Tissue consumption prediction and management system

Abstract

The invention relates to the technical field of inventory management, in particular to a tissue consumption prediction and management system which comprises a consumption behavior recognition module, an abnormal consumption analysis module, a consumption trend prediction module, a dynamic threshold management module, an inventory scheduling optimization module and an inventory reserve planning module. According to the method, through comprehensive analysis of the tissue usage amount and the storage point position data, the daily time period consumption is accurately counted, the grouping difference and proportion are clarified, the precision and timeliness of regional resource configuration are improved, the abnormal consumption is identified, the detection sensitivity and accuracy are improved through the association mark of the fluctuation and proportion difference value, the abnormal consumption data are combined, and the detection accuracy is improved. Trend prediction precision is optimized, a basis is provided for inventory regulation and control, the adaptive capacity of inventory management is enhanced through dynamic threshold value management, inventory is analyzed in real time, the threshold value is adjusted in real time, resource allocation efficiency is improved, transportation cost is reduced, the inventory matching degree is optimized through a quarterly inventory storage scheme, and resource utilization efficiency is improved.

Description

Technical Field

[0001] The invention relates to the technical field of inventory management, and in particular to a paper towel consumption prediction and management system. Background Art

[0002] The field of inventory management technology includes technical means to effectively monitor and manage the inventory levels of goods, materials or products. The core goal of inventory management is to ensure the efficient operation of the supply chain, avoid inventory backlogs and out-of-stock phenomena, thereby reducing operating costs and improving resource utilization. The technologies involved in this field include inventory forecasting, demand analysis, replenishment strategies, inventory optimization, etc. Inventory management technology is not only used in traditional retail industries, but also widely used in various fields such as manufacturing, logistics, and catering. The core method relies on data analysis technology, algorithm optimization, and automation systems. While monitoring inventory dynamics in real time, it uses historical data and trends to predict future demand, adjust inventory levels, and ensure the stability and flexibility of the supply chain.

[0003] Among them, the paper towel consumption prediction and management system refers to a system that predicts the consumption of paper towel products and manages inventory. The system mainly focuses on the consumption characteristics of paper towels in enterprises or public places, and relies on real-time collected data to analyze and predict the consumption trend of paper towels. The system monitors various environmental factors, such as frequency of use, time period changes, regional demand differences, etc., combined with historical consumption records, and uses a specific calculation model to predict future paper towel demand and adjust inventory based on this prediction result. The main technical means of the system include modeling of historical consumption data, consumption pattern analysis, and linkage mechanism with the inventory management system.

[0004] Although existing technologies can use data analysis techniques to predict and adjust inventory in inventory management, in the analysis of consumption distribution within a region, they lack refined processing of specific time periods and spatial locations, which easily leads to blindness and limitations in resource allocation. In terms of identifying abnormal consumption, existing technologies mostly rely on single-threshold settings and are unable to dynamically perceive the fluctuation characteristics within a time period, resulting in inaccurate abnormal markings, thereby increasing the monitoring difficulty of abnormal consumption. In trend prediction, existing technologies rely on the overall fitting of historical data and lack separate processing of abnormal consumption data and analysis of the increasing and decreasing amplitudes of continuous changes, resulting in insufficient prediction accuracy and inability to accurately match dynamic demands. In threshold management, threshold adjustments in existing technologies are mostly fixed settings, making it difficult to adapt to the dynamic changes in demands of different regions or time periods, resulting in a decline in inventory management efficiency. In inventory scheduling, existing technologies lack comprehensive analysis of the priority scheduling order and remaining transportation volume, leading to inefficient scheduling strategies, increasing logistics costs and time consumption. In long-term inventory planning, existing technologies pay more attention to the overall inventory level and ignore the demand distribution law within a time period, easily causing seasonal backlogs or out-of-stock phenomena, and reducing the flexibility and stability of the supply chain. Summary of the Invention

[0005] To solve the technical problems of existing technologies in inventory management, although they can use data analysis techniques to predict and adjust inventory, in the analysis of consumption distribution within a region, they lack refined processing of specific time periods and spatial locations, which easily leads to blindness and limitations in resource allocation. In terms of identifying abnormal consumption, existing technologies mostly rely on single-threshold settings and are unable to dynamically perceive the fluctuation characteristics within a time period, resulting in inaccurate abnormal markings, thereby increasing the monitoring difficulty of abnormal consumption. In trend prediction, existing technologies rely on the overall fitting of historical data and lack separate processing of abnormal consumption data and analysis of the increasing and decreasing amplitudes of continuous changes, resulting in insufficient prediction accuracy and inability to accurately match dynamic demands. In threshold management, threshold adjustments in existing technologies are mostly fixed settings, making it difficult to adapt to the dynamic changes in demands of different regions or time periods, resulting in a decline in inventory management efficiency. In inventory scheduling, existing technologies lack comprehensive analysis of the priority scheduling order and remaining transportation volume, leading to inefficient scheduling strategies, increasing logistics costs and time consumption. In long-term inventory planning, existing technologies pay more attention to the overall inventory level and ignore the demand distribution law within a time period, easily causing seasonal backlogs or out-of-stock phenomena, and reducing the flexibility and stability of the supply chain, embodiments of the present invention provide a tissue consumption prediction and management system. The technical solution is as follows:

[0006] A tissue consumption prediction and management system is provided, and the system includes:

[0007] The consumption behavior recognition module extracts the paper towel usage and storage point location data based on the consumption records of paper towel storage points in the region, counts the daily paper towel consumption in each time period, analyzes the grouping differences and proportions, and obtains the regional paper towel consumption distribution data;

[0008] The abnormal consumption analysis module analyzes the consumption fluctuation range of high-frequency time periods based on the regional paper towel consumption distribution data, identifies the time period where the ratio difference marks the out-of-range period, and generates a paper towel consumption abnormality marking result;

[0009] The consumption trend prediction module identifies the paper towel consumption data in the abnormal time period based on the abnormal paper towel consumption marking result, calculates the change range and increase / decrease ratio, fits the consumption, and generates a paper towel consumption trend distribution data set;

[0010] The dynamic threshold management module extracts the extreme value of the change ratio increase or decrease in the area based on the paper towel consumption trend distribution data set, calculates the difference between the consumption and the dynamic offset range, analyzes the offset value and the upper and lower limit dynamic range, and generates a dynamic threshold allocation table;

[0011] The inventory scheduling optimization module extracts the real-time inventory of each regional storage point based on the dynamic threshold allocation table, analyzes the difference between the inventory and the adjustment threshold, reallocates the inventory, and generates a paper towel regional inventory scheduling table;

[0012] The inventory reserve planning module identifies long-term inventory demand based on the paper towel regional inventory scheduling table, analyzes the deviation between the current inventory and the reserve allocation ratio, plans the inventory ratio by time period, and generates a quarterly inventory reserve plan for paper towels.

