Inventory dynamic adjustment system and method based on cloud computing

Abstract

The present invention relates to the technical field of inventory management, specifically to a cloud computing-based inventory dynamic adjustment system and method, wherein the system includes a demand fluctuation analysis module, an inventory optimization decision module, a data integration update module, and an inventory forecast refinement module, and identifies and records demand fluctuation patterns based on real-time sales data and product categories. The present invention analyzes demand fluctuations through real-time sales data and product categories, making the demand response mechanism more flexible and accurate, and uses a mixed integer linear programming algorithm to optimize inventory configuration, which not only reduces resource waste, but also enhances the cost efficiency of the supply chain. A long short-term memory network is used to update prediction parameters, which enhances the accuracy of prediction, thereby making inventory decisions more scientific and dynamic. By adjusting the prediction process and analysis parameters, it is possible to adapt to market changes more quickly, improve the efficiency of pharmaceutical inventory management, reduce drug expiration and waste, and effectively ensure the quality of medical services.

Description

Technical Field

[0001] The present invention relates to the technical field of inventory management, and in particular to a cloud computing-based inventory dynamic adjustment system and method. Background Art

[0002] Pharmaceutical inventory management technology aims to optimize the storage, distribution and use of drugs and medical supplies, covering all aspects from procurement and storage to drug distribution and expiration management. With the advancement of information technology, pharmaceutical inventory management can track inventory levels in real time, predict demand, automatically replenish stocks, and ensure the transparency and efficiency of the drug supply chain. Effective inventory management can reduce drug waste, avoid out-of-stock and expiration problems, and also improve the quality and response speed of medical services.

[0003] Among them, the dynamic adjustment system of pharmaceutical inventory realizes real-time monitoring and dynamic adjustment of pharmaceutical inventory. By centrally managing data, it predicts pharmaceutical demand and automatically adjusts inventory levels. Its main purpose is to ensure that medical institutions can adjust inventory configuration in time according to actual usage and market changes, thereby reducing inventory backlogs, avoiding drug expiration, and improving the efficiency of medical services.

[0004] Existing systems often affect the timeliness and accuracy of inventory decisions due to slow response speeds and untimely data updates. For example, the lack of effective real-time data analysis functions leads to overstocking or out-of-stock inventory. Especially in public health emergencies, this deficiency leads to supply chain disruptions. Existing demand forecasting technologies do not fully consider market cyclical changes and emergencies, and the forecast results deviate from actual demand, which not only increases operating costs but also affects patients' treatment progress. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a cloud computing-based inventory dynamic adjustment system and method.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a cloud computing-based inventory dynamic adjustment system, the system comprising:

[0007] The demand fluctuation analysis module identifies and records demand fluctuation patterns based on real-time sales data and product categories, analyzes the frequency and amplitude of demand changes, adjusts inventory calculation parameters based on the analysis results, determines the safety stock level, and generates inventory parameter adjustment results;

[0008] The inventory optimization decision module compares the maximum inventory level and supply chain delay based on the inventory parameter adjustment results, uses the MILP algorithm to optimize inventory, analyzes supply chain costs and inventory turnover, refines ordering frequency and quantity, and generates an optimized inventory configuration;

[0009] The data integration update module configures the point of sale data interface based on the inventory optimization configuration, collects real-time data, and uses the long short-term memory network to update the demand forecast parameters, including adjusting the time window and sensitivity, and generating forecast parameter update records;

[0010] The inventory forecast refinement module updates records based on the forecast parameters, analyzes historical sales data and market trends, calculates forecast errors, and adjusts the analysis parameters to refine the forecast process and generate refined inventory forecast results.

