks caused by strategy lags. Summary of the Invention

[0003] In view of this, the present invention provides a material demand analysis and forecasting method, device and storage medium based on big data.

[0004] The technical solution of the embodiment of the present invention is achieved as follows:

[0005] In one aspect, an embodiment of the present invention provides a material demand analysis and forecasting method based on big data, comprising the following steps:

[0006] Acquire a multi-source historical material data set for the supply chain link where the target enterprise is located, wherein the multi-source historical material data set includes multiple material flow sequences, each material flow sequence consisting of at least one material demand event and a corresponding material supply event;

[0007] Performing cross-link integration processing on the multi-source historical material data set to obtain a cross-link integrated data set, wherein the cross-link integrated data set is used to reflect the flow status changes of materials between different supply chain links;

[0008] Performing demand feature extraction processing on the cross-link integrated data set to obtain the material demand dynamic characteristics and material trend evolution characteristics of each material flow sequence;

[0009] Based on a preset dynamic analysis model, a joint trend matching process is performed on the dynamic characteristics of the material demand and the material trend evolution characteristics to generate a predicted demand distribution result of the material flow sequence;

[0010] A material supply optimization strategy is generated according to the predicted demand distribution result, and the material supply optimization strategy is fed back to the supply chain management system to trigger a material allocation operation.

[0011] In a second aspect, the present invention provides a material demand analysis and forecasting device, comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and the processor implements the steps in the above-described method when executing the program.

[0012] In a second aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above-described method when executed by a processor.

[0013] The material demand analysis and prediction method based on big data provided by the present invention obtains the full-link historical material data of multiple links in the supply chain and performs cross-link conflict analysis and data alignment to construct an integrated data set reflecting the real flow status, effectively solving the prediction deviation problem caused by the fragmentation of simple link data in traditional methods; by synchronously extracting the dynamic characteristics of material demand that characterize the real-time fluctuation state and the material trend evolution characteristics of the cross-link transmission effect from the integrated data, and combining them with the dynamic analysis model for joint matching, it realizes multi-dimensional modeling of the coupling effect of demand mutation and supply bottleneck, so that the prediction results can simultaneously capture the combined impact of local supply and demand imbalance and global transmission delay, significantly improving the accuracy of demand distribution prediction in complex supply chain scenarios; further based on the prediction results, an optimization strategy including the dynamic weights of inventory allocation and external procurement is generated and fed back to the supply chain system in real time, forming a closed-loop response mechanism that links demand fluctuation perception with resource allocation, while avoiding the lag of manual decision-making, enhancing the accuracy and anti-interference ability of strategy execution, and ensuring the stable operation of the material supply system in a changing market environment.

[0014] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and are not intended to limit the technical solutions of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of the implementation flow of a material demand analysis and forecasting method based on big data provided in an embodiment of the present invention.

[0016] Figure 2 A hardware entity diagram of a material demand analysis and forecasting device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention are further elaborated in detail below with reference to the accompanying drawings and embodiments. The described embodiments should not be regarded as limiting the present invention. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0018] An embodiment of the present invention provides a material demand analysis and forecasting method based on big data, which can be executed by a processor of a material demand analysis and forecasting device. The material demand analysis and forecasting device can be a device with data processing capabilities, such as a server, laptop, tablet computer, or desktop computer.

[0019] Figure 1 A schematic diagram of the implementation flow of a material demand analysis and forecasting method based on big data provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the method includes the following steps:

[0020] Step S100: Acquire a multi-source historical material data set of the supply chain link where the target enterprise is located. The multi-source historical material data set includes multiple material flow sequences, and each material flow sequence consists of at least one material demand event and a corresponding material supply event.

[0021] The target enterprise's supply chain segment refers to its specific position and functions within the entire supply chain system. The supply chain encompasses a series of processes, from raw material procurement, manufacturing, product distribution, and ultimately, the end customer. The target enterprise may be a raw material supplier, manufacturer, distributor, or retailer. A multi-source historical material data set is a collection of historical material-related data collected from multiple data sources. These data sources can include an enterprise's internal inventory management system, production record system, sales order system, as well as external supplier data and market research data. A material flow sequence is a series of data records describing the flow of materials within the supply chain, capturing the entire process from material demand to supply fulfillment. A material demand event refers to a demand for a material within the supply chain, such as a production plan requiring a certain raw material or a customer ordering a corresponding product. A material supply event is the supply response to a material demand event, such as a supplier shipment or a transfer from an internal warehouse.

[0022] For internal data sources, data interfaces can be used to extract relevant data from inventory management systems, production record systems, and sales order systems. For example, through the data interface of the enterprise resource planning (ERP) system, inventory change data, production material collection data, and sales order data can be extracted according to preset time ranges and data field requirements. For external data sources, data sharing agreements can be established with suppliers to obtain supplier supply records and inventory data through secure data transmission channels. Alternatively, web crawler technology can be used to obtain relevant market data from public market research websites and industry report databases. When acquiring data, it is necessary to clean and preprocess it to remove duplicate, erroneous, and incomplete data to ensure data quality. For example, if duplicate records or missing data are present in inventory data recorded in the inventory management system, deduplication and supplementation are necessary. Ultimately, the processed data is integrated into a single dataset, forming a multi-source historical material data set containing multiple material flow sequences, each of which clearly records material demand events and corresponding material supply events.

[0023] Step S200: performing cross-link integration processing on multi-source historical material data sets to obtain a cross-link integrated data set, which is used to reflect the changes in the flow status of materials between different supply chain links.

[0024] Cross-link integration processing involves comprehensively processing data from multiple historical material data sets across different supply chain links, eliminating discrepancies and conflicts, and ensuring that the data accurately reflects the flow of materials throughout the supply chain. This integrated data set contains detailed information on the flow of materials between different supply chain links and can reflect changes in material flow status, such as flow direction, flow speed, and changes in inventory levels.

[0025] As an implementation method, step S200 performs cross-link integration processing on a multi-source historical material data set to obtain a cross-link integrated data set, which may specifically include the following steps S210 to S240:

[0026] Step S210: obtaining the original data source identifier of each material flow sequence in the multi-source historical material data set, and determining the supply chain link type corresponding to the material flow sequence according to the original data source identifier.

[0027] The original data source identifier is used to identify the original source of each material flow sequence data. This can be the name, number, address, etc. of the data source. This identifier allows the data to be traced back to its source. Supply chain link types refer to the different functional links in the supply chain, such as raw material supply, manufacturing, product distribution, and retail.

[0028] Obtaining the original data source identifier can be achieved by adding an identification field to each data record during the data collection process. For example, when collecting data from different systems, add a "data source identifier" field to each data record to record which system the data comes from. After obtaining the original data source identifier, the supply chain link type corresponding to each material flow sequence is determined based on the pre-defined mapping relationship between the data source and the supply chain link type. For example, it is pre-defined that the data of the inventory management system corresponds to the production and manufacturing link, and the data of the sales order system corresponds to the product distribution link, etc. If the original data source identifier of a material flow sequence shows that it comes from the inventory management system, then it can be determined that the supply chain link type corresponding to the material flow sequence is the production and manufacturing link.

[0029] Step S220: for each supply chain link type, extract abnormal data segments in the material flow sequence, where the abnormal data segments include conflicting data segments where there is a timing conflict between the material demand event and the material supply event.

[0030] Abnormal data segments are data that do not conform to normal logic or patterns in the material flow sequence. Timing conflicts occur when the time sequence of material demand events and material supply events is unreasonable. For example, a material supply event occurs before a material demand event, or the time interval between material supply events does not conform to the normal production or supply cycle. A conflicting data segment is a specific data range that contains a timing conflict.

[0031] Time series analysis can be used to extract anomalous data segments. First, the material demand events and material supply events in each material flow sequence are sorted chronologically. Next, the temporal relationship between adjacent material demand and material supply events is examined. For example, if a material supply event is found to be earlier than its corresponding material demand event, or if the time interval between material supply events is too short or too long, exceeding the normal fluctuation range, this data segment is marked as an anomalous data segment. In specific implementations, a time interval threshold can be set. When the time interval between adjacent events is less than or greater than the threshold, a timing conflict is determined.

[0032] Step S230: performing conflict resolution processing on the abnormal data segment to generate a conflict resolution result, and performing data correction on the abnormal data segment according to the conflict resolution result to obtain a corrected material flow sequence.

[0033] Conflict resolution analyzes and explains timing conflicts within anomalous data segments to identify the causes of the conflicts. Conflict resolution results provide information on the causes and solutions. Data correction adjusts the data within the anomalous data segments based on the conflict resolution results to align with normal logic and patterns.

[0034] Conflict resolution for anomalous data segments can be performed using a combination of rule-based reasoning and machine learning. For example, conflict resolution rules can be defined such that if a material supply event occurs before a material demand event, this could be due to a data entry error or premature stocking. These rules can then be used to conduct a preliminary analysis of the anomalous data segment. Simultaneously, a machine learning algorithm can be used to learn from historical data to build a predictive model for the cause of the conflict. For example, a decision tree algorithm can be used to train the model using the features of the anomalous data segment (such as time interval and event type) as input and the cause of the conflict as output. During actual resolution, the features of the anomalous data segment are input into the trained model to predict the cause of the conflict. Data correction is then performed based on the conflict resolution results. If the cause is a data entry error, the data is corrected by manually verifying the original records. If the cause is premature stocking, the time sequence and quantity of the demand and supply events are adjusted based on the actual situation. For example, the supply event for premature stocking can be moved after the demand event, and the supply quantity can be adjusted accordingly. After data correction, a revised material flow sequence is obtained.

[0035] Step S240: Align the corrected material flow sequence with other material flow sequences that do not contain abnormal data fragments to generate an aligned cross-link data set, and use the aligned cross-link data set as a cross-link integrated data set; wherein, data alignment includes identifying and associating material flow sequences of different supply chain links according to material identification, and unifying the timestamp format and measurement unit in the material flow sequence.