[0013] As a further solution of the present invention, the regional paper towel consumption distribution data includes daily time period grouping statistics, grouping difference data and grouping ratio data, the paper towel consumption abnormal marking results include high-frequency time period consumption fluctuation range, low-frequency time period consumption ratio and out-of-range time period mark, the paper towel consumption trend distribution data set includes abnormal time period consumption data, continuous time period consumption change amplitude and increase and decrease ratio fitting results, the dynamic threshold allocation table includes the regional change ratio increase and decrease extreme values, dynamic offset range and dynamic range upper and lower limit analysis results, the paper towel regional inventory scheduling table includes the real-time inventory of regional storage points, the difference between the inventory and the adjustment threshold and the storage point priority scheduling order, and the paper towel quarterly inventory reserve plan includes long-term inventory demand, inventory reserve allocation ratio and time period inventory planning ratio.

[0014] As a further solution of the present invention, the consumption behavior identification module includes:

[0015] Use the record extraction submodule to classify the data according to daily time and location based on the consumption records of paper towel storage points in the area, group the daily consumption records by storage point, extract the time series and spatial location series, and establish a location and time data matching table;

[0016] The grouping statistical analysis submodule divides the paper towel usage into differentiated daily time periods based on the location and time data matching table, accumulates the paper towel consumption of the storage points within the time period, analyzes the segment proportions, and obtains consumption group proportion data;

[0017] The spatiotemporal consumption analysis submodule extracts the consumption ratio characteristics within the storage point time period based on the consumption grouping ratio data, associates the consumption ratio with the location data, analyzes the spatiotemporal consumption relationship of the storage points in the area, and obtains the regional paper towel consumption distribution data.

[0018] As a further solution of the present invention, the abnormal consumption analysis module includes:

[0019] The high- and low-frequency ratio analysis submodule divides the consumption data into time periods based on the regional paper towel consumption distribution data, summarizes the consumption in each time period, analyzes the total consumption ratio of the time period, distinguishes high-frequency and low-frequency time periods, and obtains a high- and low-frequency ratio data set;

[0020] The fluctuation range extraction submodule arranges the proportion data in the high-frequency time period in time series based on the high- and low-frequency ratio data set, divides the data in the continuous time period into multiple fluctuation intervals, performs change trend analysis on the data in each interval, analyzes the change amplitude within the fluctuation range, and obtains the consumption fluctuation characteristics;

[0021] Based on the consumption fluctuation characteristics, the abnormal marking submodule compares the proportion of each time period in the high- and low-frequency ratio data with the consumption fluctuation range item by item, marks the time period where the proportion difference exceeds the fluctuation range as abnormal, records the proportion and difference content of the abnormal time period, and generates the abnormal marking result of paper towel consumption.

[0022] As a further solution of the present invention, the consumption trend prediction module includes:

[0023] The abnormal marking submodule extracts the consumption data value of the abnormal time period based on the abnormal marking result of the paper towel consumption, compares it with the normal consumption range, analyzes the characteristics of the abnormal interval and the data fluctuation characteristics, and obtains the abnormal consumption time period data set;

[0024] The variation range statistics submodule groups the consumption data within the time period based on the abnormal consumption time period data set, counts the consumption difference between each group, and obtains variation range statistics;

[0025] The consumption trend fitting submodule is based on the variation amplitude statistical data, takes the variation amplitude and total consumption in the time series as data fitting parameters, performs trend fitting on the consumption changes according to the time interval, calls the cumulative characteristics of the increase and decrease amplitude, and combines the fitting results to calculate the predicted paper towel consumption trend value, and obtain the paper towel consumption trend distribution data set.

[0026] As a further solution of the present invention, the predicted paper towel consumption trend value is calculated according to the formula:

[0027] y=β 0 +β 1 x 1 +β 2 x 2 +β 3 x 3 ;

[0028] Among them, y represents the predicted trend value of paper towel consumption, β 0 represents the intercept of the regression equation, β 1 represents the influence coefficient of the change in the time series, x 1 Represents the magnitude of change within a specific time interval, β 2 Represents the influence coefficient of total consumption, x 2 represents the total consumption in a specific time interval, β 3 The weighted adjustment coefficient representing the trend of change, x 3 Represents the consumption difference between the previous time interval and the current time interval.

[0029] As a further solution of the present invention, the dynamic threshold management module includes:

[0030] The variation extreme value extraction submodule screens the extreme values ​​in the consumption change ratio in each time period based on the paper towel consumption trend distribution data set, and calculates the consumption trend at the time point where the extreme value is located to obtain variation extreme value data;

[0031] The dynamic offset analysis submodule analyzes the difference between the consumption and the dynamic offset range based on the extreme value data, counts the dynamic changes of the offset value in each time period, and analyzes the time span and dynamic characteristics of the offset interval to obtain the offset dynamic range data;

[0032] The threshold distribution acquisition submodule extracts the upper and lower limits of the time period offset value based on the offset dynamic range data, divides the dynamic threshold interval according to the time span, sorts out the upper and lower limit allocation rules of the threshold within the time period, and generates a dynamic threshold allocation table based on the consumption trend in the area.

[0033] As a further solution of the present invention, the inventory scheduling optimization module includes:

[0034] The inventory data capture submodule checks the consistency of the inventory data of each regional storage point with the data returned by the real-time interface based on the dynamic threshold allocation table, analyzes the integrity of inventory parameters, and generates a real-time inventory list of the regional storage point;

[0035] The inventory difference analysis submodule extracts the threshold range of each storage point based on the real-time inventory list of the regional storage points, compares the storage point inventory with the corresponding adjustment threshold, identifies the inventory difference of each storage point, sorts the storage points according to the absolute value of the difference, and generates the inventory difference analysis result;

[0036] Based on the inventory difference analysis results, the inventory reallocation submodule extracts the priority scheduling order and remaining transportation volume of each storage point, analyzes the distribution of insufficient storage points and excess storage points item by item, allocates excess inventory to insufficient storage points, calculates the inventory volume between storage points, and generates a paper towel area inventory scheduling table.

[0037] As a further solution of the present invention, the inventory between the storage points is calculated according to the formula

[0038]

[0039] Among them, X i,j represents the inventory quantity that needs to be allocated from storage point i to storage point j, S i represents the current inventory quantity at storage point i, D j represents the inventory quantity required for storage point j, C i represents the transportation capacity of storage point i, T j represents the transportation time, P j represents the inventory priority of storage point j, W 1 Represents the adjustment coefficient in the scheduling algorithm, W 2 Represents the adjustment coefficient in the scheduling algorithm, A i represents the remaining transportation volume of storage point i, B j Represents the change in inventory demand at storage point j.