[0011] The present invention is improved in that the analysis steps of the demand change frequency and amplitude are specifically as follows:

[0012] Based on real-time sales data and product categories, sales records are extracted, including timestamps and sales volumes, to generate a time-stamped sales record dataset;

[0013] Digital signal processing is performed on the time-stamped sales record data set to analyze the frequency components using discrete Fourier transform, using the formula:

[0014]

[0015] Get frequency and amplitude analysis data ,in, It is Sales volume at a point in time, is the total number of data points, represents the frequency index, is the base of natural logarithms, is an imaginary unit;

[0016] Based on the frequency and amplitude analysis data, the amplitude value corresponding to the frequency is evaluated, and the demand change pattern is analyzed by information quantitative analysis to obtain the amplitude value of each frequency.

[0017] The present invention is improved in that the steps of obtaining the inventory parameter adjustment result are specifically as follows:

[0018] According to the analysis results, the inventory parameters are preliminarily analyzed and the inventory response speed is recalculated using the formula:

[0019]

[0020] Adjust inventory parameters and generate adjusted inventory parameters ,in, The frequency component representing the change in demand, is the current response speed, is the adjustment factor, is the current safety stock level;

[0021] The adjusted inventory parameters are used to redetermine the safety inventory level, match sales fluctuations, reduce out-of-stock and excess inventory risks, and obtain inventory parameter adjustment results.

[0022] The present invention is improved in that the step of comparing the maximum inventory level and the supply chain delay is specifically as follows:

[0023] Based on the inventory parameter adjustment result, extract the current inventory data of multiple drugs, analyze the maximum inventory level of each drug, and obtain the maximum inventory data;

[0024] Query supply chain management information, obtain multi-stage time data for drug delivery, and generate supply chain delay records;

[0025] Comparing the maximum inventory data with the time data in the supply chain delay record, the formula is used:

[0026]

[0027] Calculate differences in inventory and supply chain efficiency and generate variance analysis results ,in, is the maximum inventory level, is the average supply chain delay, is the adjustment factor used to balance the impact of inventory and supply time, Indicates the evaluation period.

[0028] The present invention is improved in that the steps of obtaining the inventory optimization configuration are specifically as follows:

[0029] The MILP algorithm is used to simulate and optimize inventory configuration using the formula:

[0030]

[0031] Calculate optimized inventory configuration ,in, represents the adjustment factor, is the original cost, Represents the impact weight of supply chain delay adjustment, is the safety stock level, is the scaling factor;

[0032] The optimized inventory configuration is subjected to application testing to monitor the performance of the new inventory configuration in controlling inventory costs and optimizing turnover rate, and obtain the test results of the new inventory configuration.

[0033] The present invention is improved in that the step of collecting real-time data is specifically as follows:

[0034] Based on the inventory optimization configuration, configure the data interface of the sales point, including setting the data format and transmission protocol, and standardizing the information flow to obtain the interface configuration record;

[0035] Based on the interface configuration record, real-time data collection is performed using the formula:

[0036]

[0037] Generate a real-time dataset, where Indicates time Cumulative data of sales points at that time, It is Points of sale at time data, Indicates the number of sales points.

[0038] The present invention is improved in that the step of obtaining the prediction parameter update record is specifically as follows:

[0039] Based on the real-time data collected, the demand change trend is analyzed using a long short-term memory network, and the input data is modeled in a time series to generate preliminary forecast parameters;

[0040] Based on the preliminary prediction parameters, the time window and data sensitivity are adjusted, and the formula is used:

[0041]

[0042] Get the updated prediction parameters ,in, represents the weight of the time window, Indicates time demand data, is the model sensitivity adjustment coefficient, are the old prediction parameters, is the stability adjustment factor, is the total number of time points.

[0043] The present invention is improved in that the steps of obtaining the refined inventory forecast results are specifically as follows:

[0044] Based on the forecast parameter update record, analyze historical sales data and market trends, identify key market factors associated with sales volume changes, and obtain market trend analysis results;

[0045] Based on the market trend analysis results, the formula is adopted:

[0046]

[0047] Calculate and analyze forecast errors , and adjust the analysis parameters, among which, is the weight coefficient, is the current sales data, is the forecast data, is the total number of data points;

[0048] Based on the forecast error and in combination with market input data, the forecast process is refined to generate a refined inventory forecast result.