[0036] Data alignment refers to organizing and matching different material flow sequences so that they remain consistent in certain key information to facilitate subsequent analysis and processing. Material identification is information used to uniquely identify each material, such as material number, name, etc. Material identification can be used to link flow sequences involving the same material in different supply chain links. The timestamp format is a format for recording the time when an event occurs. Different data sources may use different timestamp formats. A unified timestamp format can facilitate the analysis of the chronological order of material flows. The unit of measurement is the unit for measuring the quantity, weight, volume, etc. of materials. Different links may use different units of measurement. A unified unit of measurement can ensure the accuracy and comparability of the data.

[0037] The corrected material flow sequence is aligned with other material flow sequences that do not contain the anomalous data fragments. First, all material flow sequences are grouped according to material identifiers, with sequences involving the same material grouped together. Then, the timestamp format within each group of sequences is standardized. Data conversion functions can be used to convert timestamps of different formats to a standardized format. For example, formats such as "YYYY-MM-DD HH:MM:SS" and "DD / MM / YYYY HH:MM" can be converted to "YYYY-MM-DD HH:MM:SS." Units of measurement can be standardized using predefined unit conversion rules. For example, kilograms and grams can be standardized to kilograms, and liters and milliliters can be standardized to liters. After completing identifier association, timestamp format standardization, and unit of measurement standardization, an aligned cross-link data set is generated. This cross-link integrated data set accurately reflects the changes in material flow status across different supply chain links.

[0038] Step S300: extracting and processing the demand characteristics of the cross-link integrated data set to obtain the dynamic characteristics of material demand and the material trend evolution characteristics of each material flow sequence.

[0039] Demand feature extraction is the process of extracting key information reflecting the characteristics of material demand from a cross-link integrated data set. Material demand dynamics describe the uneven distribution of material demand over time, reflecting how material demand changes over time, such as fluctuations and sudden changes. Material trend evolution reflects the delayed transmission effects and cumulative impact weights of material demand across different supply chain links, revealing the propagation patterns and changing trends of material demand within the supply chain.

[0040] As an embodiment, step S300 performs demand feature extraction processing on the cross-link integrated data set to obtain the material demand dynamic features and material trend evolution features of each material flow sequence. Specifically, the following steps S310 to S340 may be included:

[0041] Step S310: extracting the time series fluctuation data of each material flow sequence from the cross-link integrated data set, where the time series fluctuation data includes the periodic variation amplitude of material demand and the material supply response delay duration.

[0042] Time series fluctuation data describes the fluctuations in material flow over time. The cyclical fluctuation range of material demand refers to the range of changes in material demand over a certain period, reflecting the cyclical fluctuations in material demand. Material supply response delay refers to the time interval between the occurrence of a material demand event and the start of the corresponding material supply event, reflecting the supply chain's responsiveness to material demand.

[0043] Extracting time-series fluctuation data from a cross-link integrated data set allows for statistical analysis of material demand and material supply response time within each sequence, following the chronological order of the material flow sequence. To determine the cyclical variation in material demand, first determine an appropriate cycle length, such as a monthly, quarterly, or annual cycle. Then, calculate the maximum and minimum material demand within each cycle; the difference between the two represents the variation within that cycle. For the material supply response delay, compare the timestamps of each material demand event and the corresponding material supply event, calculating the time difference between the two as the supply response delay for that demand-supply pair. For example, for a material flow sequence, count the material demand for each month, calculate the maximum and minimum values ​​for each month, and determine the monthly cyclical variation. Simultaneously, record the time of each demand event and the corresponding supply event to calculate the supply response delay.

[0044] Step S320: Perform pattern recognition processing on the time series fluctuation data to determine the demand mutation nodes and supply bottleneck nodes in the material flow sequence.

[0045] Pattern recognition processing is the process of analyzing and mining time-series fluctuating data to identify hidden patterns and regularities. A demand mutation node refers to a point in the material flow sequence where material demand suddenly and significantly changes. This can be caused by factors such as a sudden increase or decrease in market demand or the launch of a new product. A supply bottleneck node refers to a point in the supply chain where material supply is delayed or restricted due to certain factors. This can be caused by factors such as insufficient supplier production capacity or transportation issues.

[0046] As an embodiment, step S320 performs pattern recognition processing on the time series fluctuation data to determine the demand mutation node and supply bottleneck node in the material flow sequence, which may specifically include the following steps S321 to S326:

[0047] Step S321: Based on the periodic variation amplitude of the material demand in the time series fluctuation data, the material flow sequence is divided into multiple fluctuation data segments, each fluctuation data segment corresponds to a periodic variation amplitude interval.

[0048] Fluctuation data segments are data segments created by dividing the material flow sequence according to the cyclical fluctuations in material demand. Cyclic fluctuation intervals are predefined ranges within which the cyclical fluctuations in material demand within each fluctuation data segment fall.

[0049] Cluster analysis can be used to segment material flow sequences into multiple fluctuating data segments based on the cyclical fluctuations in material demand. First, determine the intervals for the cyclical fluctuations, such as 0-10%, 10%-20%, and 20%-30%. Then, statistically analyze the cyclical fluctuations within each material flow sequence and segment the sequence into different fluctuating data segments based on the intervals to which they belong. For example, for a material flow sequence, calculate the cyclical fluctuations for each cycle and assign data with a fluctuation range of 0-10% to one fluctuating data segment, data with a fluctuation range of 10%-20% to another, and so on.

[0050] Step S322: Calculate the difference between adjacent time periods for the material demand in each fluctuation data segment to generate a difference value sequence, and determine the candidate mutation position based on the continuously increasing or decreasing difference direction change in the difference value sequence.

[0051] Adjacent period variance calculation calculates the difference in material demand between two adjacent periods. This calculation can reflect the changing trend of material demand. A variance sequence is a sequence of variances calculated from adjacent period variance calculations. Candidate mutation locations are locations in the variance sequence where the direction of the variance changes from increasing to decreasing, potentially indicating a demand mutation.

[0052] For each fluctuation data segment, the material demand quantity difference between adjacent time periods is calculated, sequentially calculating the difference between the material demand quantities in two adjacent time periods in chronological order. For example, for a fluctuation data segment with a material demand quantity sequence of [100, 120, 130, 110, 90], the difference values ​​between adjacent time periods are calculated to be [20, 10, -20, -20]. Next, the difference value sequence is examined for changes in direction, either increasing or decreasing. If the difference value changes from increasing to decreasing or from decreasing to increasing, the position is marked as a candidate mutation location. In the above example, the position where the difference value changes from 10 to -20 is a candidate mutation location.

[0053] Step S323: extract the material demand change rate within the time series window corresponding to the candidate mutation position. If the change rate exceeds the average change rate threshold of the fluctuation data segment, the candidate mutation position is marked as a demand mutation node.

[0054] The time window is a time range set around a candidate mutation location and is used to analyze changes in material demand near that location. The material demand change rate refers to the percentage change in material demand within the time window, reflecting the rate of change. The average change rate threshold is a critical value set based on the average change in material demand within the fluctuation data segment and is used to determine whether a change in material demand is a mutation.

[0055] To extract the rate of change of material demand within the time series window corresponding to the candidate mutation location, first determine the size of the time series window. For example, centered around the candidate mutation location, select three time periods before and after as the time series window. Then, calculate the initial and final values ​​of the material demand within the time series window. Subtract the initial value from the final value and divide it by the initial value to obtain the rate of change of material demand. For example, within a time series window, if the material demand changes from 100 to 150, the rate of change is 50%. Simultaneously, calculate the average rate of change of material demand within the corresponding fluctuation data segment and use this as the average rate of change threshold. If the rate of change of material demand corresponding to the candidate mutation location exceeds the average rate of change threshold, mark the candidate mutation location as a demand mutation node.

[0056] Step S324: for the material supply response delay duration in the time series fluctuation data, identify the abnormal delay period in which the delay duration exceeds the preset delay benchmark, and extract the supply event identifier associated with the abnormal delay period.

[0057] The preset delay benchmark is a pre-set time threshold used to determine whether a material supply response delay is abnormal. The abnormal delay period is the time period when the material supply response delay exceeds the preset delay benchmark. The supply event identifier uniquely identifies each material supply event. By extracting the supply event identifiers associated with the abnormal delay period, we can further analyze the specific supply event that caused the supply delay.

[0058] Abnormal delay periods are identified based on the length of material supply response delays. Each material supply response delay is compared to a preset delay benchmark. If the delay exceeds the preset delay benchmark, the time period corresponding to the delay is marked as an abnormal delay period. For example, if the preset delay benchmark is 7 days and a material supply response delay is 10 days, then the time period corresponding to this 10-day delay is an abnormal delay period. Supply event identifiers associated with the abnormal delay period are then extracted. These identifiers can be obtained from the cross-link integrated data set and record the specific supply events that occurred during the abnormal delay period.

[0059] Step S325: Obtain the material supply path node information corresponding to the supply event identifier, and based on the supply interruption reason type and path node load status recorded in the node information, filter out the target path node where supply delay occurs due to excessive node load, and mark the target path node as a supply bottleneck node.

[0060] Material supply path node information provides detailed information about each node in the material supply process, including its name, location, function, and load status. The supply interruption cause type categorizes the cause of a material supply interruption or delay, such as insufficient supplier capacity, transportation failure, or raw material shortage. Path node load status indicates the workload of each supply path node, including equipment utilization and personnel workload. The target path node is the node where supply delays occur due to excessive node load.

[0061] Obtaining the material supply path node information corresponding to the supply event identifier can be achieved by establishing an association relationship between the supply event identifier and the material supply path node information in the cross-link integrated data set. For example, an association field is added to each supply event to record the supply path node information corresponding to the event. Then, based on the supply interruption cause type and path node load status recorded in the node information, the target path nodes whose supply is delayed due to excessive node load are screened out. If the load status of a node shows that its equipment utilization rate exceeds 90%, and the supply interruption cause type is "insufficient production capacity", it can be determined that the node is a target path node whose supply is delayed due to excessive node load, and it is marked as a supply bottleneck node.

[0062] Step S326: Perform spatiotemporal correlation analysis on the demand mutation node and the supply bottleneck node to determine the causal correlation strength between the two in the material flow sequence. If the causal correlation strength reaches the preset correlation threshold, configure a supply bottleneck impact identifier for the demand mutation node.