[0040] As a further solution of the present invention, the inventory reserve planning module includes:

[0041] The inventory distribution analysis submodule matches the regional and inventory demand data based on the paper towel regional inventory scheduling table, verifies the data integrity, removes abnormal values, reorganizes the regional inventory and demand data, and generates a regional inventory and demand distribution data table;

[0042] The reserve deviation analysis submodule extracts the inventory and long-term demand of each region based on the regional inventory and demand distribution data table, compares the data according to regional classification, calculates the difference between regional inventory and demand, converts the difference into a reserve allocation deviation parameter, sorts the regional deviations, and generates a reserve allocation ratio deviation table;

[0043] The inventory ratio planning submodule extracts regional inventory demand distribution data based on the reserve allocation ratio deviation table, matches reserve demand with current inventory, adjusts reserve ratio and inter-regional coordination, and generates a quarterly paper towel inventory reserve plan.

[0044] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0045] By combining and analyzing the paper towel usage and storage point location data, we can achieve accurate grouping statistics of daily time period consumption, further clarify the group differences and proportions, and present the dynamic distribution of regional paper towel consumption in a more detailed manner. This detailed analysis based on time periods and spatial locations improves the accuracy and timeliness of resource allocation in the region. By extracting the consumption ratios of high- and low-frequency time periods and conducting in-depth analysis of consumption fluctuations in high-frequency time periods, we can identify abnormal consumption time periods and establish an association marking mechanism for fluctuations and ratio differences, thereby improving the sensitivity and accuracy of abnormal consumption detection and reducing the possibility of omissions. In trend prediction, by combining abnormal consumption data with the difference in changes in continuous time periods, we can quantify the increase or decrease and propose It is in line with future consumption trends, significantly optimizes the accuracy of trend prediction, and provides a scientific basis for inventory control. The allocation of dynamic thresholds not only enhances the flexibility of threshold adjustment by calculating the change ratio of extreme values ​​and dynamic offset range, but also generates a more suitable dynamic range in combination with time period changes and regional distribution to ensure the dynamic adaptability of inventory management. Through real-time analysis of the difference between the inventory volume of storage points and the adjustment threshold, as well as the optimal combination of priority scheduling sequences, the efficiency and flexibility of resource allocation are achieved, and the transportation cost and scheduling time within the region are reduced. The quarterly inventory reserve plan is based on the combination of long-term demand analysis and time period planning, which significantly improves the matching degree of seasonal inventory, reduces the risk of inventory backlog, and improves the resource utilization efficiency of the supply chain. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0047] Figure 1 is a schematic diagram of a paper towel consumption prediction and management system provided by an embodiment of the present invention;

[0048] Figure 2 It is a schematic diagram of the system framework of the present invention;

[0049] Figure 3 It is a flow chart of the consumption behavior identification module in the present invention;

[0050] Figure 4 It is a flow chart of the abnormal consumption analysis module in the present invention;

[0051] Figure 5 It is a flow chart of the consumption trend prediction module in the present invention;

[0052] Figure 6 This is a flow chart of the dynamic threshold management module in the present invention;

[0053] Figure 7 It is a flow chart of the inventory scheduling optimization module in the present invention;

[0054] Figure 8 It is a flow chart of the inventory reserve planning module in the present invention. DETAILED DESCRIPTION

[0055] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0056] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0057] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0058] In the embodiments of the present invention, sometimes the subscripts such as W 1 It may be written in non-subscript form such as W1. When the difference is not emphasized, the meaning is the same.

[0059] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0060] The embodiment of the present invention provides a paper towel consumption prediction and management system. Figure 1-2 The schematic diagram of the paper towel consumption prediction and management system shown in the figure includes:

[0061] The consumption behavior recognition module extracts the consumption of paper towels and the location data of the storage points based on the consumption records of the paper towel storage points in the area, and counts the consumption by grouping according to the daily time period, analyzes the group differences and proportions, and obtains the regional paper towel consumption distribution data;

[0062] The abnormal consumption analysis module extracts the proportion of paper towel usage in high- and low-frequency time periods based on regional paper towel consumption distribution data, analyzes the consumption fluctuation range in high-frequency time periods, identifies the time periods where the proportion difference marks the out-of-range period, and generates abnormal paper towel consumption marking results;

[0063] The consumption trend prediction module identifies the consumption data of paper towels in abnormal time periods based on the abnormal marking results of paper towel consumption, combines the consumption change difference in continuous time periods, calculates the change range and increase / decrease ratio, fits the consumption, and generates a paper towel consumption trend distribution data set;

[0064] The dynamic threshold management module extracts the extreme value of the change ratio increase or decrease in the area based on the paper towel consumption trend distribution data set, calculates the difference between the consumption and the dynamic offset range, analyzes the dynamic range between the offset value and the upper and lower limits, and generates a dynamic threshold allocation table based on the time period changes and regional distribution;

[0065] The inventory scheduling optimization module extracts the real-time inventory of each regional storage point based on the dynamic threshold allocation table, analyzes the difference between the inventory and the adjustment threshold, and combines the priority scheduling order of the storage point with the remaining transportation volume to reallocate the inventory and generate a regional inventory scheduling table for paper towels;

[0066] The inventory reserve planning module is based on the regional paper towel inventory scheduling table, extracts the inventory and demand distribution data of each region, identifies the long-term inventory demand, analyzes the deviation between the current inventory and reserve allocation ratio, plans the inventory ratio by time period, and generates a quarterly paper towel inventory reserve plan.

[0067] The regional paper towel consumption distribution data includes the daily time period group statistics, group difference data and group ratio data. The abnormal marking results of paper towel consumption include the consumption fluctuation range of high-frequency time periods, the consumption ratio of low-frequency time periods and the out-of-range time period marking. The paper towel consumption trend distribution data set includes the consumption data of abnormal time periods, the consumption change amplitude and increase and decrease ratio fitting results of continuous time periods. The dynamic threshold allocation table includes the increase and decrease extreme values ​​of the change ratio in the region, the dynamic offset range and the upper and lower limits of the dynamic range analysis results. The regional inventory scheduling table of paper towels includes the real-time inventory of regional storage points, the difference between the inventory and the adjustment threshold and the priority scheduling order of storage points. The quarterly inventory reserve plan of paper towels includes the long-term inventory demand, inventory reserve allocation ratio and time period inventory planning ratio.