[0049] The cloud computing-based inventory dynamic adjustment method is executed based on the cloud computing-based inventory dynamic adjustment system, and includes the following steps:

[0050] S1: Based on real-time sales data and product categories, identify demand fluctuation patterns, record fluctuation frequency and amplitude, adjust inventory calculation parameters according to analysis results, determine safety stock levels, and generate inventory parameter adjustment results;

[0051] S2: Based on the inventory parameter adjustment results, the maximum inventory level and supply chain delay are compared, the MILP algorithm is applied to optimize the inventory, and the supply chain cost and inventory turnover rate are analyzed, the ordering frequency and quantity are refined, and the inventory optimization configuration is generated;

[0052] S3: Based on the inventory optimization configuration, configure the point of sale data interface, collect real-time data of the point of sale, use the long short-term memory network to adjust the prediction parameters, analyze the time window and sensitivity of the parameters, and generate the prediction parameter update record;

[0053] S4: Based on the forecast parameter update record, analyze historical sales data and market trends, calculate forecast errors, adjust the analysis parameters, refine the forecast process, and generate refined inventory forecast results;

[0054] S5: Based on the refined inventory forecast results, the inventory forecast process is optimized, the inventory level is determined, and the inventory quantity is dynamically adjusted according to market demand to obtain a dynamic inventory adjustment configuration.

[0055] Compared with the prior art, the advantages and positive effects of the present invention are:

[0056] In the present invention, demand fluctuations are analyzed through real-time sales data and product categories, making the demand response mechanism more flexible and accurate, and using a mixed integer linear programming algorithm to optimize inventory allocation, which not only reduces resource waste but also enhances the cost efficiency of the supply chain. Long short-term memory networks are used to update prediction parameters, which enhances the accuracy of predictions and makes inventory decisions more scientific and dynamic. By adjusting the prediction process and analysis parameters, it is possible to adapt to market changes more quickly, improve the efficiency of pharmaceutical inventory management, reduce drug expiration and waste, and effectively ensure the quality of medical services. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 The present invention proposes a module diagram of a cloud computing-based inventory dynamic adjustment system;

[0058] Figure 2 It is a flow chart for analyzing the frequency and amplitude of demand changes in the present invention;

[0059] Figure 3 It is a flow chart for obtaining inventory parameter adjustment results in the present invention;

[0060] Figure 4 A flow chart showing the comparison between the maximum inventory level and the supply chain delay in the present invention;

[0061] Figure 5 A flowchart for obtaining the inventory optimization configuration in the present invention;

[0062] Figure 6 This is a flow chart of collecting real-time data in the present invention;

[0063] Figure 7 A flowchart for obtaining the prediction parameter update record in the present invention;

[0064] Figure 8 The figure is a flowchart for obtaining the refined inventory forecast results in the present invention. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0066] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0067] Example

[0068] See also Figure 1 The present invention provides a technical solution: a cloud computing-based inventory dynamic adjustment system comprising:

[0069] The demand fluctuation analysis module identifies and records demand fluctuation patterns based on real-time sales data and product categories, analyzes the frequency and amplitude of demand changes, adjusts inventory calculation parameters based on the analysis results, determines the safety stock level, and generates inventory parameter adjustment results;

[0070] The inventory optimization decision module compares the maximum inventory level and supply chain delay based on the inventory parameter adjustment results, uses the MILP algorithm to optimize inventory, analyzes supply chain costs and inventory turnover, refines ordering frequency and quantity, and generates inventory optimization configuration;

[0071] The data integration update module configures the point-of-sale data interface based on inventory optimization configuration, collects real-time data, and uses the long short-term memory network to update demand forecast parameters, including adjusting the time window and sensitivity, and generating forecast parameter update records;

[0072] The inventory forecast refinement module updates records based on forecast parameters, analyzes historical sales data and market trends, calculates forecast errors, and adjusts analysis parameters to refine the forecast process and generate refined inventory forecast results.

