Data-driven WMS storage optimization method and system

Abstract

The invention discloses a data-driven WMS storage optimization method and system, and relates to the technical field of storage management, and the method comprises the steps: obtaining the actual inventory number of a WMS system of a middle library material and the theoretical inventory number of an ERP system at the end of each preset statistical period, and calculating the inventory difference rate; continuously collecting inventory difference rates of a plurality of statistical periods to form a difference rate time sequence; calculating a difference direction consistency coefficient and a progressive trend change rate of the difference rate time sequence; when the difference direction consistency coefficient and the gradual trend change rate both exceed respective threshold values, judging that the material has systematic deviation; for the material with systematic deviation, calculating a BOM revision coefficient according to the average value of the difference rate time sequence, and generating a BOM revision suggested value; according to the method, automatic and reasonable constraint on the BOM revision amplitude is realized through a difference rate revision technology combining a difference rate average value, a statistical period number and a confidence coefficient adjustment coefficient.

Description

Technical Field

[0001] This invention relates to the field of warehouse management technology, and more specifically, to a data-driven WMS warehouse optimization method and system. Background Technology

[0002] In modern manufacturing enterprises' warehouse management, the Warehouse Management System (WMS) and the Enterprise Resource Planning (ERP) system are two core information systems. The WMS system is mainly responsible for recording the actual inbound and outbound operations of materials and real-time inventory levels, while the ERP system calculates and manages theoretical inventory based on Bill of Materials (BOM) quotas. Ideally, the inventory data in the two systems should be consistent, but in actual operation, there are often continuous discrepancies between them.

[0003] The causes of such inventory discrepancies are complex and varied. On the one hand, they may stem from random factors such as natural material loss, operational errors, and inventory deviations. On the other hand, they may also be due to systematic deviations caused by inaccurate BOM unit quota settings. The BOM unit quota refers to the theoretical quantity of a certain material required to produce one unit of finished product. This value is usually determined during the product design phase, but in actual production, due to factors such as process improvements, equipment adjustments, and changes in operating habits, the actual consumption may deviate from the initial set value.

[0004] Currently, most companies use periodic inventory checks to identify inventory discrepancies. When these discrepancies accumulate to a certain level, the causes are manually analyzed, and a decision is made on whether to adjust the BOM (Bill of Materials) quotas. This traditional method has the following problems: It usually only attracts attention when discrepancies significantly impact production or inventory management, by which time substantial inventory distortion and management costs may have already occurred. Furthermore, manual analysis struggles to effectively distinguish between random discrepancies and systematic biases, potentially misinterpreting normal fluctuations as BOM problems or attributing genuine BOM deviations to operational errors. Summary of the Invention

[0005] The purpose of this invention is to provide a data-driven WMS warehouse optimization method and system to solve the problems existing in the background art.

[0006] The above-mentioned technical objective of the present invention is achieved through the following technical solution: Firstly, this application provides a data-driven WMS warehouse optimization method, comprising the following steps: Step S1: At the end of each preset statistical period, obtain the actual inventory quantity of intermediate warehouse materials in the WMS system and the theoretical inventory quantity in the ERP system, and calculate the inventory difference rate. Step S2: Collect inventory difference rates for multiple statistical periods continuously to form a time series of difference rates; Step S3: Calculate the consistency coefficient of the difference direction and the rate of change of the asymptotic trend of the difference rate time series; Step S4: When both the consistency coefficient of the difference direction and the rate of change of the gradual trend exceed their respective thresholds, it is determined that the material has a systematic deviation; for materials with systematic deviations, the BOM revision coefficient is calculated based on the average value of the difference rate time series, and a BOM revision recommendation value is generated. Step S5: Verify the BOM revision recommendation value and complete the revision. Collect the inventory difference rate for multiple statistical periods after the revision, calculate the average difference rate change before and after the revision, and determine the revision effect.

[0007] Based on the above technical solution, the present invention can be further improved as follows.

[0008] Furthermore, obtaining the theoretical inventory quantity and the actual inventory quantity in step S1 above includes: Obtain the actual inventory quantity of materials at the end of the statistical period from the WMS system; Obtain the beginning inventory quantity, incoming quantity, completed quantity, and BOM unit quota of materials from the ERP system within the statistical period; The theoretical inventory quantity is obtained by subtracting the sum of the products of all completed quantities and the corresponding BOM unit quota from the sum of the beginning inventory quantity and the quantity received.