[0063] Spatiotemporal correlation analysis analyzes the temporal and spatial relationships between demand mutation nodes and supply bottleneck nodes to determine the causal relationship between them. Causal correlation strength measures the closeness of the causal relationship between demand mutation nodes and supply bottleneck nodes. The preset correlation threshold is a pre-set critical value used to determine whether the causal correlation strength is sufficient. The supply bottleneck impact indicator indicates whether a demand mutation node is affected by a supply bottleneck.

[0064] To analyze the spatiotemporal correlations between demand mutation nodes and supply bottleneck nodes, association rule mining can be used. First, the time and location information of the demand mutation nodes and supply bottleneck nodes are matched to identify pairs of nodes that are close in time and space. The causal strength of the association between each node pair is then calculated. For example, conditional probability can be used to calculate the probability of the demand mutation node occurring given the supply bottleneck node, using this probability as the causal strength. If the causal strength reaches a preset correlation threshold, a supply bottleneck impact flag is assigned to the demand mutation node.

[0065] Step S330: Generate dynamic characteristics of material demand based on demand mutation nodes and supply bottleneck nodes. The dynamic characteristics of material demand are used to characterize the non-balanced distribution state of material demand in the time dimension.

[0066] Material demand dynamics describe the uneven distribution of material demand over time. They reflect how material demand changes over time, such as fluctuations and sudden changes. Generating material demand dynamics based on demand mutation nodes and supply bottleneck nodes comprehensively considers the location and magnitude of demand mutation nodes and the impact of supply bottleneck nodes.

[0067] The specific generation process is as follows. First, a quantitative analysis of the demand mutation nodes is performed, recording information such as the occurrence time of each demand mutation node and the change in material demand. For example, if the demand mutation node occurs on the 10th day, the material demand increases from 100 to 200, a change of 100%. Then, the impact of the supply bottleneck node on the demand mutation node is considered. If a demand mutation node is configured with a supply bottleneck impact flag, it means that the demand mutation may be restricted by the supply bottleneck, and this impact needs to be taken into account when generating the dynamic characteristics of material demand. The degree of impact of the supply bottleneck can be reflected by setting different weights. For example, for demand mutation nodes configured with a supply bottleneck impact flag, the change is multiplied by a weight coefficient less than 1 to reflect the inhibitory effect of the supply bottleneck on demand realization. Finally, the information of all demand mutation nodes is integrated to form the dynamic characteristics of material demand, which can comprehensively reflect the uneven distribution of material demand in the time dimension.

[0068] Step S340: Extract the flow rate difference data of the material flow sequence between different supply chain links, and perform trend fitting processing on the flow rate difference data to generate material trend evolution characteristics; the material trend evolution characteristics are used to reflect the transmission delay effect and cumulative impact weight of material demand between different supply chain links.

[0069] Turnover rate variance data describes the differences in material flow speeds between different supply chain links. It reflects the speed of material flow from one link to another in the supply chain. Trend fitting is the process of analyzing and modeling this turnover rate variance data to identify trends and patterns in the data. Material trend evolution characteristics reflect the transmission delay effect and cumulative impact weight of material demand between different supply chain links. They can reveal the propagation patterns and changing trends of material demand within the supply chain.

[0070] As an embodiment, step S340 extracts the flow rate difference data between different supply chain links of the material flow sequence, and performs trend fitting processing on the flow rate difference data to generate material trend evolution characteristics. Specifically, the following steps S341 to S346 may be included:

[0071] Step S341: Based on the periodic variation amplitude of material demand and the material supply response delay duration in the time series fluctuation data, the rate difference between the input material rate and the output material rate of each supply chain link is calculated to obtain the initial turnover rate of each supply chain link.

[0072] The input material rate refers to the rate at which materials enter a supply chain link, typically expressed as the number of materials entering the link per unit time. The output material rate refers to the rate at which materials exit a supply chain link, also expressed as the number of materials leaving the link per unit time. The initial turnover rate is calculated by calculating the difference between the input material rate and the output material rate, and it reflects the net flow rate of materials within that supply chain link.

[0073] The initial flow rate for each supply chain link is calculated based on the cyclical variation in material demand and the material supply response delay. First, the cyclical variation in material demand within each cycle is determined based on the cyclical variation in material demand. For example, if the cyclical variation in material demand within a cycle is 20% and the initial demand is 100, then the change in demand within that cycle is 20. Then, the time points at which materials enter and exit that link are determined based on the material supply response delay. For example, if the material supply response delay is 3 days, then materials will not enter the link until 3 days after the demand event occurs. Based on this information, the input and output material rates for each cycle are calculated. For example, if the input material quantity is 80 and the output material quantity is 60 over a 7-day period, then the input material rate is 80 / 7 and the output material rate is 60 / 7, resulting in an initial flow rate of (80 / 7) - (60 / 7) = 20 / 7.

[0074] Step S342: Compare the initial turnover rates of adjacent supply chain links by comparing their rate differences, generate turnover rate difference data between adjacent links, and record the supply chain link connection identifiers corresponding to the turnover rate difference data.

[0075] Adjacent supply chain links refer to two directly adjacent links in the supply chain, such as the raw material supply link and the manufacturing link, or the manufacturing link and the product distribution link. Turnover rate difference data is the difference between the initial turnover rates of adjacent supply chain links, reflecting the change in the material flow rate between adjacent links. The supply chain link connection identifier uniquely identifies the connection between adjacent supply chain links. By recording this identifier, the specific link connection corresponding to the turnover rate difference data can be clearly identified.

[0076] Compare the initial turnover rates of adjacent supply chain links and calculate the difference in their initial turnover rates. For example, if the initial turnover rate of a raw material supply link is 10, and the initial turnover rate of the adjacent manufacturing link is 8, then the turnover rate difference between the two links is 2. At the same time, record the corresponding supply chain link connection identifier, such as "Raw Material Supply - Manufacturing." By performing this comparison and recording for all adjacent supply chain links, a complete set of turnover rate difference data between adjacent links is generated.

[0077] Step S343: Perform fluctuation range analysis on the turnover rate difference data in the time dimension, extract abnormal fluctuation ranges whose fluctuation ranges exceed the preset fluctuation benchmark, and locate the corresponding supply chain link pairs according to the connection identifiers within the abnormal fluctuation ranges.

[0078] Fluctuation amplitude analysis analyzes the temporal changes in turnover rate variance data and calculates its fluctuation amplitude. The abnormal fluctuation interval is the period during which the fluctuation amplitude of the turnover rate variance data exceeds a preset fluctuation baseline. The preset fluctuation baseline is a pre-set fluctuation amplitude threshold used to determine whether a fluctuation is abnormal. A supply chain link pair consists of adjacent supply chain links corresponding to the connection identifiers within the abnormal fluctuation interval.

[0079] The moving window method can be used to analyze the fluctuation amplitude of the turnover rate difference data in the time dimension. First, set the size of a moving window, for example, 7 days as a window. Then, calculate the maximum and minimum values ​​of the turnover rate difference data in each window, and the difference between the two is the fluctuation amplitude in the window. Compare the fluctuation amplitude of each window with the preset fluctuation benchmark. If it exceeds the preset fluctuation benchmark, the time period corresponding to the window is marked as an abnormal fluctuation interval. According to the connection identifier in the abnormal fluctuation interval, locate the corresponding supply chain link pair. For example, if the connection identifier in the abnormal fluctuation interval is "manufacturing-product distribution", then the corresponding supply chain link pair is the manufacturing link and the product distribution link.

[0080] Step S344: Identify the supply-demand correlation between the material supply response delay time and the periodic change amplitude of the material demand in the supply chain link pair. If the supply-demand correlation is lower than the preset correlation benchmark, mark the supply chain link pair as a transmission abnormality path.

[0081] The supply-demand correlation is an indicator that measures the correlation between the delay in material supply response and the cyclical fluctuations in material demand. It reflects the degree of coordination between material supply and demand. The preset correlation benchmark is a pre-set critical value used to determine whether the supply-demand correlation is sufficiently high. An abnormal transmission path refers to a supply chain link pair where the supply-demand correlation falls below the preset correlation benchmark, indicating that there may be abnormalities in the transmission of material demand within this link pair.

[0082] Correlation analysis can be used to identify the supply-demand correlation between the material supply response delay and the cyclical variation in material demand within a supply chain link pair. For example, the Pearson correlation coefficient between the material supply response delay and the cyclical variation in material demand can be calculated and used as an indicator of supply-demand correlation. If this coefficient falls below a preset correlation benchmark, the supply chain link pair is marked as a transmission anomaly path. For example, if the preset correlation benchmark is 0.5 and the Pearson correlation coefficient between the material supply response delay and the cyclical variation in material demand within a supply chain link pair is 0.3, then the supply chain link pair is marked as a transmission anomaly path.

[0083] Step S345: performing multi-stage state tracking on the flow rate difference data in the conduction abnormality path to obtain a state change trajectory of the difference data within a preset time span.

[0084] Multi-stage state tracking continuously monitors and records the status of flow rate difference data at different stages within the conduction anomaly path. The preset time span is a pre-defined time range that determines the observation period for the state change trajectory. The state change trajectory records the changes in flow rate difference data within the preset time span, which can reflect the trends and patterns of data changes.

[0085] Time series analysis can be used to track the multi-stage state of the flow rate difference data in the conduction abnormality path. First, the preset time span is divided into multiple stages in chronological order, for example, divided into months. Then, the state of the flow rate difference data, such as the average value, maximum value, minimum value, etc., is recorded in each stage. By continuously tracking and recording these state data, the state change trajectory of the flow rate difference data within the preset time span is obtained. For example, within a preset time span of 12 months, the average value of the flow rate difference data in the conduction abnormality path is recorded each month to form a state change trajectory containing 12 data points.

[0086] Step S346: Perform segmented slope fitting based on the state change trajectory to generate a trend conduction curve corresponding to the conduction abnormality path, and smooth the slope mutation points in the trend conduction curve to generate material trend evolution characteristics; wherein, the material trend evolution characteristics include a conduction path stability index and a trend diffusion direction vector. The conduction path stability index is used to quantify the ability to resist supply and demand fluctuations between supply chain links, and the trend diffusion direction vector is used to indicate the conduction priority order of material demand changes in the supply chain.