[0073] Inventory optimization configuration includes optimized ordering frequency, optimized ordering quantity, and cost efficiency indicators. Forecast parameter update records include updated time window size, updated sensitivity settings, and data collection frequency. Refined inventory forecast results include forecast level, trend adaptation indicators, and error analysis indicators.

[0074] See also Figure 2 , the analysis steps of demand change frequency and amplitude are as follows:

[0075] Based on real-time sales data and product categories, sales records are extracted, including timestamps and sales volumes, to generate a time-stamped sales record dataset;

[0076] When extracting sales records for each drug category, the timestamp and sales volume of the records are used to construct a time series according to the formula The sales records are organized into time series data, where: Represents a point in time, Represents at time Sales volume;

[0077] The construction of the time series is based on real-time data monitoring. For example, the sales volume at 08:00 is 50 copies, and the sales volume at 10:00 is 65 copies. Substitute the data points into the formula:

[0078]

[0079]

[0080] By comparing sales data at different time points, sales trends and patterns can be analyzed to understand the distribution and changes of sales activities throughout the day, providing data support for further frequency analysis and demand fluctuation forecasting.

[0081] Digital signal processing is performed on the time-stamped sales record data set to analyze the frequency components using discrete Fourier transform, using the formula:

[0082]

[0083] Get frequency and amplitude analysis data ,in, It is Sales volume at a point in time, is the total number of data points, represents the frequency index, is the base of natural logarithms, is an imaginary unit;

[0084] For example, sales data at four points in time within a day ,but ,calculate of , the process is as follows:

[0085]

[0086]

[0087] The result indicates that the total sales volume is 260, which provides an overall trend in sales and is further used to analyze demand fluctuations.

[0088] Based on the frequency and amplitude analysis data, the amplitude value corresponding to the frequency is evaluated, and the demand change pattern is analyzed quantitatively through information to obtain the amplitude value of each frequency;

[0089] After analyzing the frequency and amplitude data, the amplitude values ​​of different frequencies were evaluated using statistical methods. abs Calculate the amplitude value of each frequency, where represents the amplitude value, It is the frequency and amplitude analysis data, and the abs function is used to calculate the modulus of the complex number;

[0090] For example, if the output above is , then the amplitude of the frequency component is calculated as:

[0091]

[0092]

[0093] The calculation results show that the amplitude of frequency 1 is about 31.62, which helps to understand the degree of sales fluctuation at a specific frequency. Through amplitude analysis, the most significant fluctuation pattern in sales data can be identified, thereby optimizing inventory and supply chain management strategies.

[0094] See also Figure 3 , the specific steps for obtaining the inventory parameter adjustment results are:

[0095] According to the analysis results, the inventory parameters are preliminarily analyzed and the inventory response speed is recalculated using the formula:

[0096]

[0097] Adjust inventory parameters and generate adjusted inventory parameters ,in, The frequency component representing the change in demand, is the current response speed, is the adjustment factor, is the current safety stock level;

[0098] Detected current 10 / day, 15 / day, For 200 copies, adjustment factor Set to 0.5, the calculation is as follows:

[0099]

[0100]

[0101]

[0102] The adjusted safety stock level is approximately 245 units, and the calculation reflects the strategy of appropriately increasing inventory levels in response to increased demand, aiming to reduce the risk of out-of-stocks due to demand fluctuations.

[0103] Use the adjusted inventory parameters to redefine the safety stock level, match sales fluctuations, reduce out-of-stock and excess inventory risks, and obtain inventory parameter adjustment results;

[0104] After applying the adjusted inventory parameters, the safety stock level is recalculated according to the formula Adjust to accommodate sales fluctuations, where represents the new safety stock level, is the adjusted inventory parameter, It is the market volatility factor;

[0105] Adjusted inventory parameters 245 copies, market volatility factor If it is 1.1 (indicating that the market demand is expected to increase by 10%), the calculation process is as follows:

[0106]

[0107] This means that the new safety stock level should be approximately 270 copies, ensuring that inventory levels can effectively cope with future sales peaks while reducing costs caused by excess inventory.