[0009] Furthermore, the calculation of the consistency coefficient in the direction of difference in step S3 above includes: For any two adjacent periods in the difference rate time series, subtract the inventory difference rate of the previous period from the inventory difference rate of the later period, and calculate the difference rate change series accordingly. Traverse the sequence of changes in difference rate, count the number of changes greater than zero (recorded as positive values), and count the number of changes less than zero (recorded as negative values). Calculate the difference between the number of positive values ​​and the number of negative values, take the absolute value of the difference, and divide the absolute value by the total number of elements in the difference rate change sequence to obtain the difference direction consistency coefficient.

[0010] Furthermore, the calculation of the gradual trend change rate in step S3 above includes: When the consistency coefficient of the difference direction is greater than 0.6, calculate the average rate of change of the difference rate change sequence; Extract the inventory difference rate of the first statistical period from the difference rate time series, and denote it as the initial difference rate; Dividing the average rate of change by the initial rate of difference yields the gradual trend rate of change.

[0011] Furthermore, step S4 above includes the following steps: Step S41: When the consistency coefficient of the difference direction is greater than the corresponding judgment threshold and the absolute value of the gradual trend change rate is greater than the corresponding judgment threshold, it is determined that the material has a systematic deviation and enters the subsequent revision process; when either condition is not met, it is determined that the material does not have a systematic deviation and the processing flow of the material ends. Step S42: For materials with systematic deviations, calculate the BOM revision coefficient based on the average value of the difference rate time series, and generate the BOM revision recommendation value.

[0012] Furthermore, step S42 above includes the following steps: For materials with systematic deviations, iterate through all inventory difference rates in the difference rate time series and calculate the arithmetic mean, which is recorded as the average difference rate. The total number of statistical periods included in the statistical difference rate time series; When the total number of statistical periods is less than six, the confidence adjustment factor is set to 0.5; when the total number of statistical periods is greater than or equal to six and less than or equal to twelve, the confidence adjustment factor is set to 0.7; when the total number of statistical periods is greater than twelve, the confidence adjustment factor is set to 1.0.

[0013] Furthermore, step S43 above also includes the following steps: The product of the average difference rate and the confidence adjustment coefficient is used as the weighted difference rate; Calculate the standard deviation of the difference rate between each inventory variance rate and the corresponding weighted variance rate; The data fluctuation coefficient is obtained by dividing the standard deviation of the variance rate by the absolute value of the average variance rate. If the data volatility coefficient is greater than 0.3, the weighted difference rate is multiplied by the volatility suppression factor of 0.8 to obtain the revised difference rate; otherwise, the weighted difference rate is used as the revised difference rate. The BOM revision coefficient is calculated using the revision difference rate to generate the BOM revision recommendation value.

[0014] Furthermore, the above-mentioned calculation of the BOM revision coefficient using the revision difference rate to generate the BOM revision recommendation value includes: The negative of the revision difference rate is taken as the initial revision coefficient; Determine if the absolute value of the preliminary revision coefficient exceeds 0.5. If it exceeds 0.5, limit the preliminary revision coefficient to ±0.5 to obtain the BOM revision coefficient; if it does not exceed 0.5, use the preliminary revision coefficient as the BOM revision coefficient. Multiply the unit price in the BOM by one and the sum of the BOM revision factor to obtain the preliminary revised price; The preliminary revised quotas are processed for precision based on the material's unit of measurement attribute to obtain the BOM revision recommendation value.

[0015] Furthermore, step S5 above includes the following steps: Step S51: After verifying the BOM revision suggestion value, update the BOM unit quota of the material in the ERP system; Step S52: After the BOM revision is completed, collect the inventory difference rate for multiple statistical periods after the revision; Step S53: Sum the revised inventory difference rates for multiple statistical periods and calculate the revised average difference rate; Step S54: Take the absolute values ​​of the average difference rate before revision and the average difference rate after revision respectively, and subtract the two absolute values ​​to obtain the improvement amount; divide the improvement amount by the absolute value of the average difference rate before revision to obtain the improvement range of the difference rate. Step S55: Determine the revision effect by the improvement range of the difference rate, and associate and store the material code, the unit quota of BOM before and after the revision, the average difference rate before and after the revision, the improvement range of the difference rate, and the revision effect to form a revision effect record.