[0068] Specifically, if Figure 2 , 3 As shown, the consumption behavior identification module includes:

[0069] Use the record extraction submodule to classify the data according to daily time and location based on the consumption records of paper towel storage points in the area, group the daily consumption records by storage point, extract the time series and spatial location series, and establish a location and time data matching table;

[0070] By setting the parameter range of daily time series and storage point location series, a detailed classification standard is constructed. The specific operations include decomposing the timestamp information in the paper towel consumption record, dividing the 24 hours of a day into fixed time nodes, and extracting the geographical location information of the storage point to ensure that the data classification covers all storage points and time periods. In the classification process, the area number and storage point number are used for marking, and the corresponding relationship between location and time data is established to form an accurate match of daily consumption data. Through data cleaning and deduplication operations, duplicate records and abnormal data are eliminated, such as sudden usage fluctuations caused by non-business hours or extreme weather, to ensure the accuracy of the classification results. According to the sorted daily consumption data, a location and time data matching table is generated, and clear daily time periods and storage point location correspondence records are exported to ensure the smooth implementation of subsequent group statistical analysis.

[0071] The grouping statistical analysis submodule divides the paper towel usage into differentiated daily time periods based on the location and time data matching table, accumulates the paper towel consumption of the storage points within the time period, analyzes the segment proportion, and obtains the consumption group proportion data;

[0072] Accurate data segmentation is achieved through preset time period rules. For example, a day is divided into four time periods: morning peak, noon, evening peak and night. During the time period, the paper towel usage of each storage point is accumulated and counted. The specific operations include aggregating the storage point number of each data in the matching table, accumulating the consumption values ​​of the same storage point in the corresponding time period one by one, and recording the total amount of the segment. The proportion of each time period is calculated based on the accumulation result, and the segment proportion data is accurately output as visual data of paper towel usage patterns. In order to improve the reliability of the analysis, the interference of extreme values ​​on the grouped data, such as abnormal consumption on special holidays, is excluded during trend analysis, and historical data is introduced for verification. The consumption group ratio data is obtained according to the group statistical results, providing a differentiated allocation basis for regional paper towel management.

[0073] The spatiotemporal consumption analysis submodule extracts the consumption ratio characteristics of the storage point within the time period based on the consumption grouping ratio data, associates the consumption ratio with the location data, analyzes the spatiotemporal consumption relationship of the storage points in the region, and obtains the regional paper towel consumption distribution data;

[0074] Taking the storage point number as the primary key, the paper towel consumption ratio within the time period is associated with the corresponding geographical location to generate a spatiotemporal consumption feature table for all storage points in the region. During the association process, the consumption differences of adjacent storage points are analyzed based on the spatial distribution of the storage points, and the regional paper towel replenishment strategy is further optimized. The consumption characteristics within the time period are classified and marked, such as the time period characteristics of high-consumption areas and low-consumption areas, and the matching relationship between peak periods and specific storage points is clarified. The historical spatiotemporal consumption data is referred to, and the interference of abnormal peaks is eliminated to obtain the regional paper towel consumption distribution data. According to the comprehensive analysis results, the dynamic distribution model of paper towel consumption in the region is output to provide a spatiotemporal reference basis for daily management and paper towel consumption prediction, and guide the optimization of regional replenishment strategies.

[0075] Specifically, if Figure 2 , 4 As shown, the abnormal consumption analysis module includes:

[0076] The high- and low-frequency ratio analysis submodule divides the consumption data into time periods based on the regional paper towel consumption distribution data, summarizes the consumption in each time period, analyzes the total consumption ratio of the time period, distinguishes high-frequency and low-frequency time periods, and obtains a high- and low-frequency ratio data set;

[0077] The data is segmented by daily time nodes, for example, a day is divided into four time periods: morning peak, noon, evening peak and night. The specific operations include accumulating the paper towel consumption data in each time period and calculating the total consumption of each time period. Based on the consumption data of all storage points in the area, records are extracted one by one, and the consumption of each time period is accumulated respectively. At the same time, the geographical location and number information corresponding to each storage point are recorded. By calculating the proportion of consumption in each time period to the total consumption, the total consumption proportion of each time period is analyzed, and high-frequency and low-frequency time periods are distinguished. The high-frequency time period is defined as the time period when the total consumption proportion reaches a certain threshold, while the low-frequency time period is the time period with a low total consumption proportion. The high- and low-frequency ratio data set is obtained through statistical analysis to clarify the paper towel consumption pattern in different time periods, providing data support for the subsequent fluctuation range extraction.

[0078] The fluctuation range extraction submodule is based on the high- and low-frequency ratio data set. It arranges the ratio data in the high-frequency time period in time series, divides the data in the continuous time period into multiple fluctuation intervals, analyzes the change trend of the data in each interval, analyzes the change range within the fluctuation range, and obtains the consumption fluctuation characteristics.

[0079] The daily high-frequency time period data is divided into multiple continuous fluctuation intervals, and the data is segmented through the time axis. The data of adjacent time periods are classified into the same interval based on the time nodes, and the total consumption and proportion changes in each interval are calculated. For each fluctuation interval, the change trend of the data is further analyzed, including the growth rate, decrease rate and fluctuation range of the consumption ratio. In the process of change trend analysis, the difference of the continuous proportion values ​​in the time period is calculated to measure the change range of each interval, mark the range of consumption fluctuations, eliminate the interference of extreme values ​​on the results, such as short-term peak consumption caused by special events or holidays, ensure the reliability of the fluctuation range, output the consumption fluctuation characteristics in the high-frequency time period, clarify the changing rules of paper towel consumption in the high-frequency time period, and provide a reference basis for abnormal marking.

[0080] Based on the consumption fluctuation characteristics, the abnormal marking submodule compares the proportion of each time period in the high- and low-frequency ratio data with the consumption fluctuation range item by item, marks the time period where the proportion difference exceeds the fluctuation range as abnormal, records the proportion and difference content of the abnormal time period, and generates the abnormal marking result of paper towel consumption;

[0081] By matching the proportion values ​​in the high- and low-frequency proportion data with the upper and lower limits of the fluctuation range one by one, it is determined whether there is an abnormal situation beyond the range. The specific operations include extracting the proportion data of each time period, calculating the difference between it and the change range of the corresponding fluctuation interval, marking the data points whose difference exceeds the fluctuation range as abnormal, and recording the specific proportion value and deviation value of the abnormal time period. While recording the abnormal data, the time period to which it belongs is numbered, and a complete abnormal marking record table is generated by combining the geographical location of the storage point and other relevant data. To ensure the accuracy of the marking results, the time periods interfered by external special factors, such as special events or temporary emergencies, are eliminated during the comparison process. Finally, the abnormal marking results of paper towel consumption are generated, which will provide an early warning mechanism and optimization strategy for regional paper towel replenishment management, thereby improving management efficiency and resource utilization.