[0087] Segmented slope fitting is the process of analyzing and modeling the trajectory of state changes, dividing it into multiple segments, and calculating the slope of each segment. The trend transmission curve is a curve generated by segmented slope fitting. It reflects the changing trend of the flow rate difference data in the transmission anomaly path. Slope mutation points are points in the trend transmission curve where the slope suddenly changes significantly. Smoothing is the process of adjusting these mutation points to make the curve smoother and more continuous. The transmission path stability index is an indicator used to quantify the ability of supply and demand fluctuations between supply chain links to resist interference. It can be measured by the stability of the trend transmission curve. The trend diffusion direction vector is a vector used to indicate the transmission priority order of material demand changes in the supply chain. It can be determined by the direction and size of the slope of the trend transmission curve.

[0088] Segmented slope fitting is performed based on the state change trajectory. First, the state change trajectory is divided into multiple segments. Cluster analysis or breakpoint detection methods can be used to determine the segment division points. Then, the slope of each segment is calculated, and a trend transmission curve is plotted based on the slope. Moving average or spline interpolation methods can be used to smooth the slope mutation points in the trend transmission curve. For example, using the moving average method, several data points around the slope mutation point are averaged and the average value is used to replace the mutation point value, making the curve smoother. The resulting material trend evolution characteristics include a transmission path stability index and a trend diffusion direction vector. The transmission path stability index can be obtained by calculating the standard deviation or variance of the trend transmission curve. A smaller standard deviation or variance indicates a more stable transmission path and greater resilience to supply and demand fluctuations between supply chain links. The trend diffusion direction vector can be determined based on the direction and magnitude of the slope of the trend transmission curve. A positive slope indicates an upward trend in material demand change, and a larger slope indicates a higher transmission priority. A negative slope indicates a downward trend in material demand change, and a larger absolute value of the slope indicates a lower transmission priority.

[0089] Step S400: Based on a preset dynamic analysis model, a joint trend matching process is performed on the dynamic characteristics of material demand and the material trend evolution characteristics to generate a predicted demand distribution result of the material flow sequence.

[0090] The preset dynamic analysis model is a pre-established model used to analyze material demand trends. It can be based on machine learning algorithms, such as neural network models or support vector machine models. Joint trend matching is the process of matching and analyzing the dynamic characteristics of material demand and the evolution of material trends with the preset dynamic analysis model to identify the correlations and patterns between them. The predicted demand distribution, derived from the joint trend matching process, predicts the future demand distribution of the material flow sequence. It includes information such as the material demand priority sequence for each supply chain link within a specified future time period.

[0091] As an embodiment, step S400 performs a joint trend matching process on the dynamic characteristics of material demand and the material trend evolution characteristics based on a preset dynamic analysis model to generate a predicted demand distribution result of the material flow sequence. Specifically, the steps S410 to S450 may be included:

[0092] Step S410: Input the dynamic characteristics of material demand into the demand quantification module of the dynamic analysis model to generate a set of demand quantification indicators. The set of demand quantification indicators includes a demand peak forecast value and a demand valley fluctuation coefficient.

[0093] The demand quantification module is used within the dynamic analysis model to quantitatively analyze material demand. It converts the dynamic characteristics of material demand into specific quantitative indicators. The demand quantification indicator set is a collection of quantitative indicators generated by the module. The demand peak forecast predicts the maximum potential material demand will reach in the future, while the demand valley fluctuation coefficient measures the degree of fluctuation in material valley demand.

[0094] The dynamic characteristics of material demand are input into the demand quantification module of the dynamic analysis model. The demand quantification module can perform quantitative analysis using regression analysis. For example, a linear regression model is used to train the model, taking the amplitude and time of the demand mutation nodes in the dynamic characteristics of material demand as input variables and historical demand peak and valley data as output variables. When new dynamic characteristics of material demand are input, the model makes predictions based on the trained parameters, generating a demand peak forecast value and a demand valley fluctuation coefficient. Assume that the trained linear regression model is y=a*x1+b*x2+c, where x1 and x2 are the amplitude and time of the demand mutation nodes in the dynamic characteristics of material demand, and a, b, and c are the model parameters. After inputting the new dynamic characteristics of material demand, the calculated y value is the demand peak forecast value, and the demand valley fluctuation coefficient is calculated based on the model's error analysis.

[0095] Step S420: Inputting the material trend evolution characteristics into the trend conduction module of the dynamic analysis model to generate a trend conduction indicator set, which includes a trend diffusion rate and a trend attenuation threshold.

[0096] The trend transmission module, part of the dynamic analysis model, analyzes the transmission of material demand trends. It converts material trend evolution characteristics into specific trend transmission indicators. The trend transmission indicator set is a collection of indicators generated by the module. The trend diffusion rate refers to the speed at which changes in material demand propagate through the supply chain, and the trend attenuation threshold is the critical value at which changes in material demand begin to decay during the transmission process.

[0097] The material trend evolution characteristics are input into the trend transmission module of the dynamic analysis model. This module can perform analysis using time series analysis. For example, the Autoregressive Integrated Moving Average (ARIMA) model uses the state change trajectory data in the material trend evolution characteristics as input to predict the trend of material demand changes. The trend diffusion rate and trend decay threshold are calculated based on the predicted results. Assuming that the ARIMA model's forecast of material demand changes is a time series, the trend diffusion rate is calculated by calculating the rate of change between adjacent time points. The trend decay threshold is also determined based on the changing trend of the series.

[0098] Step S430: Calculate the demand transmission matching degree between different supply chain links of the material flow sequence based on the demand quantification indicator set and the trend transmission indicator set.

[0099] Demand transmission matching is an indicator to measure the coordination of material demand transmission between different supply chain links. It reflects the matching degree between demand quantitative indicators and trend transmission indicators in different links.

[0100] The demand transmission match degree can be calculated using a weighted summation method based on a set of quantitative demand indicators and a set of trend transmission indicators. First, determine the weights for the quantitative demand indicators and trend transmission indicators. For example, set the weight of the peak demand forecast value to 0.6, the weight of the valley demand fluctuation coefficient to 0.2, the weight of the trend diffusion rate to 0.1, and the weight of the trend attenuation threshold to 0.1. Then, for each supply chain link, perform a weighted sum of the quantitative demand indicators and trend transmission indicators for that link according to their weights to obtain the demand transmission score for that link. Finally, calculate the difference in the demand transmission scores between adjacent supply chain links, and use the absolute value of the difference as the demand transmission match degree. For example, if the demand transmission score of a supply chain link is 80 and the demand transmission score of the adjacent link is 70, then the demand transmission match degree between the two links is |80 - 70| = 10.

[0101] Step S440: If the demand transmission matching degree is lower than the preset matching threshold, the weights of the material demand dynamic characteristics and the material trend evolution characteristics are adjusted to generate an adjusted joint feature set.

[0102] The preset matching threshold is a critical value used to determine whether the demand transmission matching degree is high enough. Weight adjustment is the process of redistributing the weights of the material demand dynamic characteristics and material trend evolution characteristics in the joint analysis. The adjusted joint feature set is the combination of the material demand dynamic characteristics and material trend evolution characteristics after weight adjustment.

[0103] If the demand transmission matching degree falls below the preset matching threshold, it indicates that the transmission coordination of material demand across different supply chain links is poor, and the weighting of the dynamic characteristics of material demand and the material trend evolution characteristics needs to be adjusted. Optimization algorithms, such as genetic algorithms, can be used to determine the optimal weight distribution scheme. Genetic algorithms simulate the biological evolution process and continuously iterate to find the optimal solution. Using the demand transmission matching degree as the objective function and the weights of the dynamic characteristics of material demand and the material trend evolution characteristics as variables, the genetic algorithm continuously adjusts the weights through selection, crossover, and mutation operations until the demand transmission matching degree reaches the preset matching threshold or approaches the optimal value.

[0104] Step S450: Input the adjusted joint feature set into the distribution prediction module of the dynamic analysis model to generate a predicted demand distribution result, which includes the material demand priority sequence of each supply chain link in the future specified time period.

[0105] The distribution forecasting module is used within the dynamic analysis model to predict the distribution of material demand based on a set of input features. The predicted demand distribution results, processed by the module, are a forecast of the distribution of material demand across various supply chain links within a specified future timeframe. The material demand priority sequence is a ranking of supply chain links based on the importance or urgency of material demand.

[0106] The adjusted joint feature set is input into the distribution prediction module of the dynamic analysis model. This module can use classification algorithms in machine learning, such as decision tree algorithms, to perform predictions. The decision tree algorithm constructs a decision tree model to classify and predict material demand based on the input feature set. The adjusted joint feature set is used as input to the decision tree model. The model uses the trained rules to predict material demand for each supply chain link within a specified future time period, generating a predicted demand distribution. For example, based on the input features, the decision tree model may determine that a particular supply chain link will have high material demand in the next month and prioritize it accordingly.

[0107] Step S500: Generate a material supply optimization strategy based on the predicted demand distribution result, and feed the material supply optimization strategy back to the supply chain management system to trigger material allocation operations.

[0108] A material supply optimization strategy is a strategy developed based on forecasted demand distribution to optimize material supply. It includes decisions on aspects such as inventory transfer ratios and external procurement weights. A supply chain management system (SCM) is an information system used by enterprises to manage supply chain operations. It receives SCM strategies and triggers corresponding material allocation operations, such as inventory transfers and procurement.

[0109] As an embodiment, step S500 generates a material supply optimization strategy based on the predicted demand distribution result, and feeds the material supply optimization strategy back to the supply chain management system to trigger the material allocation operation. Specifically, the following steps S510 to S550 may be included:

[0110] Step S510: extracting the priority sequence identifier from the predicted demand distribution result, and determining the urgency level of material supply according to the priority sequence identifier.

[0111] The priority sequence identifier is used to identify the priority of material demand at each supply chain link in the forecast demand distribution results. It can be a number, an alphabetic code, etc. The material supply urgency classification is based on the priority sequence identifier to classify the urgency of material supply into three levels, such as high, medium, and low.