[0108] See also Figure 4 , the comparison steps of maximum inventory level and supply chain delay are as follows:

[0109] Based on the inventory parameter adjustment results, extract the current inventory data of multiple drugs, analyze the maximum inventory level of each drug, and obtain the maximum inventory data;

[0110] Extract the current inventory data of each drug from the inventory management system according to the formula Calculate the maximum inventory level of each drug and obtain the maximum inventory level data , where Indicates the inventory quantity of a single drug;

[0111] There are three drugs with inventory quantities of 50, 75 and 30, calculated according to the formula:

[0112]

[0113] Displaying a maximum inventory level of 75 helps identify peak inventory levels that should be addressed in inventory strategies, ensuring that inventory management strategies are effective and responsive.

[0114] Query supply chain management information, obtain multi-stage time data for drug delivery, and generate supply chain delay records;

[0115] Based on the maximum inventory level data, the transportation and handling time of different supply chain nodes is compared according to the formula Calculating average delay in supply chain , where Indicates The delay time of each node, is the total number of nodes;

[0116] For example, if a supply chain includes three nodes with delay times of 2, 3, and 1 day respectively, according to the formula:

[0117]

[0118] result days, indicating that the average processing and transportation time of the entire supply chain is 2 days. This indicator has important reference value for optimizing supply chain structure and improving logistics efficiency.

[0119] Compare the maximum inventory data with the time data in the supply chain delay record, using the formula:

[0120]

[0121] Calculate differences in inventory and supply chain efficiency and generate variance analysis results ,in, is the maximum inventory level, is the average supply chain delay, is the adjustment factor used to balance the impact of inventory and supply time, Indicates the evaluation cycle;

[0122] Monitored , Day, set adjustment factor , evaluation cycle Days, substitute into the formula and calculate as follows:

[0123]

[0124] The calculation results show that there is a difference of about 23.04 units between inventory levels and supply chain delays, which means that inventory strategies need to be further adjusted to cope with the impact of supply chain delays on inventory management.

[0125] See also Figure 5 , the specific steps for obtaining the optimal inventory configuration are:

[0126] The MILP algorithm is used to simulate and optimize inventory configuration using the formula:

[0127]

[0128] Calculate optimized inventory configuration ,in, represents the adjustment factor, is the original cost, Represents the impact weight of supply chain delay adjustment, is the safety stock level, is the scaling factor;

[0129] set up , , ;

[0130] Collected Yuan, share;

[0131] Calculation process:

[0132]

[0133]

[0134]

[0135] The results show that after adopting the new inventory strategy, the adjusted expected cost is approximately 11,136.36 yuan, showing the cost-effectiveness brought by the new strategy.

[0136] Conduct application tests on the optimized inventory configuration, monitor the performance of the new inventory configuration in controlling inventory costs and optimizing turnover rates, and obtain the test results of the new inventory configuration;

[0137] Collect key performance indicators from actual application tests and use the formula Calculate the efficiency and cost control performance of new inventory strategies , where Represents the inventory cost and turnover rate values ​​obtained from actual monitoring, Represents the preset inventory management target value;

[0138] The actual monitored inventory cost is 9,000 yuan, the preset target is 10,000 yuan, the actual monitored turnover rate is 0.75, the preset target is 0.80;

[0139] Calculation process:

[0140] Inventory cost efficiency:

[0141]

[0142] Turnover efficiency:

[0143]

[0144] The results show that the new inventory strategy showed 90% efficiency in cost control, while the turnover rate reached the expected efficiency of 93.75%, which helps to determine whether the performance of the new strategy in actual application meets the expected goals.