[0082] Specifically, if Figure 2 , 5 As shown, the consumption trend prediction module includes:

[0083] The abnormal marking submodule extracts the consumption data values ​​of the abnormal time period based on the abnormal marking results of paper towel consumption, compares them with the normal consumption range, analyzes the characteristics of the abnormal interval and the data fluctuation characteristics, and obtains the abnormal consumption time period data set;

[0084] The consumption records of each storage point within the abnormal time period are extracted one by one, and compared with the normal consumption range item by item. In the specific implementation, a normal consumption upper and lower limit range model is established to map all data points of the abnormal time period into the range. For records outside the range, they are classified and sorted according to the time period, and the fluctuation range of the consumption value and the concentration of the consumption data in the spatial distribution are recorded. The frequency of change and the increase and decrease trend of the consumption data within the abnormal time period are calculated, and the main features of the sudden increase or decrease in consumption are clarified, such as whether it is concentrated in high-frequency usage areas or whether it shows a continuous abnormal trend. By further analyzing the data of the abnormal time period and combining the fluctuation characteristics of the abnormal data with the spatial location distribution, an abnormal consumption time period data set is finally formed to provide basic data support for subsequent change range statistics.

[0085] The variation range statistics submodule groups the consumption data within the time period based on the abnormal consumption time period data set, counts the consumption difference between each group, and obtains the variation range statistics;

[0086] By aggregating the data points of each abnormal time period according to their storage point numbers, the abnormal consumption records of the same storage point are classified into the same group. Within each group, the consumption records are sorted according to the time series, the consumption differences between adjacent records are counted, and the overall change amplitude within the group is further calculated. The abnormal consumption fluctuation characteristics of the storage point are used as a reference, the data is divided according to the time node interval, the data increase and decrease trends of each time period are recorded one by one, and the difference in change amplitude between different storage points is analyzed. After comprehensive statistics of all grouped data, the change amplitude distribution and change interval characteristics between each group are obtained, such as the situation where the change amplitude is concentrated in a specific time period or a specific storage point. Finally, the change amplitude statistical data is obtained to provide parameter input for subsequent consumption trend fitting.

[0087] The consumption trend fitting submodule is based on the statistical data of the change amplitude, takes the change amplitude and total consumption in the time series as data fitting parameters, performs trend fitting on the consumption change according to the time interval, calls the cumulative characteristics of the increase and decrease amplitude, and combines the fitting results to calculate the predicted paper towel consumption trend value, and obtain the paper towel consumption trend distribution data set;

[0088] Calculate the predicted paper towel consumption trend value according to the formula:

[0089] y=β 0 +β 1 x 1 +β 2 x 2 +β 3 x 3 ;

[0090] Among them, y represents the predicted trend value of paper towel consumption, β 0represents the intercept of the regression equation, β 1 represents the influence coefficient of the magnitude of change in the time series, x 1 Represents the magnitude of change within a specific time interval, β 2 Represents the influence coefficient of total consumption, x 2 Represents the total consumption in a specific time interval, β 3 The weighted adjustment coefficient representing the trend of change, x 3 Represents the consumption difference between the previous time interval and the current time interval;

[0091] Detailed explanation of the formula and the process of formula calculation and derivation:

[0092] The multiple linear regression model is used to estimate the trend of paper towel consumption. This model is often used in statistical analysis to predict the value of a dependent variable (y, i.e., the value of the paper towel consumption trend distribution data set) based on multiple independent variables;

[0093] β 0 : Intercept, reflecting the initial consumption when there is no influence of any independent variable. This value is usually obtained through the training of the linear regression model and is set according to the center point of the data;

[0094] β 1 : The influence coefficient of the change amplitude in the time series, which indicates how the paper towel consumption changes when the change amplitude increases by one unit. This coefficient is obtained by performing linear regression analysis on historical data to ensure that the coefficient value is within the sensitivity of the actual change amplitude;

[0095] x 1 : The change range within a specific time interval is recorded and calculated by the real-time monitoring system. The quantitative standard is the difference between the peak and trough of consumption within the time interval;

[0096] β 2 : The impact coefficient of total consumption, which quantifies the impact of total consumption on the trend of paper towel consumption. It is obtained through historical data model analysis. The coefficient reflects the average impact of changes in total consumption on the consumption trend;

[0097] x 2 : The total amount of consumption within a specific time interval is accumulated through the data collection system;

[0098] β 3 : Weighted adjustment coefficient of the change trend, used to fine-tune the model to be closer to the actual changes, and adjusted according to the model verification results to improve the prediction accuracy;

[0099] x 3 : The difference between the consumption of the previous time interval and the current time interval, calculated as the current interval consumption minus the previous interval consumption.

[0100] Substitute the parameters into the formula for calculation: Set the parameter value to β 0 =50,β 1 =2,β 2 =0.5,β 3 =1.5, assuming a real data scenario, where x 1 =10(variation range), x 2 =200 (total consumption), x 3 =20 (difference in consumption before and after);

[0101] The calculation process is as follows:

[0102] y=50+2×10+0.5×200+1.5×20;

[0103] y = 50 + 20 + 100 + 30 = 200;

[0104] The result y=200 shows that considering the given change amplitude, total consumption, and difference before and after, the predicted paper towel consumption trend data set value is 200, which indicates that the predicted consumption of paper towels has increased significantly during the time period considered. The model effectively reveals the dynamics and patterns of consumption changes and supports more refined inventory and supply chain management decisions.

[0105] Specifically, if Figure 2 , 6 As shown, the dynamic threshold management module includes:

[0106] The extreme value extraction submodule is based on the paper towel consumption trend distribution data set, screens the extreme values ​​in the consumption change ratio in each time period, and counts the consumption trend at the time point where the extreme value is located to obtain the extreme value data;

[0107] By extracting the change ratio of all consumption data within the time period one by one, sorting the change trend within each time period, determining the maximum and minimum consumption ratio values, comparing the data points of each time period with the data of its adjacent time periods one by one, recording the maximum and minimum values ​​of the increase or decrease in the adjacent change ratios, marking the corresponding time points and storage point locations, and correlating the extreme values ​​with the consumption trends in their time periods, such as recording the time synchronization between the extreme values ​​of the consumption change ratio and the peak or valley values ​​of the overall trend, and further combining the geographical location of the storage point and the regularity of the time nodes, the change extreme value data are classified into the corresponding time periods and spatial locations, and finally the change extreme value data covering the regularity of paper towel consumption throughout the day are obtained, providing basic information support for subsequent dynamic offset analysis.