[0112] The priority sequence identifier in the forecast demand distribution result can be directly obtained from the data structure of the forecast demand distribution result. For example, the forecast demand distribution result is stored in a table, in which a column records the priority sequence identifier of each supply chain link. The urgency level of material supply is determined according to the priority sequence identifier, and the corresponding relationship between the priority sequence identifier and the urgency level can be pre-set. For example, the urgency level of the supply chain link with a priority sequence identifier of 1-3 is high, 4-6 is medium, and 7-10 is low. If the priority sequence identifier of a supply chain link is 2, then the urgency level of material supply in this link is high.

[0113] Step S520: Based on the urgency level, a corresponding supply resource quota is configured for each supply chain link. The supply resource quota includes the inventory allocation ratio and the external procurement weight.

[0114] The supply resource quota is the quantity and proportion of resources allocated to each supply chain link to meet material needs. The inventory allocation ratio refers to the proportion of materials allocated from the company's internal inventory, and the external procurement weight refers to the proportion of materials purchased from external suppliers.

[0115] Based on the urgency level, supply resource quotas are allocated to each supply chain link using pre-set allocation rules. For example, for a supply chain link with a high urgency level, the inventory transfer ratio is set at 70%, and the external procurement weighting is set at 30%. For a supply chain link with a medium urgency level, the inventory transfer ratio is set at 50%, and the external procurement weighting is set at 50%. For a supply chain link with a low urgency level, the inventory transfer ratio is set at 30%, and the external procurement weighting is set at 70%. Based on the urgency level of each supply chain link, supply resource quotas are allocated according to the above rules.

[0116] Step S530: Generate an initial supply strategy based on the inventory transfer ratio and the external procurement weight, and perform resource conflict detection on the initial supply strategy.

[0117] The initial supply strategy is a preliminary material supply plan based on the inventory transfer ratio and external procurement weight. It includes information such as the inventory transfer quantity and the external procurement quantity. Resource conflict detection checks the initial supply strategy to determine whether there are resource shortages or conflicts, such as insufficient inventory to meet transfer requirements or insufficient material supply from external suppliers.

[0118] An initial supply strategy is generated based on the inventory transfer ratio and external procurement weight. First, the total material demand for each supply chain link is calculated. For example, the total material demand for a particular supply chain link is 100 pieces. Next, the inventory transfer quantity and external procurement quantity are calculated based on the inventory transfer ratio and external procurement weight. If the inventory transfer ratio is 60% and the external procurement weight is 40%, the inventory transfer quantity is 100 * 60% = 60 pieces, and the external procurement quantity is 100 * 40% = 40 pieces. After the initial supply strategy is generated, resource conflict detection is performed.

[0119] As an implementation method, in step S530, resource conflict detection is performed on the initial provisioning strategy, which may specifically include the following steps S531 to S537:

[0120] Step S531: Acquire real-time inventory status data of the current supply chain link, where the real-time inventory status data includes available inventory and in-transit inventory.

[0121] Real-time inventory status data reflects the actual inventory situation in the current supply chain. Available inventory refers to the amount of inventory that can be immediately allocated, and in-transit inventory refers to the amount of inventory that has been ordered but has not yet arrived.

[0122] Obtaining real-time inventory status data for each supply chain link can be achieved through data interaction with the company's inventory management system. The inventory management system records the inventory status of each supply chain link in real time. By calling the inventory management system's data interface, the corresponding real-time inventory status data can be obtained according to the supply chain link identifier. For example, for a certain manufacturing link, the available inventory is 80 pieces and the in-transit inventory is 20 pieces.

[0123] Step S532: Calculate the target allocation quantity based on the inventory allocation ratio, and determine whether the target allocation quantity exceeds the sum of the available inventory and the in-transit inventory.

[0124] The target transfer quantity is the quantity of materials to be transferred from inventory, calculated based on the inventory transfer ratio. By determining whether the target transfer quantity exceeds the sum of available inventory and in-transit inventory, you can determine if there is an inventory shortage.

[0125] Calculate the target transfer quantity based on the inventory transfer ratio by multiplying the total material demand by the inventory transfer ratio. For example, if the total material demand is 100 pieces and the inventory transfer ratio is 60%, the target transfer quantity is 100 * 60% = 60 pieces. Compare the target transfer quantity with the sum of available inventory and in-transit inventory. If available inventory is 80 pieces and in-transit inventory is 20 pieces, for a total of 100 pieces, and the target transfer quantity of 60 pieces is less than the total, then inventory can meet the transfer demand. If the target transfer quantity is 120 pieces, which is greater than the total, then inventory is insufficient.

[0126] Step S533: If it exceeds, it is determined that a first type of resource conflict exists, and a first conflict identifier is generated.

[0127] A first-class resource conflict occurs when the amount of inventory transferred exceeds the sum of the available inventory and the inventory in transit. The first-class conflict indicator is used to indicate the existence of a first-class resource conflict. It can be a code or a mark.

[0128] If the target transfer quantity exceeds the sum of available inventory and in-transit inventory, a Type I resource conflict exists. For example, if the target transfer quantity is 120 pieces and the sum of available inventory and in-transit inventory is 100 pieces, a Type I resource conflict exists. A Type I conflict flag, such as "C1," is generated to indicate a Type I resource conflict.

[0129] Step S534: Acquire the real-time supply capacity data of the external supplier, where the real-time supply capacity data includes the maximum committed quantity and the minimum delivery cycle.

[0130] Real-time supply capacity data reflects the current actual supply capacity of external suppliers. The maximum committed quantity refers to the maximum quantity of materials that a supplier can provide within a certain period of time, and the minimum supply cycle refers to the shortest time required for a supplier to provide materials from the time an order is received.

[0131] Obtaining real-time supply capacity data from external suppliers can be achieved by connecting to their information systems. Suppliers' information systems record their supply capacity in real time, and data such as their maximum available quantity and minimum lead time can be obtained through secure data transmission channels. For example, after connecting to a particular supplier's data, the supplier's maximum available quantity and minimum lead time can be determined to be 50 units and 5 days.

[0132] Step S535: Calculate the target purchase quantity based on the external purchase weight, and determine whether the target purchase quantity exceeds the maximum committed quantity or whether the minimum delivery cycle is later than the demand time window.

[0133] The target purchase quantity is the quantity of materials to be purchased from external suppliers, calculated based on external procurement weights. The demand window is the timeframe within which material demand must be met. By determining whether the target purchase quantity exceeds the maximum committed quantity or the minimum lead time is later than the demand window, you can determine if there are insufficient external procurement resources or delivery delays.

[0134] Calculate the target purchase quantity based on the external procurement weight by multiplying the total material demand by the external procurement weight. For example, if the total material demand is 100 pieces and the external procurement weight is 40%, then the target purchase quantity is 100*40%=40 pieces. Compare the target purchase quantity with the maximum committable quantity and the minimum lead time with the demand time window. If the maximum committable quantity is 50 pieces, the target purchase quantity of 40 pieces is less than the maximum committable quantity, and the minimum lead time is 5 days, and the demand time window is 7 days, then external procurement can meet the demand. If the target purchase quantity is 60 pieces, which is greater than the maximum committable quantity, or the minimum lead time is 10 days, which is 7 days later than the demand time window, then there is a shortage of external procurement resources or a supply delay.

[0135] Step S536: If it is longer or later than, it is determined that there is a second type of resource conflict, and a second conflict identifier is generated.

[0136] The second type of resource conflict refers to a conflict in which the target purchase quantity exceeds the maximum committed quantity or the minimum delivery cycle is later than the demand time window. The second conflict indicator is used to mark the existence of the second type of resource conflict. It can be a code or a mark.

[0137] If the target purchase quantity exceeds the maximum promised quantity or the minimum lead time is later than the required time window, a Type II resource conflict is identified. For example, if the target purchase quantity is 60 units and the maximum promised quantity is 50 units, a Type II resource conflict exists. Alternatively, if the minimum lead time is 10 days and the required time window is 7 days, a Type II resource conflict also exists. A Type II conflict flag, such as "C2," is generated to indicate a Type II resource conflict.

[0138] Step S537: Determine the priority order of conflict resolution according to the type combination of the first conflict identifier and the second conflict identifier.

[0139] The priority order of conflict resolution is determined by the combination of the first conflict identifier and the second conflict identifier type to determine the order in which conflicts are handled. It can help enterprises rationally arrange resources and prioritize the resolution of important conflicts.

[0140] The priority order for conflict resolution is determined based on the combination of the first and second conflict identifiers. Priority rules can be pre-defined for different types of combinations. For example, if only the first type of resource conflict exists (only the first conflict identifier "C1"), the priority is to resolve insufficient inventory. If only the second type of resource conflict exists (only the second conflict identifier "C2"), the priority is to resolve insufficient externally procured resources or delayed delivery. If both the first and second types of resource conflicts exist (both "C1" and "C2"), the priority is to resolve insufficient inventory before resolving external procurement issues. The priority order for conflict resolution is determined based on the specific combination of conflict identifier types according to the above rules.

[0141] Step S540: If a resource conflict is detected in the initial supply strategy, the inventory allocation ratio and external procurement weight are dynamically adjusted based on the conflict type to generate a material supply optimization strategy after the conflict is resolved.

[0142] Dynamic adjustment refers to the process of real-time adjustments to inventory allocation ratios and external procurement weights based on the type and severity of resource conflicts. The optimized material supply strategy after conflict resolution is the one that resolves resource conflicts through dynamic adjustment.

[0143] As an implementation method, step S540 dynamically adjusts the inventory transfer ratio and external procurement weight based on the conflict type to generate a material supply optimization strategy after the conflict is resolved. Specifically, the following steps S541 to S545 may be included:

[0144] Step S541: If there is a first type of resource conflict, the inventory transfer ratio is reduced, and the target transfer quantity is recalculated based on the reduced inventory transfer ratio until the target transfer quantity does not exceed the sum of the available inventory and the in-transit inventory.

[0145] If a first-type resource conflict is detected, it means that the inventory allocation quantity exceeds the sum of the available inventory and the inventory in transit, and the inventory allocation ratio needs to be reduced. The inventory allocation ratio can be reduced gradually, for example, by 10% each time. After reducing the inventory allocation ratio, recalculate the target allocation quantity based on the new ratio. For example, the original inventory allocation ratio was 60%, the target allocation quantity was 120 pieces, and the sum of the available inventory and the inventory in transit was 100 pieces. There is a first-type resource conflict. Reduce the inventory allocation ratio to 50%, then the new target allocation quantity is 100*50%=50 pieces, which is less than the total and meets the requirements. Repeat this process until the target allocation quantity does not exceed the sum of the available inventory and the inventory in transit.