[0016] Secondly, this application provides a data-driven WMS warehouse optimization system, applied to any one of the data-driven WMS warehouse optimization methods in the first aspect, characterized in that it includes: The discrepancy calculation module is used to obtain the actual inventory quantity of intermediate warehouse materials in the WMS system and the theoretical inventory quantity in the ERP system at the end of each preset statistical period, and to calculate the inventory discrepancy rate. The sequence construction module is used to continuously collect inventory difference rates for multiple statistical periods to form a time series of difference rates; The trend analysis module is used to calculate the coefficient of consistency of the direction of difference and the rate of change of the gradual trend in the time series of difference rates. The deviation determination module is used to determine that there is a systematic deviation in the material when both the consistency coefficient of the difference direction and the rate of change of the gradual trend exceed their respective thresholds; for materials with systematic deviations, the BOM revision coefficient is calculated based on the average value of the difference rate time series, and a BOM revision recommendation value is generated. The effect verification module is used to verify the BOM revision suggestions and complete the revision, collect the inventory difference rate for multiple statistical periods after the revision, calculate the average difference rate change before and after the revision, and determine the revision effect.

[0017] Thirdly, this application provides an electronic device, including: at least one processor, at least one memory, and a data bus; In this system, the processor and memory communicate with each other via a data bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the method as described in any of the first aspects.

[0018] Fourthly, this application provides a non-transitory computer-readable storage medium that stores computer instructions that cause a computer to perform any of the methods in the first aspect.

[0019] Compared with the prior art, the present invention has at least the following beneficial effects: This invention establishes a quantifiable and traceable closed-loop calibration mechanism by using the time-series difference between actual inventory in the WMS and theoretical inventory in the ERP system as the driving force, thereby significantly improving the accuracy and reliability of material quota management and warehouse operations. First, by periodically collecting and constructing a time series of difference rates, scattered inventory deviations can be transformed into continuous trend information, facilitating the identification of long-term systematic deviations rather than short-term fluctuations. Second, by introducing two judgment indicators—the consistency coefficient of difference direction and the rate of change of gradual trend—it achieves filtering based on both directionality and rate, reducing misjudgments caused by occasional anomalies or noise, ensuring that BOM revisions are triggered only when the direction is consistent and the trend is significant, thus avoiding the chain risks caused by blind adjustments. Furthermore, by employing a confidence adjustment coefficient and a volatility suppression mechanism, data with different sample sizes and volatility levels are weighted and suppressed, balancing prudence in small sample cases with full utilization of historical information in large sample cases, enhancing the robustness of revision conclusions. Simultaneously, by setting an upper limit on the revision coefficient and combining it with measurement unit precision processing, it ensures the effectiveness of the revision while also considering the operability of process and production execution.

[0020] This invention also incorporates a quantitative comparison of the difference rate before and after revision into closed-loop verification. It assesses the revision effect by evaluating the magnitude of improvement and records the complete change trajectory, supporting post-audit and strategy optimization, thus forming a data-driven continuous improvement process. Overall, this method can reduce inaccurate material preparation, frequent material requisition adjustments, and inventory backlog caused by BOM quota deviations, thereby reducing material costs and operational fluctuations, and improving the accuracy of production planning and supply chain coordination. Furthermore, through automated data collection, judgment, and recording, it reduces the burden of manual verification, shortens the decision-making cycle, and establishes a scalable quota management standard and rule base for enterprises, promoting the long-term implementation of lean inventory management and precise cost control. Attached Figure Description

[0021] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the steps of the warehouse optimization method in an embodiment of the present invention; Figure 2 This is a data flow diagram of the warehouse optimization system in an embodiment of the present invention; Figure 3 This is a time series trend chart of the inventory difference rate in an embodiment of the present invention; Figure 4 This is a schematic diagram of the modules of the warehouse optimization system in an embodiment of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0023] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0024] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0025] In the description of the embodiments of the present invention, "multiple" means at least two.

[0026] Example: This example provides a data-driven WMS warehouse optimization method, such as... Figure 1 As shown, the method includes the following steps: Step S1: At the end of each preset statistical period, obtain the actual inventory quantity of the intermediate warehouse materials in the WMS system and the theoretical inventory quantity in the ERP system, and calculate the inventory difference rate.

[0027] Step S2: Collect inventory difference rates for multiple statistical periods to form a time series of difference rates.

[0028] Step S3: Calculate the consistency coefficient of the difference direction and the rate of change of the asymptotic trend of the difference rate time series.

[0029] Step S4: When both the consistency coefficient of the difference direction and the rate of change of the gradual trend exceed their respective thresholds, it is determined that the material has a systematic deviation; for materials with systematic deviations, the BOM revision coefficient is calculated based on the average value of the difference rate time series, and a BOM revision recommendation value is generated.