[0145] See also Figure 6 ,The specific steps for collecting real-time data are:

[0146] Based on the inventory optimization configuration, configure the data interface of the sales point, including setting the data format and transmission protocol, and standardizing the information flow to obtain the interface configuration record;

[0147] When configuring the data interface of the point of sale, the data format and transmission protocol need to be set to ensure the standardization of the data flow and the stability of the transmission. Use the formula To calculate the transmission rate, Represents at time The transmission rate under Representative time The packet size, Represents the length of transmission time;

[0148] If the data packet size at each point of sale is 500MB and the transmission time is 10 seconds, the formula is:

[0149]

[0150] The results show that each sales point The data transmission rate is 50MB / s, so the data transmission rate of each sales point is reasonably configured to ensure the real-time and stability of data collection.

[0151] Based on the interface configuration record, real-time data collection is performed using the formula:

[0152]

[0153] Generate a real-time dataset, where Indicates time Cumulative data of sales points at that time, It is Points of sale at time data, Indicates the number of sales points;

[0154] There are three sales points at a time The data are 100, 150 and 200 respectively, then:

[0155]

[0156] The calculation example shows that at time The cumulative data volume of all sales points is 450, which effectively reflects the real-time status of sales activities and provides a data basis for further analysis.

[0157] See also Figure 7 , the specific steps for obtaining the prediction parameter update record are:

[0158] Based on the collected real-time data, the long short-term memory network is used to analyze the demand change trend, and the input data is modeled into time series to generate preliminary forecast parameters;

[0159] When performing long short-term memory network analysis, it is necessary to perform time series processing on the collected real-time data, using the formula Normalize the time series data, where represents the normalized data point, is the original data point, is the mean value of the data set, represents standard deviation;

[0160] The average value of a set of data is 100, standard deviation is 15, and a data point is 115, the formula is:

[0161]

[0162] The results show that the data points The standard deviation is 1 from the mean. This standardization makes data of different sizes comparable and provides accurate input for the LSTM model.

[0163] Based on the preliminary prediction parameters, adjust the time window and data sensitivity, and use the formula:

[0164]

[0165] Get the updated prediction parameters ,in, represents the weight of the time window, Indicates time demand data, is the model sensitivity adjustment coefficient, are the old prediction parameters, is the stability adjustment factor, is the total number of time points;

[0166] Data were collected at 4 time points:

[0167] They are 1, 2, 3 and 4 respectively;

[0168] They are 100, 200, 300 and 400 respectively;

[0169] is 50;

[0170] is 0.5;

[0171] is 10.

[0172] The calculation is as follows:

[0173]

[0174]

[0175]

[0176]

[0177] This calculation shows that after considering the data weights and adjustment factors, the new prediction parameters It is 303, which helps to adjust the model parameters more accurately to adapt to the actual changes in the data.

[0178] See also Figure 8 , the steps to obtain the refined inventory forecast results are as follows:

[0179] Based on the forecast parameter update records, analyze historical sales data and market trends, identify key market factors associated with sales volume changes, and obtain market trend analysis results;

[0180] When analyzing historical sales data and market trends, use formulas The sales data is seasonally adjusted and trended, where: represents the adjusted trend component, It's in time The actual observed value of It is the trend component of the previous period. It is the seasonal ingredient of the previous issue. is a smoothing parameter used to control the impact of actual observations on the trend component;

[0181] In time of is 200, is 190, is 10, let is 0.5;

[0182] The calculation is as follows:

[0183]

[0184]

[0185] The result shows that after considering seasonal adjustment and trend decomposition, the trend component of the current period is 200, indicating that the sales trend remains stable to reflect seasonal changes and market trends.

[0186] Based on the market trend analysis results, the formula is adopted:

[0187]

[0188] Calculate and analyze forecast errors , and adjust the analysis parameters, among which, is the weight coefficient, is the current sales data, is the forecast data, is the total number of data points;

[0189] There are three data points for monitoring:

[0190] ;

[0191] [100, 150, 120];

[0192] ;

[0193] The calculation is as follows:

[0194]

[0195]

[0196] The results show that the forecast error is 20, which is used to further refine the forecast model to ensure that the adjustments can effectively reflect the actual market dynamics.