[0146] Step S542: If there is a second type of resource conflict, reduce the external procurement weight and recalculate the target procurement quantity based on the reduced external procurement weight until the target procurement quantity does not exceed the maximum committed quantity and the minimum delivery cycle is earlier than the demand time window.

[0147] If a second-type resource conflict is detected, it means that the target procurement quantity exceeds the maximum committed quantity or the minimum supply cycle is later than the demand time window, and the external procurement weight needs to be reduced. The external procurement weight can be reduced gradually, for example, by 10% each time. After reducing the external procurement weight, recalculate the target procurement quantity based on the new weight. For example, the original external procurement weight was 40%, the target procurement quantity was 60 pieces, and the maximum committed quantity was 50 pieces. There is a second-type resource conflict. Reduce the external procurement weight to 30%, then the new target procurement quantity is 100*30%=30 pieces, which is less than the maximum committed quantity and meets the requirements. At the same time, check whether the minimum supply cycle is earlier than the demand time window. If not, continue to reduce the external procurement weight until the target procurement quantity does not exceed the maximum committed quantity and the minimum supply cycle is earlier than the demand time window.

[0148] Step S543: If both the first and second resource conflicts exist, the inventory allocation ratio is reduced first. After the target allocation quantity corresponding to the reduced inventory allocation ratio stabilizes, the external procurement weight is reduced.

[0149] If both the first and second types of resource conflicts exist, the inventory allocation ratio shall be reduced first according to the priority order of conflict resolution. Reduce the inventory allocation ratio according to the method of step S541 until the target allocation quantity does not exceed the sum of the available inventory and the in-transit inventory. After the target allocation quantity stabilizes, reduce the external procurement weight according to the method of step S542 until the target procurement quantity does not exceed the maximum committed quantity and the minimum delivery cycle is earlier than the demand time window. For example, if both the first and second types of resource conflicts exist, first reduce the inventory allocation ratio from 60% to 50% so that the target allocation quantity meets the requirements. Then, reduce the external procurement weight from 40% to 30% so that the target procurement quantity and delivery cycle meet the requirements.

[0150] Step S544: regenerate the supply resource quota according to the adjusted inventory allocation ratio and external procurement weight, and update the initial supply strategy based on the regenerated supply resource quota.

[0151] Based on the adjusted inventory transfer ratio and external procurement weight, the inventory transfer quantity and external procurement quantity are recalculated to generate a new supply resource quota. For example, if the adjusted inventory transfer ratio is 50%, the external procurement weight is 30%, and the total material demand is 100 pieces, the inventory transfer quantity is 100 * 50% = 50 pieces, and the external procurement quantity is 100 * 30% = 30 pieces. Based on the regenerated supply resource quota, the initial supply strategy is updated, and the new inventory transfer quantity and external procurement quantity are updated to the initial supply strategy.

[0152] Step S545: Use the updated initial supply strategy as the material supply optimization strategy after conflict resolution.

[0153] The updated initial supply strategy is used as the material supply optimization strategy after conflict resolution. This strategy can solve the resource conflict problem existing in the initial supply strategy and achieve the optimization of material supply.

[0154] Step S550: Encode the inventory transfer instructions and purchase instructions in the material supply optimization strategy into executable operation codes, and transmit the executable operation codes to the strategy execution interface of the supply chain management system.

[0155] Stock transfer instructions are specific instructions for stock transfers within a material supply optimization strategy, including information such as the type, quantity, source, and destination of the material being transferred. Procurement instructions are specific instructions for external purchases, including information such as the type, quantity, and supplier of the material being purchased. Executable operation codes convert stock transfer and procurement instructions into computer-readable and executable code, enabling automated control of material allocation operations. The strategy execution interface is the interface within the supply chain management system used to receive and execute material supply optimization strategies.

[0156] As an implementation manner, the method provided in the embodiment of the present invention may further include the following steps S600 to S900:

[0157] Step S600: monitor the strategy execution status of the supply chain management system in real time, and obtain actual execution effect data of the material allocation operation.

[0158] Real-time monitoring is the process of continuously tracking and monitoring the execution of material supply optimization strategies within a supply chain management system. Strategy execution status refers to the progress and results of implementing material supply optimization strategies within the supply chain management system, such as whether inventory transfers have been completed and purchased goods have arrived. Actual execution performance data reflects the actual effects of material allocation operations, including actual demand fulfillment rates and actual supply delay durations.

[0159] Real-time monitoring of the policy execution status of the supply chain management system can be achieved through data exchange with the supply chain management system's log system. The supply chain management system's log system records the execution process and results of each material allocation operation. By regularly reading log data, policy execution status information can be obtained. For example, log data can be read every hour to check whether inventory allocation operations have been completed and whether purchase orders have been shipped. To obtain data on the actual execution results of material allocation operations, relevant information can be extracted from the supply chain management system's business data. For example, actual sales quantities can be obtained from the sales order system to calculate the actual demand fulfillment rate; material arrival times can be obtained from the logistics management system to calculate the actual supply delay.

[0160] Step S700: Perform deviation analysis on the actual execution effect data and the predicted demand distribution result to generate a deviation analysis report.

[0161] Deviation analysis is the process of comparing and analyzing actual performance data with predicted demand distribution results to identify discrepancies. A deviation analysis report summarizes and records the results of the deviation analysis, including information such as the deviation type and the level of impact.

[0162] As an implementation method, step S700 performs deviation analysis on the actual execution effect data and the predicted demand distribution result to generate a deviation analysis report, which may specifically include the following steps S710 to S750:

[0163] Step S710: extracting the actual demand satisfaction rate and the actual supply delay duration from the actual execution effect data.

[0164] The actual demand fulfillment rate is the ratio of the actual quantity of materials provided to the actual quantity required, reflecting the degree to which material supply meets demand. The actual supply delay is the time interval from the occurrence of a material demand event to the completion of the actual supply event, reflecting the timeliness of material supply.

[0165] The actual demand fulfillment rate and actual supply delay duration can be extracted from the actual execution performance data and statistically calculated from relevant business data. For example, the actual sales quantity and order quantity can be obtained from the sales order system to calculate the actual demand fulfillment rate; the material demand time and arrival time can be obtained from the logistics management system to calculate the actual supply delay duration.

[0166] Step S720: extracting the predicted demand satisfaction rate range and the predicted supply delay threshold from the predicted demand distribution result.

[0167] The predicted demand fulfillment rate range is the estimated percentage of material demand that can be met within the predicted demand distribution results. It reflects the expected range of future demand fulfillment. The predicted supply delay threshold is a pre-set time limit used to measure the maximum allowable duration of the predicted material supply delay.

[0168] Extracting the predicted demand fulfillment rate range and predicted supply delay threshold from the predicted demand distribution results can be done directly from the predicted demand distribution results' data structure. These results are typically stored as structured data, such as in a database table or data file. For example, a database table storing predicted demand distribution results may contain dedicated fields for recording the predicted demand fulfillment rate range (e.g., 85%-95%) and the predicted supply delay threshold (e.g., 5 days). By querying this database table, the corresponding data can be accurately extracted.

[0169] Step S730: Calculate a first deviation between the actual demand fulfillment rate and the median of the predicted demand fulfillment rate range, and a second deviation between the actual supply delay duration and the predicted supply delay threshold.

[0170] The first deviation is the difference between the actual demand fulfillment rate and the midpoint of the predicted demand fulfillment rate range, reflecting the degree of difference in the proportion of actual demand fulfillment compared to the predicted situation. The second deviation is the difference between the actual supply delay duration and the predicted supply delay threshold, reflecting the degree of difference in the time between the actual supply delay and the predicted situation.

[0171] To calculate the first deviation, first calculate the median of the predicted demand fulfillment rate range. Assuming the predicted demand fulfillment rate range is a%-b%, the median is (a+b) / 2%. Then, subtract this median from the actual demand fulfillment rate to obtain the first deviation. For example, if the predicted demand fulfillment rate range is 85%-95%, the median is (85+95) / 2 = 90%, and the actual demand fulfillment rate is 80%, then the first deviation is 80%-90% = -10%. To calculate the second deviation, simply subtract the predicted supply delay threshold from the actual supply delay duration.

[0172] Step S740: Determine a deviation coefficient according to the weighted sum of the first deviation amount and the second deviation amount.

[0173] The coefficient of deviation comprehensively considers the impact of the primary and secondary deviations on the overall deviation. To accurately measure the overall deviation, it's necessary to assign different weights to the primary and secondary deviations. These weights can be determined based on actual business circumstances and the importance of each indicator.

[0174] The deviation coefficient is determined based on the weighted sum of the first and second deviations. First, set the weight of the first deviation to w1 and the weight of the second deviation to w2, with w1 + w2 = 1. The formula for calculating the deviation coefficient is: Deviation coefficient = first deviation × w1 + second deviation × w2. For example, if w1 = 0.6 and w2 = 0.4, the first deviation is -10% (converted to a decimal of -0.1), and the second deviation is 2 days, assuming the number of days is also normalized to a decimal (for example, based on 10 days, 2 days is normalized to 0.2), then the deviation coefficient = -0.1 × 0.6 + 0.2 × 0.4 = -0.06 + 0.08 = 0.02.

[0175] Step S750: If the deviation coefficient exceeds the deviation threshold, a deviation analysis report including the deviation type and deviation impact level is generated; wherein the deviation type includes demand forecast deviation and supply response deviation, and the deviation impact level grades the material shortage risk of the supply chain link according to the weighted sum.

[0176] The deviation threshold is a pre-set critical value used to determine whether the deviation coefficient exceeds the acceptable range. When the deviation coefficient exceeds the deviation threshold, it indicates that there is a significant difference between the actual execution and the forecast, and in-depth analysis is required to generate a deviation analysis report. Deviation types are divided into demand forecast deviation and supply response deviation. Demand forecast deviation mainly refers to the difference between actual demand satisfaction and forecast demand satisfaction, while supply response deviation focuses on the difference between actual supply delays and forecast supply delays. The deviation impact level is a grading of the material shortage risk in the supply chain based on the weighted sum, such as high, medium, and low, to assess the impact of the deviation on the supply chain.