[0030] Step S5: Verify the BOM revision recommendation value and complete the revision; collect the inventory difference rate for multiple statistical periods after the revision, calculate the average difference rate change before and after the revision, and determine the revision effect.

[0031] See Figure 2 This paper illustrates the overall structure of a data-driven WMS warehouse optimization system according to an embodiment of the present invention. The system is located between the WMS system and the ERP system. It receives actual inventory quantities from the WMS via the main data stream and obtains initial inventory quantities, incoming quantities, completed quantities, and BOM unit quotas from the ERP system for calculating theoretical inventory and conducting inventory variance analysis.

[0032] like Figure 2 As shown, the system contains multiple processing modules, including: a data receiving and preprocessing module, a difference calculation and trend analysis module, a systematic deviation determination module, and a BOM revision suggestion value generation module. The data receiving and preprocessing module unifies the format of data from multiple systems, filters anomalies, and completes fields. The difference calculation and trend analysis module identifies the pattern of difference changes based on the difference between actual and theoretical inventory, combined with statistical periodic sequences. The systematic deviation determination module determines whether the difference is caused by random fluctuations or long-term deviations. If a systematic deviation exists, the generation module calculates the revision coefficient based on the revision difference rate and generates a BOM revision suggestion value.

[0033] Figure 2 The system's data storage structure is also illustrated, used to store key data before and after revision, including material codes, quotas before and after revision, changes in variance rates, and evaluation of revision effectiveness. This repository also receives historical data feedback from statistical periods, enabling the system to continuously track inventory discrepancies after revisions and achieve a closed-loop optimization based on real data. Finally, the system-generated BOM revision recommendations are input into the ERP system to update the BOM unit quotas, achieving automatic quota calibration and management process optimization.

[0034] Furthermore, obtaining the theoretical and actual inventory quantities in step S1 includes: Obtain the actual inventory quantity of materials at the end of the statistical period from the WMS system; In one embodiment, a data collection task is triggered at the end of a pre-set statistical period (e.g., by day, by shift, or by week). During task execution, the system reads the real-time inventory of each material registered in the intermediate warehouse according to the material code through the interface service of the WMS system or the database query mechanism. This inventory is usually automatically accumulated by warehousing actions such as shelving, requisitioning, and returning materials, without the need for manual re-entry.

[0035] For example, if a material code is A1001, and at the end of a certain statistical period, the WMS system shows that its inventory quantity is 125 units, the system will directly read and use "125" as the actual inventory quantity for that period.

[0036] Obtain the beginning inventory quantity, incoming quantity, completed quantity, and BOM unit quota of materials from the ERP system within the statistical period; In one embodiment, the system reads the initial inventory quantity of each target material from the ERP system at the start of the statistical period and uses it as the baseline value for calculating the theoretical inventory for that period. Simultaneously, the system also accumulates the quantity of the material received within the period based on purchase receipts, production receipts, and other receipt-type documents recorded in the ERP system. Furthermore, based on the completion reports of production orders, the system calculates the total quantity of all completed work using that material and simultaneously reads the corresponding BOM unit quota for that material in the ERP system for subsequent theoretical consumption calculations.

[0037] For example, if the initial inventory of material A1001 in a certain statistical period is 80 units, the purchase of goods during the period increases by 50 units, and the finished quantity of products made from this material as raw material is 10 units, while its unit quota in the BOM is 3 units, then the system will record the initial inventory of 80, the quantity purchased is 50, the finished quantity is 10, and the BOM unit quota is 3 units, as inputs for the theoretical inventory calculation.

[0038] The theoretical inventory quantity is obtained by subtracting the sum of the products of all completed quantities and the corresponding BOM unit quota from the sum of the beginning inventory quantity and the quantity received.

[0039] In one embodiment, after data collection, the initial inventory quantity is summed with the cumulative inbound quantity within the statistical period to form the total available quantity for production consumption and other uses. Subsequently, based on all completed quantities recorded in the ERP system, the unit quota of the material in the corresponding product's BOM is multiplied to obtain the theoretical consumption during the period. If the material is used by multiple products simultaneously, the product of the completed quantity and unit quota is calculated for each relevant product, and these products are summed to obtain the total theoretical consumption. Finally, the theoretical consumption is subtracted from the aforementioned total available quantity to obtain the theoretical inventory quantity.