[0197] Based on the forecast error and combined with market input data, the forecast process is refined to generate refined inventory forecast results;

[0198] When applying the adjusted forecast parameters to the forecast model, the formula The new market data and adjusted parameters are input into the model to recalculate and generate refined inventory forecast results, where: represents the updated prediction result, is the previous prediction result, is the newly entered market data, is the smoothing coefficient;

[0199] Before collecting is 200, is 220, Set to 0.8;

[0200] The calculation is as follows:

[0201]

[0202]

[0203] Based on current market inputs and adjusted parameters, the updated inventory forecast result is 204, ensuring the responsiveness and accuracy of the forecast model and adapting to market changes.

[0204] The inventory dynamic adjustment method based on cloud computing includes the following steps:

[0205] S1: Based on real-time sales data and product categories, identify demand fluctuation patterns, record fluctuation frequency and amplitude, adjust inventory calculation parameters according to analysis results, determine safety stock levels, and generate inventory parameter adjustment results;

[0206] S2: Based on the inventory parameter adjustment results, the maximum inventory level and supply chain delay are compared, the MILP algorithm is applied to optimize inventory, and the supply chain cost and inventory turnover rate are analyzed to refine the ordering frequency and quantity and generate the inventory optimization configuration;

[0207] S3: Based on inventory optimization configuration, configure the point-of-sale data interface, collect real-time data from the point of sale, use the long short-term memory network to adjust the prediction parameters, analyze the time window and sensitivity of the parameters, and generate prediction parameter update records;

[0208] S4: Update records based on forecast parameters, analyze historical sales data and market trends, calculate forecast errors, adjust analysis parameters, refine forecast processes, and generate refined inventory forecast results;

[0209] S5: Based on the refined inventory forecast results, the inventory forecast process is optimized, the inventory level is determined, and the inventory quantity is dynamically adjusted according to market demand to obtain a dynamic inventory adjustment configuration.