[0108] The dynamic offset analysis submodule analyzes the difference between the consumption and the dynamic offset range based on the extreme value data, counts the dynamic changes of the offset value in each time period, and analyzes the time span and dynamic characteristics of the offset interval to obtain the offset dynamic range data;

[0109] By calculating the difference between the actual consumption in each time period and the reference value within its dynamic offset range, the changing trend of the dynamic offset range is clarified. First, with the extreme value data as a reference, the fluctuation range of the consumption ratio in each time period is analyzed, the difference change between adjacent time periods is calculated, and the increase or decrease direction and numerical amplitude of each offset are recorded. The dynamic offset values ​​of each time period are grouped and the offset changes in each group are counted. The distribution of the changes on the time axis is analyzed. For larger offset values, their impact range is further analyzed, such as whether the time span covers high-frequency time periods or whether it is synchronized with abnormal consumption of certain storage points. Comprehensive statistics are used to obtain offset dynamic range data, including the change amplitude, direction and time span of the offset value, to provide accurate data support for the delineation of dynamic threshold distribution.

[0110] The threshold distribution acquisition submodule extracts the upper and lower limits of the time period offset value based on the offset dynamic range data, divides the dynamic threshold interval according to the time span, sorts out the upper and lower limit allocation rules of the threshold within the time period, and generates a dynamic threshold allocation table based on the consumption trend in the area;

[0111] By analyzing the distribution characteristics of the offset values ​​in each time period one by one, the dynamic changes of the upper and lower limits are clarified. Based on the offset dynamic range data, the maximum and minimum ranges of the offset values ​​in each time period are calculated, and the upper and lower limits are classified and divided according to the time span to form a dynamic threshold interval. According to the distribution characteristics of the upper and lower limits in the time period, the upper and lower limit allocation rules for the thresholds in each time period are sorted out. For example, a narrower offset threshold range is applicable to high-frequency time periods, while a wider threshold range is applicable to low-frequency time periods. Combined with the paper towel consumption trend in the region, the dynamic threshold allocation rules for each time period are comprehensively summarized, and finally a dynamic threshold allocation table covering the entire day's time range is generated, which provides a direct basis for the optimization of the regional paper towel replenishment plan and ensures that the management strategy can dynamically adapt to the characteristics and laws of consumption changes.

[0112] Specifically, if Figure 2 , 7 As shown, the inventory scheduling optimization module includes:

[0113] The inventory data capture submodule verifies the consistency of the inventory data of each regional storage point with the data returned by the real-time interface based on the dynamic threshold allocation table, analyzes the integrity of inventory parameters, and generates a real-time inventory list of regional storage points;

[0114] The inventory data capture submodule verifies the consistency of the inventory data of each regional storage point with the data returned by the real-time interface based on the dynamic threshold allocation table. By calling the inventory data of each storage point returned by the real-time interface, it compares it item by item with the upper and lower limits of the threshold in the dynamic threshold allocation table to ensure that the inventory quantity is within a reasonable range. The integrity of the data returned by the interface is checked, such as checking whether there are missing records or inconsistent inventory parameters, and the geographic location number and inventory record of each storage point are verified one by one to confirm that the records match correctly. When analyzing the integrity of inventory parameters, the storage point number and corresponding time point of abnormal data are extracted, and the storage points with missing or erroneous data are marked. The abnormal values ​​are corrected through historical inventory records to ensure that the generated inventory list is accurate. Finally, the real-time inventory list of regional storage points is output, which contains the current inventory quantity, geographic location number and inventory status of each storage point in the corresponding time period, providing basic data support for inventory difference analysis.

[0115] The inventory difference analysis submodule extracts the threshold range of each storage point based on the real-time inventory list of regional storage points, compares the inventory quantity of the storage point with the corresponding adjustment threshold, identifies the inventory difference of each storage point, sorts the storage points according to the absolute value of the difference, and generates the inventory difference analysis results;

[0116] By comparing the actual inventory in the list with the upper and lower limit values ​​in the dynamic threshold allocation table, the inventory difference of each storage point is identified, and the difference between the inventory of each storage point and the corresponding threshold range is calculated item by item. For storage points with inventory below the lower limit, they are marked as insufficient inventory points, and the absolute value of the difference is recorded. For storage points with inventory above the upper limit, they are marked as excess inventory points, and the absolute value of their difference is also recorded. The difference data of all storage points are sorted from large to small according to the absolute value to give priority to storage points with larger differences. At the same time, the storage point number and location distribution corresponding to the difference ranking are output. For multiple storage points in the same area, the consistency of their inventory status is further compared to identify regional inventory anomalies. Finally, the inventory difference analysis results are generated to provide a priority reference for inventory reallocation.

[0117] The inventory redistribution submodule extracts the priority scheduling order and remaining transportation volume of each storage point based on the inventory difference analysis results, analyzes the distribution of insufficient storage points and excess storage points item by item, allocates excess inventory to insufficient storage points, calculates the inventory volume between storage points, and generates a paper towel regional inventory scheduling table;

[0118] Calculate the inventory between storage points according to the formula

[0119]

[0120] Among them, X i,j represents the inventory quantity that needs to be allocated from storage point i to storage point j, Si represents the current inventory quantity at storage point i, D j represents the inventory quantity required for storage point j, C i represents the transportation capacity of storage point i, T j represents the transportation time, P j represents the inventory priority of storage point j, W 1 Represents the adjustment coefficient in the scheduling algorithm, W 2 Represents the adjustment coefficient in the scheduling algorithm, A i represents the remaining transportation volume of storage point i, B j Represents the change in inventory demand at storage point j;

[0121] Detailed explanation of the formula and the process of formula calculation and derivation:

[0122] This formula is used to calculate the inventory allocation quantity X from storage point i to storage point j i,j ,The results are used to optimize inventory allocation and ensure that the inventory level at each storage point meets demand;

[0123] S i : The current inventory quantity of storage point i is obtained through real-time monitoring of the inventory management system. For example, set S i =500 units;

[0124] D j : The demand for storage point j is calculated by the sales forecast model. For example, set D j =700 units;

[0125] C i : The transportation capacity of storage point i, which indicates the maximum transportation volume that can be handled per unit time, is obtained through the analysis of historical transportation data. For example, set C i = 1000 units / day;

[0126] T j : The transportation time from storage point i to storage point j is obtained through the transportation time data of the logistics system. For example, set T j =2 days;

[0127] P j : The inventory priority of storage point j. The higher the value, the higher the priority. It is set by business strategy and is usually quantified based on factors such as sales importance and customer demand. For example, set P j =1.5;

[0128] W 1 : Adjustment coefficient, used to adjust the impact of the remaining transportation volume of storage point i on the allocation volume. It is set according to historical data and business needs, reflecting the importance of the remaining transportation volume. For example, set W 1 =0.6;

[0129] W 2 : Adjustment coefficient, used to adjust the impact of the inventory demand change of storage point j on the allocation quantity. It is set according to historical data and business needs, reflecting the importance of demand changes. For example, set W 2 =0.4;

[0130] A i : The remaining transportation volume of storage point i, which means the transportation capacity that storage point i can still use for allocation under the current transportation plan, is obtained through real-time transportation plan data. For example, set A i =300 units;