[0040] For example, the initial inventory of material A1001 is 80 units, the quantity received is 50 units, and the total available quantity is 130 units. If 10 units of product P1 and 5 units of product P2 are completed during this period, and the unit quota of this material in P1 is 3 and the unit quota of P2 is 2, then the theoretical consumption is (10×3)+(5×2)=40 units, and the theoretical inventory quantity is 130-40=90 units.

[0041] Of particular importance is that the calculation of the inventory discrepancy rate in step S1 includes: The inventory difference is obtained by subtracting the theoretical inventory from the actual inventory quantity. The ratio of the inventory difference to the theoretical inventory quantity is used as the inventory difference rate. When the theoretical inventory quantity is 0, skip the calculation of the inventory difference rate for that statistical period; The positive or negative sign of the inventory difference rate indicates the direction of deviation from the BOM quota. A positive value indicates that the actual inventory is more than the theoretical inventory, suggesting that the BOM unit quota may be too high. A negative value indicates that the actual inventory is less than the theoretical inventory, suggesting that the BOM unit quota may be too low.

[0042] Furthermore, the calculation of the consistency coefficient in the direction of difference in step S3 includes: For any two adjacent periods in the difference rate time series, subtract the inventory difference rate of the previous period from the inventory difference rate of the later period, and calculate the difference rate change series accordingly. In one embodiment, a pre-constructed time series of difference rates is obtained. This series consists of inventory difference rates for multiple consecutive statistical periods, arranged chronologically. Starting from the first element of the series, using a sliding window of "adjacent periods," the difference rates of the previous and next periods are sequentially extracted, and a difference operation is performed: the difference rate of the next period is subtracted from the difference rate of the previous period to obtain the change in difference rate for that pair of periods. This operation is repeated until all adjacent period pairs in the series are traversed, thereby generating a sequence of difference rate changes with a length of "the length of the difference rate time series minus one."

[0043] For example, if the time series of the difference rate is 0.05, 0.08, 0.10, 0.07, then the system calculates the changes sequentially: 0.08 - 0.05 = 0.03, 0.10 - 0.08 = 0.02, 0.07 - 0.10 = -0.03, thus forming the change series 0.03, 0.02, -0.03. It should be noted that the change series reflects the direction and magnitude of the difference rate change, serving as the basis for subsequent judgments of trend consistency, rather than being directly used to determine the magnitude of the deviation.

[0044] Traverse the sequence of changes in difference rates, count the number of changes greater than zero (recorded as positive values), and count the number of changes less than zero (recorded as negative values). In one embodiment, after obtaining the sequence of changes, the system iterates through the sequence item by item and evaluates the sign of each change through conditional judgment. If a change is greater than zero, the system increments the positive counter; if a change is less than zero, the system increments the negative counter. Changes equal to zero are not included in these two statistical results because they do not reflect directional characteristics. After completing the traversal of the entire sequence of changes, the system obtains two indicators: the number of positive values ​​and the number of negative values, which reflect the overall directionality of the sequence.

[0045] For example, for the change sequence 0.03, 0.02, -0.03, the system identifies 0.03 and 0.02 as positive values ​​and counts them twice, while -0.03 is a negative value and is counted once. Therefore, the number of positive values ​​is 2 and the number of negative values ​​is 1.

[0046] Calculate the difference between the number of positive values ​​and the number of negative values, take the absolute value of the difference, and divide the absolute value by the total number of elements in the difference rate change sequence to obtain the difference direction consistency coefficient.

[0047] In one embodiment, after obtaining the number of positive and negative values, the system first calculates the difference between them to reflect the degree of imbalance between positive and negative changes. Then, the system takes the absolute value of this difference, ensuring that the final result is unaffected by the sign of the change (more positive than negative) or vice versa, representing only the degree of directional consistency. Next, the system calculates the total number of elements in the change sequence, using this as the normalized denominator, and divides the aforementioned absolute difference by this total number to obtain the difference direction consistency coefficient. This coefficient ranges from 0 to 1, with values ​​closer to 1 indicating greater directional consistency.

[0048] For example, if the number of positive values ​​is 2, the number of negative values ​​is 1, and the total number of changes is 3, then the difference is 2-1=1, the absolute value is 1, and the coefficient of consistency of the direction of difference is 1÷3≈0.33.

[0049] Furthermore, the calculation of the gradual trend rate of change in step S3 includes: When the consistency coefficient of the direction of difference is greater than 0.6, the rate of change of the gradual trend is calculated; In one embodiment, after calculating the consistency coefficient of the difference direction, it is first determined whether the coefficient is strictly greater than 0.6; only when this condition is met does the subsequent calculation process of the asymptotic trend rate of change proceed. This determination step is to ensure that the sequence has sufficient consistency in direction, so that the subsequent ratio based on the average rate of change to the initial difference rate is statistically significant.