[0177] If the deviation coefficient exceeds the deviation threshold, the deviation type is determined based on the specific circumstances of the first and second deviations. A larger first deviation indicates a primary demand forecast deviation; a larger second deviation indicates a primary supply response deviation. The deviation impact level can be categorized based on the magnitude of the deviation coefficient. For example, when the deviation coefficient is between 0 and 0.05, the deviation impact level is low; between 0.05 and 0.1, it's medium; and greater than 0.1, it's high. For example, if the deviation coefficient is 0.15, exceeding the deviation threshold of 0.1, and the first deviation is relatively small while the second deviation is large, the deviation type is supply response deviation, and the deviation impact level is high. Based on this information, a deviation analysis report is generated, detailing the deviation type, deviation impact level, and relevant data analysis results, providing a basis for subsequent adjustments.

[0178] Step S800: If the deviation coefficient in the deviation analysis report exceeds the preset deviation threshold, the parameter correction operation of the dynamic analysis model is triggered; wherein, the parameter correction operation includes: adjusting the weight distribution ratio of the demand quantification module and the trend conduction module in the dynamic analysis model according to the deviation type in the actual execution effect data.

[0179] The preset deviation threshold is a pre-set standard value used to determine whether the deviation has reached a level that requires adjustment to the dynamic analysis model. When the deviation coefficient exceeds this threshold, it indicates that the dynamic analysis model's prediction results deviate significantly from the actual situation and the model needs to be corrected. Parameter correction is the process of adjusting the parameters in the dynamic analysis model. Specifically, based on the type of deviation in the actual execution performance data, the weight distribution ratio of the demand quantification module and the trend transmission module is adjusted to improve the model's forecast accuracy.

[0180] If the deviation coefficient in the deviation analysis report exceeds the preset deviation threshold, appropriate weight adjustments will be made based on the deviation type. If the deviation is a demand forecast deviation, this indicates that the demand quantification module's forecasting performance is poor. The weight of the demand quantification module can be appropriately increased, while the weight of the trend transmission module can be reduced. For example, if the original weight of the demand quantification module was 0.6 and the weight of the trend transmission module was 0.4, the weight of the demand quantification module could be increased to 0.7 and the weight of the trend transmission module could be reduced to 0.3. If the deviation is a supply response deviation, this indicates that the trend transmission module's forecasting capability is insufficient. The weight of the trend transmission module can be increased, while the weight of the demand quantification module can be reduced. For example, the weight of the trend transmission module could be increased from 0.4 to 0.5, while the weight of the demand quantification module could be reduced from 0.6 to 0.5. These weight adjustments enable the dynamic analysis model to better adapt to actual conditions and improve forecast accuracy.

[0181] Step S900: After the corrected dynamic analysis model regenerates an updated forecast demand distribution result, the material supply optimization strategy is iteratively optimized based on the updated forecast demand distribution result.

[0182] The calibrated dynamic analysis model, obtained after parameter correction, more accurately reflects the changing trends in material demand. The updated forecast demand distribution is a prediction of the future distribution of material demand generated by the calibrated dynamic analysis model. Iterative optimization of the material supply strategy involves continuously adjusting and improving the existing material supply optimization strategy based on the updated forecast demand distribution to ensure that material supply better meets actual demand.

[0183] After the corrected dynamic analysis model regenerates an updated forecast demand distribution, the method described in step S500 and its substeps is repeated to generate a new material supply optimization strategy based on the updated forecast demand distribution. During this generation process, the material supply urgency classification, supply resource allocation, and other factors are redefined. For example, if the updated forecast demand distribution indicates an increase in material demand for a particular supply chain link, the new material supply optimization strategy will increase the supply resource allocation for that link, increase the inventory transfer ratio, or increase the weight of external procurement. Through continuous iterative optimization, the material supply optimization strategy can promptly adapt to changes in material demand, improving the operational efficiency and stability of the supply chain.

[0184] As an implementation manner, the method provided in the embodiment of the present invention may further include the following steps S1000 to S1300:

[0185] Step S1000: configuring a data lifecycle identifier for each material flow sequence, where the data lifecycle identifier is used to indicate a valid period of data for the material flow sequence.

[0186] The data lifecycle identifier is assigned to each material flow sequence, defining the timeframe within which the material flow sequence data is valid. The data validity period is the time period during which the material flow sequence data maintains reference value and usability. Beyond this timeframe, the data may lose accuracy and validity due to market changes, business adjustments, and other factors.

[0187] A data lifecycle identifier is configured for each material flow sequence, and the data validity period can be determined based on the characteristics of the material flow sequence and business needs. For some material flow sequences with high stability and small market changes, the data validity period can be set to a longer period, such as one or two years; for some material flow sequences that are greatly affected by market fluctuations, the data validity period can be set to a shorter period, such as one quarter or six months. For example, for the material flow sequence of a certain type of commonly used standard parts, the market demand is relatively stable, and the data validity period can be set to one year; while for the material flow sequence of a certain type of fashionable electronic products, due to the rapid market updates, the data validity period can be set to one quarter. Add a "Data Validity Period" field to each material flow sequence, record its start time and end time, and use it as a data lifecycle identifier.

[0188] Step S1100: When it is detected that the current time exceeds the data validity period, the corresponding material flow sequence is marked as expired data, and the expired data is migrated to the archive storage space.

[0189] The current time refers to the system's current timestamp, used for comparison with the data validity period in the data lifecycle identifier. Expired data refers to material flow sequence data whose current time has exceeded its validity period. While this data may still have some historical reference value, it is no longer applicable for current analysis and forecasting. Archive storage is a dedicated storage area for expired data. It can be an internal enterprise disk array, tape library, or other storage device, or a storage service provided by a cloud storage service provider.

[0190] When it is detected that the current time exceeds the data validity period, the system will automatically identify the corresponding material flow sequence as expired data. You can write a scheduled task program to regularly (such as every day) check the relationship between the data validity period and the current time of all material flow sequences. When it is found that the end time of a material flow sequence is earlier than the current time, mark the sequence as expired data. For example, add an "expired mark" field to the record of the sequence in the database table and set its value to "yes". Then, migrate the expired data from the original business database to the archive storage space. You can use the data migration tool to copy expired data to the archive storage space according to preset rules, and delete this data from the original database to save storage space.

[0191] Step S1200: When generating the predicted demand distribution result, expired data in the archive storage space is excluded, and feature extraction and trend matching processing are performed only based on the non-expired material flow sequence.

[0192] When generating forecast demand distribution results, to ensure accuracy and effectiveness, it's necessary to exclude expired data from archived storage. This is because expired data may no longer reflect current market conditions and business needs, and using this data for analysis and forecasting can introduce errors. Performing feature extraction and trend matching based solely on non-expired material flow sequences can ensure forecast results are more realistic.

[0193] When generating the predicted demand distribution results, the system automatically filters out material flow sequences that have not expired. During the data query phase, query conditions are used to filter out material flow sequence data marked with an "expired" flag. Feature extraction and trend matching are then performed on the filtered, unexpired material flow sequences. Following the method described in step S300 and its substeps, dynamic material demand characteristics and material trend evolution characteristics are extracted. Joint trend matching is performed based on these characteristics to generate the predicted demand distribution results. For example, during the feature extraction process, only the time-series fluctuation data of unexpired material flow sequences is analyzed to identify demand mutation nodes and supply bottleneck nodes, thereby obtaining accurate dynamic material demand characteristics.

[0194] Step S1300: regularly performing statistical analysis on expired data in the archive storage space, generating a historical trend summary report, and inputting the historical trend summary report into the dynamic analysis model for long-term trend learning.

[0195] Regular statistical analysis of expired data in archived storage can uncover hidden historical trends and patterns within the data. A historical trend summary report summarizes and refines the results of the statistical analysis of expired data, including information on trends and cyclical patterns in material demand over the past period. By feeding this historical trend summary report into a dynamic analysis model for long-term trend learning, the model can absorb insights and patterns from historical data, improving its ability to predict future material demand.

[0196] Data analysis tools and algorithms can be used to perform statistical analysis of expired data in archived storage space on a regular basis (e.g., annually). For example, a time series analysis algorithm can be used to analyze the material demand in expired data to identify its cyclical patterns and long-term trends. A historical trend summary report is generated based on the analysis results, which includes information on material demand growth trends, seasonal fluctuations, and so on. The historical trend summary report is input into the dynamic analysis model in an appropriate data format (such as a CSV file, a JSON file, etc.). After receiving the historical trend summary report, the dynamic analysis model combines the information in the report with the current training data to adjust and optimize the model parameters to achieve long-term trend learning. For example, based on the demand growth patterns for a certain material in a specific season of each year discovered in the historical trend summary report, the model adjusts its forecast parameters for future material demand in that season to improve the accuracy of the forecast.

[0197] In addition, it can be understood that the various algorithms involved in the above-mentioned introduction of the embodiments of the present invention can be learned from the relevant content in the prior art. In order to save space, they will not be expanded too much in the embodiments of the present invention. In addition, when implementing the scheme of the present invention, those skilled in the art can supplement the details according to the common knowledge in this field. For example, according to the common knowledge in this field, normalization can be used to eliminate dimensional conflicts before feature fusion, interpolation can be used to eliminate dimensional differences, and historical data, experience or business scenario requirements can be combined to reasonably set thresholds, and models can be trained based on general model training methods, etc. The present invention will no longer provide redundant introductions to overly detailed implementation processes.

[0198] An embodiment of the present invention provides a material demand analysis and forecasting device, including a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the program, some or all of the steps in the above method are implemented.

[0199] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements some or all of the steps in the above method. The computer-readable storage medium may be transient or non-transient.

[0200] Figure 2 A schematic diagram of the hardware entity of the material demand analysis and prediction device provided by an embodiment of the present invention is shown as follows: Figure 2 As shown, the hardware entity of the material demand analysis and prediction device 1000 includes: a processor 1001 and a memory 1002, wherein the memory 1002 stores a computer program that can be run on the processor 1001, and when the processor 1001 executes the program, the steps in the method of any of the above embodiments are implemented.