[0131] B j : The change in inventory demand at storage point j, which indicates the increase or decrease in current demand relative to the previous demand forecast, is obtained through sales data and market analysis. For example, if B j =200 units;

[0132] Substitute the parameters into the formula for calculation:

[0133] Calculate the absolute value of the inventory difference: |S i -D j |=|500-700|=200;

[0134] Calculate the square of the product of transit time and priority: (T j ·P j ) 2 =(2·1.5) 2 =3 2 =9;

[0135] Calculate the transport capacity multiplied by the above result: C i ·(T j ·P j ) 2 =1000·9=9000;

[0136] Calculate the square root of the above result:

[0137] Calculate the sum of the product of the adjustment coefficient and the corresponding parameter: W 1 ·A i +W 2 ·B j =0.6·300+0.4·200=180+80=260;

[0138] Substitute the above results into the formula to calculate the final inventory allocation:

[0139] Result Xi,j ≈548.6 indicates that allocating about 549 units of inventory from storage point i to storage point j helps to meet the demand of storage point j and optimize the overall inventory allocation.

[0140] Specifically, if Figure 2 , 8 As shown, the inventory reserve planning module includes:

[0141] The inventory distribution analysis submodule matches regional and inventory demand data based on the paper towel regional inventory scheduling table, verifies data integrity, removes abnormal values, reorganizes regional inventory and demand data, and generates regional inventory and demand distribution data tables;

[0142] By comparing the regional inventory with the corresponding demand data one by one, we ensure that the inventory data and demand data in each region correspond one by one. At the same time, we check the integrity of the inventory and demand data, extract abnormal values ​​in the inventory scheduling table, such as extreme inventory or data that obviously deviates from the regional demand average, and compare them with historical records and demand distribution data in neighboring areas to correct or eliminate unreasonable values. In the case of missing data, we perform supplementary calculations in combination with regional historical data to ensure the accuracy of the generated regional inventory and demand distribution data. The cleaned and sorted regional inventory data and demand data are reclassified according to the regional number to generate an inventory and demand distribution data table covering all regions, which includes the real-time inventory of each region, the corresponding demand and the cleaned data status, providing reliable data input for subsequent reserve deviation analysis.

[0143] The reserve deviation analysis submodule extracts the inventory and long-term demand of each region based on the regional inventory and demand distribution data table, compares the data according to regional classification, calculates the difference between regional inventory and demand, converts the difference into reserve allocation deviation parameters, sorts regional deviations, and generates a reserve allocation ratio deviation table;

[0144] By comparing the regional inventory data with the long-term demand data in the corresponding time period one by one, the difference between the inventory and demand in each region is calculated, and the difference is converted into a reserve allocation deviation parameter. In the calculation process, the average demand in the historical data is statistically analyzed for each region as a reference value for the long-term demand. At the same time, the difference data between the inventory and demand is normalized to facilitate subsequent sorting analysis. For areas with large differences, their demand patterns and inventory distribution characteristics are further analyzed. For example, whether there is insufficient inventory in areas with long-term high demand, or excess inventory in areas with large demand fluctuations. The calculated deviation parameters are sorted from large to small according to the absolute value, and the areas with large deviations are marked. The reserve allocation ratio deviation table is output to clearly show the reserve deviations between regions and provide an optimization basis for inventory ratio planning.

[0145] The inventory ratio planning submodule extracts regional inventory demand distribution data based on the reserve allocation ratio deviation table, matches reserve demand with current inventory, adjusts reserve ratios and inter-regional coordination, and generates a quarterly inventory reserve plan for paper towels;

[0146] By analyzing the reserve demand and current inventory status of each region in the deviation table, the reserve ratio is adjusted region by region to achieve coordination between regions. Priority is given to adjusting the reserves of regions with larger absolute values ​​of deviation parameters. For regions with insufficient inventory, the reserve ratio is increased and inventory is allocated from regions with excess inventory in combination with their demand growth trend and long-term demand forecast. For regions with excess inventory, inventory space is released by reducing the reserve ratio, and the reserve strategy is adjusted according to the demand law of the region. For example, for regions with long-term high demand but small fluctuations, the reserve ratio can be appropriately increased, while regions with large demand fluctuations can set flexible reserve ranges. The reserve coordination data between regions is comprehensively sorted out to generate a quarterly inventory reserve plan for paper towels to ensure that the inventory and demand in each region are reasonably matched, providing a basis for long-term optimization of inventory management.

[0147] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A paper towel consumption prediction and management system, characterized in that: The system comprises: The consumption behavior recognition module extracts the paper towel usage and storage point location data based on the consumption records of paper towel storage points in the region, counts the daily paper towel consumption in each time period, analyzes the grouping differences and proportions, and obtains the regional paper towel consumption distribution data; The abnormal consumption analysis module analyzes the consumption fluctuation range of high-frequency time periods based on the regional paper towel consumption distribution data, identifies the time period where the ratio difference marks the out-of-range period, and generates a paper towel consumption abnormality marking result; The consumption trend prediction module identifies the paper towel consumption data in the abnormal time period based on the abnormal paper towel consumption marking result, calculates the change range and increase / decrease ratio, fits the consumption, and generates a paper towel consumption trend distribution data set; The dynamic threshold management module extracts the extreme value of the change ratio increase or decrease in the area based on the paper towel consumption trend distribution data set, calculates the difference between the consumption and the dynamic offset range, analyzes the offset value and the upper and lower limit dynamic range, and generates a dynamic threshold allocation table; The inventory scheduling optimization module extracts the real-time inventory of each regional storage point based on the dynamic threshold allocation table, analyzes the difference between the inventory and the adjustment threshold, reallocates the inventory, and generates a paper towel regional inventory scheduling table; The inventory reserve planning module identifies long-term inventory demand based on the paper towel regional inventory scheduling table, analyzes the deviation between the current inventory and the reserve allocation ratio, plans the inventory ratio by time period, and generates a quarterly inventory reserve plan for paper towels.

2. The paper towel consumption prediction and management system according to claim 1, characterized in that: The regional paper towel consumption distribution data includes daily time period grouping statistics, grouping difference data and grouping ratio data; the paper towel consumption abnormal marking results include high-frequency time period consumption fluctuation range, low-frequency time period consumption ratio and out-of-range time period mark; the paper towel consumption trend distribution data set includes abnormal time period consumption data, continuous time period consumption change amplitude and increase / decrease ratio fitting results; the dynamic threshold allocation table includes the increase / decrease extreme value of the change ratio in the region, the dynamic offset range and the upper and lower limits of the dynamic range analysis results; the paper towel regional inventory scheduling table includes the real-time inventory of regional storage points, the difference between the inventory and the adjustment threshold and the priority scheduling order of the storage points; the paper towel quarterly inventory reserve plan includes the long-term inventory demand, the inventory reserve allocation ratio and the time period inventory planning ratio.