[0210] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. The cloud computing-based inventory dynamic adjustment system is characterized by: The system comprises: The demand fluctuation analysis module identifies and records demand fluctuation patterns based on real-time sales data and product categories, analyzes the frequency and amplitude of demand changes, adjusts inventory calculation parameters based on the analysis results, determines the safety stock level, and generates inventory parameter adjustment results; The inventory optimization decision module compares the maximum inventory level and supply chain delay based on the inventory parameter adjustment results, uses the MILP algorithm to optimize inventory, analyzes supply chain costs and inventory turnover, refines ordering frequency and quantity, and generates an optimized inventory configuration; The data integration update module configures the point of sale data interface based on the inventory optimization configuration, collects real-time data, and uses the long short-term memory network to update the demand forecast parameters, including adjusting the time window and sensitivity, and generating forecast parameter update records; The inventory forecast refinement module updates records based on the forecast parameters, analyzes historical sales data and market trends, calculates forecast errors, and adjusts the analysis parameters to refine the forecast process and generate refined inventory forecast results; The analysis steps of the demand change frequency and amplitude are specifically as follows: Based on real-time sales data and product categories, sales records are extracted, including timestamps and sales volumes, to generate a time-stamped sales record dataset; Digital signal processing is performed on the time-stamped sales record data set to analyze the frequency components using discrete Fourier transform, using the formula: Get frequency and amplitude analysis data ,in, It is Sales volume at a point in time, is the total number of data points, represents the frequency index, is the base of natural logarithms, is an imaginary unit; Based on the frequency and amplitude analysis data, the amplitude value corresponding to the frequency is evaluated, and the amplitude value of each frequency is obtained by quantitatively analyzing the demand change pattern through information; The steps for obtaining the inventory parameter adjustment result are specifically as follows: According to the analysis results, the inventory parameters are preliminarily analyzed and the inventory response speed is recalculated using the formula: Adjust inventory parameters and generate adjusted inventory parameters ,in, The frequency component representing the change in demand, is the current response speed, is the adjustment factor, is the current safety stock level; Using the adjusted inventory parameters, re-determine the safety inventory level, match sales fluctuations, reduce out-of-stock and excess inventory risks, and obtain inventory parameter adjustment results; The specific steps for comparing the maximum inventory level and supply chain delay are: Based on the inventory parameter adjustment result, extract the current inventory data of multiple drugs, analyze the maximum inventory level of each drug, and obtain the maximum inventory data; Query supply chain management information, obtain multi-stage time data for drug delivery, and generate supply chain delay records; Comparing the maximum inventory data with the time data in the supply chain delay record, the formula is used: Calculate differences in inventory and supply chain efficiency and generate variance analysis results ,in, is the maximum inventory level, is the average supply chain delay, is the adjustment factor used to balance the impact of inventory and supply time, represents the evaluation cycle; The steps for obtaining the inventory optimization configuration are specifically as follows: The MILP algorithm is used to simulate and optimize inventory configuration using the formula: Calculate optimized inventory configuration ,in, represents the adjustment factor, is the original cost, Represents the impact weight of supply chain delay adjustment, is the safety stock level, is the scaling factor; Perform application testing on the optimized inventory configuration, monitor the performance of the new inventory configuration in controlling inventory costs and optimizing turnover rate, and obtain test results of the new inventory configuration; The steps of collecting real-time data are specifically as follows: Based on the inventory optimization configuration, configure the data interface of the sales point, including setting the data format and transmission protocol, and standardizing the information flow to obtain the interface configuration record; Based on the interface configuration record, real-time data collection is performed using the formula: Generate a real-time dataset, where Indicates time Cumulative data of sales points at that time, It is Points of sale at time data, Indicates the number of sales points; The steps for obtaining the prediction parameter update record are specifically as follows: Based on the real-time data collected, the demand change trend is analyzed using a long short-term memory network, and the input data is modeled in a time series to generate preliminary forecast parameters; Based on the preliminary prediction parameters, the time window and data sensitivity are adjusted, and the formula is used: Get the updated prediction parameters ,in, represents the weight of the time window, Indicates time demand data, is the model sensitivity adjustment coefficient, are the old prediction parameters, is the stability adjustment factor, is the total number of time points; The steps for obtaining the refined inventory forecast results are specifically as follows: Based on the forecast parameter update record, analyze historical sales data and market trends, identify key market factors associated with sales volume changes, and obtain market trend analysis results; Based on the market trend analysis results, the formula is adopted: Calculate and analyze forecast errors , and adjust the analysis parameters, among which, is the weight coefficient, is the current sales data, is the forecast data, is the total number of data points; Based on the forecast error and in combination with market input data, the forecast process is refined to generate a refined inventory forecast result.

2. The method for dynamic inventory adjustment based on cloud computing is characterized in that: The cloud computing-based inventory dynamic adjustment system according to claim 1 comprises the following steps: Based on real-time sales data and product categories, identify demand fluctuation patterns, record fluctuation frequency and amplitude, adjust inventory calculation parameters according to analysis results, determine safety stock levels, and generate inventory parameter adjustment results; Based on the inventory parameter adjustment results, the maximum inventory level and supply chain delay are compared, the MILP algorithm is applied to optimize inventory, and the supply chain cost and inventory turnover rate are analyzed, the ordering frequency and quantity are refined, and the inventory optimization configuration is generated; Based on the inventory optimization configuration, configure the point of sale data interface, collect real-time data of the point of sale, use the long short-term memory network to adjust the prediction parameters, analyze the time window and sensitivity of the parameters, and generate the prediction parameter update record; Based on the forecast parameter update record, analyze historical sales data and market trends, calculate forecast errors, and adjust the analysis parameters to refine the forecast process and generate refined inventory forecast results; Based on the refined inventory forecast results, the inventory forecast process is optimized, the inventory level is determined, and the inventory quantity is dynamically adjusted according to market demand to obtain a dynamic inventory adjustment configuration.

Images