[0201] The memory 1002 stores computer programs that can be run on the processor. The memory 1002 is configured to store instructions and applications executable by the processor 1001. It can also cache data to be processed or processed by the processor 1001 and each module in the material demand analysis and forecasting device 1000 (for example, image data, audio data, voice communication data and video communication data). It can be implemented through flash memory (FLASH) or random access memory (RAM).

[0202] When the processor 1001 executes the program, it implements the steps of any of the above-mentioned methods for analyzing and predicting material requirements based on big data. The processor 1001 generally controls the overall operation of the material requirement analysis and prediction device 1000.

[0203] An embodiment of the present invention provides a computer storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the steps of the material demand analysis and forecasting method based on big data as in any of the above embodiments.

[0204] The above description is only an embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A material demand analysis and forecasting method based on big data, characterized in that: The following steps are involved: Acquire a multi-source historical material data set for the supply chain link where the target enterprise is located, wherein the multi-source historical material data set includes multiple material flow sequences, each material flow sequence consisting of at least one material demand event and a corresponding material supply event; Performing cross-link integration processing on the multi-source historical material data set to obtain a cross-link integrated data set, wherein the cross-link integrated data set is used to reflect the flow status changes of materials between different supply chain links; Extracting time series fluctuation data of each material flow sequence from the cross-link integrated data set, wherein the time series fluctuation data includes the periodic variation amplitude of material demand and the delay duration of material supply response; Based on the periodic variation amplitude of the material demand in the time series fluctuation data, the material flow sequence is divided into a plurality of fluctuation data segments, each fluctuation data segment corresponding to a periodic variation amplitude interval; Calculate the difference between adjacent time periods for the material demand within each fluctuation data segment to generate a difference value sequence, and determine the candidate mutation position based on the continuously increasing or decreasing difference direction changes in the difference value sequence; Extract the material demand change rate within the time series window corresponding to the candidate mutation position. If the change rate exceeds the average change rate threshold of the fluctuation data segment, mark the candidate mutation position as a demand mutation node. For the material supply response delay duration in the time series fluctuation data, identifying an abnormal delay period in which the delay duration exceeds a preset delay benchmark, and extracting a supply event identifier associated with the abnormal delay period; Obtaining the material supply path node information corresponding to the supply event identifier, and based on the supply interruption cause type and path node load status recorded in the node information, screening out the target path node where supply delay occurs due to node load overload, and marking the target path node as a supply bottleneck node; Performing a spatiotemporal correlation analysis on the demand mutation node and the supply bottleneck node to determine the causal correlation strength between the two in the material flow sequence; if the causal correlation strength reaches a preset correlation threshold, assigning a supply bottleneck impact indicator to the demand mutation node; Generate a material demand dynamic feature based on the demand mutation node and the supply bottleneck node, wherein the material demand dynamic feature is used to characterize the non-balanced distribution state of material demand in the time dimension; Extracting the flow rate difference data between different supply chain links of the material flow sequence, and performing trend fitting processing on the flow rate difference data to generate material trend evolution characteristics; the material trend evolution characteristics are used to reflect the transmission delay effect and cumulative impact weight of material demand between different supply chain links; Based on a preset dynamic analysis model, a joint trend matching process is performed on the dynamic characteristics of the material demand and the material trend evolution characteristics to generate a predicted demand distribution result of the material flow sequence; A material supply optimization strategy is generated according to the predicted demand distribution result, and the material supply optimization strategy is fed back to the supply chain management system to trigger a material allocation operation.

2. The method according to claim 1, wherein The cross-link integration processing is performed on the multi-source historical material data set to obtain a cross-link integrated data set, including: Obtaining an original data source identifier of each material flow sequence in the multi-source historical material data set, and determining a supply chain link type corresponding to the material flow sequence based on the original data source identifier; For each supply chain link type, extract abnormal data segments in the material flow sequence, wherein the abnormal data segments include conflicting data segments where there is a timing conflict between a material demand event and a material supply event; Performing conflict resolution processing on the abnormal data segment to generate a conflict resolution result, and performing data correction on the abnormal data segment according to the conflict resolution result to obtain a corrected material flow sequence; The corrected material flow sequence is aligned with other material flow sequences that do not contain abnormal data fragments to generate an aligned cross-link data set, and the aligned cross-link data set is used as the cross-link integrated data set; wherein, the data alignment includes identifying and associating the material flow sequences of different supply chain links according to the material identification, and unifying the timestamp format and measurement unit in the material flow sequence.

3. The method according to claim 1, wherein The method of performing a joint trend matching process on the dynamic characteristics of material demand and the material trend evolution characteristics based on a preset dynamic analysis model to generate a predicted demand distribution result of the material flow sequence includes: Inputting the material demand dynamic characteristics into the demand quantification module of the dynamic analysis model to generate a demand quantification index set, wherein the demand quantification index set includes a demand peak forecast value and a demand valley fluctuation coefficient; Inputting the material trend evolution characteristics into the trend conduction module of the dynamic analysis model to generate a trend conduction indicator set, wherein the trend conduction indicator set includes a trend diffusion rate and a trend attenuation threshold; Calculating the demand transmission matching degree between different supply chain links of the material flow sequence based on the demand quantification indicator set and the trend transmission indicator set; If the demand transmission matching degree is lower than a preset matching threshold, weight adjustment is performed on the material demand dynamic characteristics and the material trend evolution characteristics to generate an adjusted joint feature set; The adjusted joint feature set is input into the distribution prediction module of the dynamic analysis model to generate the predicted demand distribution result, which includes the material demand priority sequence of each supply chain link in a specified future time period.

4. The method according to claim 3, wherein Generating a material supply optimization strategy based on the predicted demand distribution result, and feeding the material supply optimization strategy back to the supply chain management system to trigger a material allocation operation, includes: Extracting a priority sequence identifier from the predicted demand distribution result, and determining an urgency level of material supply according to the priority sequence identifier; Based on the urgency level, a corresponding supply resource quota is allocated to each supply chain link, wherein the supply resource quota includes an inventory allocation ratio and an external procurement weight; generating an initial supply strategy based on the inventory allocation ratio and the external procurement weight, and performing resource conflict detection on the initial supply strategy; If a resource conflict is detected in the initial supply strategy, the inventory allocation ratio and the external procurement weight are dynamically adjusted based on the conflict type to generate a material supply optimization strategy after the conflict is resolved; The inventory transfer instructions and procurement instructions in the material supply optimization strategy are encoded into executable operation codes, and the executable operation codes are transmitted to the strategy execution interface of the supply chain management system.

5. The method according to claim 4, wherein The performing resource conflict detection on the initial supply strategy includes: Obtaining real-time inventory status data for the current supply chain, including available inventory and in-transit inventory; Calculating a target allocation quantity based on the inventory allocation ratio, and determining whether the target allocation quantity exceeds the sum of the available inventory quantity and the in-transit inventory quantity; If it exceeds, it is determined that a first type of resource conflict exists, and a first conflict identifier is generated; Obtaining real-time supply capacity data of external suppliers, including maximum promised quantity and minimum delivery cycle; Calculating a target purchase quantity based on the external purchase weight, and determining whether the target purchase quantity exceeds the maximum committed quantity or whether the minimum delivery cycle is later than the demand time window; If it is greater than or later than, it is determined that a second type of resource conflict exists, and a second conflict identifier is generated; A priority order for conflict resolution is determined according to a combination of types of the first conflict identifier and the second conflict identifier.

6. The method according to claim 5, wherein The dynamically adjusting the inventory allocation ratio and the external procurement weight based on the conflict type to generate a material supply optimization strategy after conflict resolution includes: If the first type of resource conflict exists, the inventory allocation ratio is reduced, and the target allocation quantity is recalculated based on the reduced inventory allocation ratio until the target allocation quantity does not exceed the sum of the available inventory quantity and the in-transit inventory quantity; If the second type of resource conflict exists, the external procurement weight is reduced, and the target procurement quantity is recalculated according to the reduced external procurement weight until the target procurement quantity does not exceed the maximum committed quantity and the minimum supply cycle is earlier than the demand time window; If both the first and second resource conflicts exist, the inventory allocation ratio will be reduced first. After the target allocation volume corresponding to the reduced inventory allocation ratio stabilizes, the external procurement weight will be reduced. regenerating the supply resource quota according to the adjusted inventory allocation ratio and external procurement weight, and updating the initial supply strategy based on the regenerated supply resource quota; The updated initial supply strategy is used as the material supply optimization strategy after conflict resolution.

7. The method according to claim 1, wherein The method further comprises: Real-time monitoring of the strategy execution status of the supply chain management system and acquisition of actual execution effect data of the material allocation operation; Performing deviation analysis on the actual execution effect data and the predicted demand distribution result to generate a deviation analysis report; If the deviation coefficient in the deviation analysis report exceeds a preset deviation threshold, a parameter correction operation on the dynamic analysis model is triggered; The parameter correction operation includes: adjusting the weight distribution ratio of the demand quantification module and the trend transmission module in the dynamic analysis model according to the deviation type in the actual execution effect data; After the corrected dynamic analysis model regenerates an updated forecast demand distribution result, the material supply optimization strategy is iteratively optimized based on the updated forecast demand distribution result.

8. The method according to claim 7, wherein The performing deviation analysis on the actual execution effect data and the predicted demand distribution result to generate a deviation analysis report includes: Extracting the actual demand satisfaction rate and the actual supply delay duration from the actual execution effect data; Extracting a predicted demand satisfaction rate range and a predicted supply delay threshold from the predicted demand distribution result; Calculating a first deviation between the actual demand fulfillment rate and a median of the predicted demand fulfillment rate range, and a second deviation between the actual supply delay duration and the predicted supply delay threshold; determining the deviation coefficient according to a weighted sum of the first deviation amount and the second deviation amount; If the deviation coefficient exceeds the deviation threshold, the deviation analysis report including the deviation type and the deviation impact level is generated; wherein, the deviation type includes demand forecast deviation and supply response deviation, and the deviation impact level grades the material shortage risk of the supply chain link according to the weighted sum.

9. A material demand analysis and forecasting device, comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, characterized in that: When the processor executes the program, the steps of the method according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images