3. The paper towel consumption prediction and management system according to claim 1, characterized in that: The consumption behavior identification module includes: Use the record extraction submodule to classify the data according to daily time and location based on the consumption records of paper towel storage points in the area, group the daily consumption records by storage point, extract the time series and spatial location series, and establish a location and time data matching table; The grouping statistical analysis submodule divides the paper towel usage into differentiated daily time periods based on the location and time data matching table, accumulates the paper towel consumption of the storage points within the time period, analyzes the segment proportions, and obtains the consumption group proportion data; The spatiotemporal consumption analysis submodule extracts the consumption ratio characteristics within the storage point time period based on the consumption grouping ratio data, associates the consumption ratio with the location data, analyzes the spatiotemporal consumption relationship of the storage points in the area, and obtains the regional paper towel consumption distribution data.

4. The paper towel consumption prediction and management system according to claim 1, characterized in that: The abnormal consumption analysis module includes: The high- and low-frequency ratio analysis submodule divides the consumption data into time periods based on the regional paper towel consumption distribution data, summarizes the consumption in each time period, analyzes the total consumption ratio of the time period, distinguishes high-frequency and low-frequency time periods, and obtains a high- and low-frequency ratio data set; The fluctuation range extraction submodule arranges the proportion data in the high-frequency time period in time series based on the high- and low-frequency ratio data set, divides the data in the continuous time period into multiple fluctuation intervals, performs change trend analysis on the data in each interval, analyzes the change amplitude within the fluctuation range, and obtains the consumption fluctuation characteristics; Based on the consumption fluctuation characteristics, the abnormal marking submodule compares the proportion of each time period in the high- and low-frequency ratio data with the consumption fluctuation range item by item, marks the time period where the proportion difference exceeds the fluctuation range as abnormal, records the proportion and difference content of the abnormal time period, and generates the abnormal marking result of paper towel consumption.

5. The paper towel consumption prediction and management system according to claim 1, characterized in that: The consumption trend prediction module includes: The abnormal marking submodule extracts the consumption data value of the abnormal time period based on the abnormal marking result of the paper towel consumption, compares it with the normal consumption range, analyzes the characteristics of the abnormal interval and the data fluctuation characteristics, and obtains the abnormal consumption time period data set; The variation range statistics submodule groups the consumption data within the time period based on the abnormal consumption time period data set, counts the consumption difference between each group, and obtains variation range statistics; The consumption trend fitting submodule is based on the variation amplitude statistical data, takes the variation amplitude and total consumption in the time series as data fitting parameters, performs trend fitting on the consumption changes according to the time interval, calls the cumulative characteristics of the increase and decrease amplitude, and combines the fitting results to calculate the predicted paper towel consumption trend value, and obtain the paper towel consumption trend distribution data set.

6. The paper towel consumption prediction and management system according to claim 5, characterized in that: Calculate the predicted paper towel consumption trend value according to the formula: y=β0+β1x1+β2x2+β3x3; Among them, y represents the predicted trend value of paper towel consumption, β0 represents the intercept of the regression equation, β1 represents the influence coefficient of the change amplitude in the time series, x1 represents the change amplitude in a specific time interval, β2 represents the influence coefficient of the total consumption, x2 represents the total consumption in a specific time interval, β3 represents the weighted adjustment coefficient of the change trend, and x3 represents the difference in consumption between the previous time interval and the current time interval.

7. The paper towel consumption prediction and management system according to claim 1, characterized in that: The dynamic threshold management module includes: The variation extreme value extraction submodule screens the extreme values ​​in the consumption change ratio in each time period based on the paper towel consumption trend distribution data set, and calculates the consumption trend at the time point where the extreme value is located to obtain variation extreme value data; The dynamic offset analysis submodule analyzes the difference between the consumption and the dynamic offset range based on the extreme value data, counts the dynamic changes of the offset value in each time period, and analyzes the time span and dynamic characteristics of the offset interval to obtain the offset dynamic range data; The threshold distribution acquisition submodule extracts the upper and lower limits of the time period offset value based on the offset dynamic range data, divides the dynamic threshold interval according to the time span, sorts out the upper and lower limit allocation rules of the threshold within the time period, and generates a dynamic threshold allocation table based on the consumption trend in the area.

8. The paper towel consumption prediction and management system according to claim 1, characterized in that: The inventory scheduling optimization module includes: The inventory data capture submodule checks the consistency of the inventory data of each regional storage point with the data returned by the real-time interface based on the dynamic threshold allocation table, analyzes the integrity of inventory parameters, and generates a real-time inventory list of the regional storage point; The inventory difference analysis submodule extracts the threshold range of each storage point based on the real-time inventory list of the regional storage points, compares the storage point inventory with the corresponding adjustment threshold, identifies the inventory difference of each storage point, sorts the storage points according to the absolute value of the difference, and generates the inventory difference analysis result; Based on the inventory difference analysis results, the inventory reallocation submodule extracts the priority scheduling order and remaining transportation volume of each storage point, analyzes the distribution of insufficient storage points and excess storage points item by item, allocates excess inventory to insufficient storage points, calculates the inventory volume between storage points, and generates a paper towel area inventory scheduling table.

9. The paper towel consumption prediction and management system according to claim 8, characterized in that: Calculate the inventory between the storage points according to the formula Among them, X i,j represents the inventory quantity that needs to be allocated from storage point i to storage point j, S i represents the current inventory quantity at storage point i, D j represents the inventory quantity required for storage point j, C i represents the transportation capacity of storage point i, T j represents the transportation time, P j represents the inventory priority of storage point j, W1 represents the adjustment coefficient in the scheduling algorithm, W2 represents the adjustment coefficient in the scheduling algorithm, and A i represents the remaining transportation volume of storage point i, B j Represents the change in inventory demand at storage point j.

10. The paper towel consumption prediction and management system according to claim 1, characterized in that: The inventory reserve planning module includes: The inventory distribution analysis submodule matches the regional and inventory demand data based on the paper towel regional inventory scheduling table, verifies the data integrity, removes abnormal values, reorganizes the regional inventory and demand data, and generates a regional inventory and demand distribution data table; The reserve deviation analysis submodule extracts the inventory and long-term demand of each region based on the regional inventory and demand distribution data table, compares the data according to regional classification, calculates the difference between regional inventory and demand, converts the difference into a reserve allocation deviation parameter, sorts the regional deviations, and generates a reserve allocation ratio deviation table; The inventory ratio planning submodule extracts regional inventory demand distribution data based on the reserve allocation ratio deviation table, matches reserve demand with current inventory, adjusts reserve ratio and inter-regional coordination, and generates a quarterly paper towel inventory reserve plan.

Images