[0050] For example, if the coefficient of consistency of difference direction calculated for a certain material is 0.72 (>0.6), the system will continue to calculate the average rate of change of the difference rate change sequence of the material and calculate the gradual trend change rate accordingly.

[0051] If the consistency coefficient is 0.58 (≤0.6), the system stops calculating the gradual trend change rate and determines that the material does not yet meet the trend judgment conditions for systematic deviation.

[0052] Calculate the average rate of change of the difference rate sequence; In one embodiment, the system first filters effective elements on the generated sequence of difference rate changes (removing changes that are explicitly zero and defined as "no direction"—as mentioned in the aforementioned method, zero values ​​are not included in the direction statistics, but can be selected to be included in the average rate of change calculation according to the design; in this example, the arithmetic mean of all changes is taken to reflect the overall rate of change), and then adds each element in the change sequence in chronological order and divides it by the number of change elements to obtain the arithmetic mean change value of the sequence. This average value is the average rate of change (the unit is "difference rate / statistical period").

[0053] For example, if the change sequence is 0.03, 0.02, -0.03, then the average rate of change = (0.03 + 0.02 - 0.03) ÷ 3 ≈ 0.0067 (that is, an average increase of 0.0067 per statistical period).

[0054] It should be noted that if you want to reduce the impact of random extreme values, you can first perform preprocessing on the change series such as median substitution, truncated average, or weighted average. However, these preprocessings are implementation details and do not change the overall method of dividing the numerator by the initial difference rate.

[0055] Extract the inventory difference rate of the first statistical period from the difference rate time series, and denote it as the initial difference rate; In one embodiment, the system directly reads the first element of the pre-constructed difference rate time series as the initial difference rate (denoted as ) in this trend calculation. This value is used to normalize the average rate of change, thus obtaining the relative asymptotic rate of change.

[0056] For example, if the time series of the difference rate is 0.05, 0.08, 0.10, 0.13, then the initial difference rate... =0.05.

[0057] It should be noted that when the initial difference rate is zero or very close to zero, directly using... Using the denominator can lead to division by zero or extreme amplification effects, so this needs to be addressed in the implementation. zero or (For example, a very small threshold that can be set by the system) In cases where the gradual trend change rate cannot be calculated, a constraint strategy is adopted (e.g., the material is deemed to require manual verification or the gradual trend change rate cannot be calculated) to avoid numerical instability without changing the core logic of the original algorithm.

[0058] Dividing the average rate of change by the initial rate of difference yields the gradual trend rate of change.

[0059] In one embodiment, the calculated average rate of change (denoted as ) () is the dividend, and the initial difference rate of the first period of the difference rate time series is used as the divisor. If the divisor is used, perform division directly: Rate of change of asymptotic trend = ÷ The ratio retains its sign to reflect the directionality of the trend (positive values ​​indicate the average rate of increase relative to the initial rate of difference, and negative values ​​indicate the average rate of decrease), and can be used to compare with a preset threshold (e.g., 0.15 as defined in S4) to determine whether there is a significant gradual change.

[0060] For example, continuing with the previous example, the average rate of change 0.0067 and initial difference rate =0.05, then the gradual trend change rate is approximately 0.0067 ÷ 0.05 ≈ 0.134; if the absolute value of this value is greater than 0.15, it means that the judgment threshold has been reached (in this example, it has not been exceeded).

[0061] Furthermore, step S4 includes the following steps: Step S41: Set the threshold for determining the consistency coefficient of the difference direction to 0.6, and set the threshold for determining the gradual trend change rate to 0.15; In one embodiment, two key thresholds are set for the trend determination module: the threshold for determining the consistency coefficient of the difference direction is fixed at 0.6, and the threshold for determining the gradual trend change rate is fixed at 0.15. These two thresholds are determined comprehensively based on the enterprise's historical data statistical experience, the characteristics of difference fluctuations, and the stability factors of the production organization structure. They are used to ensure that materials are marked as having systematic deviations only when the "direction is sufficiently consistent" and the "change rate is significant". The system writes these two thresholds into the configuration table during the initialization phase and directly reads the corresponding thresholds for determination when performing trend analysis.

[0062] For example, when a company analyzed historical data on the difference rate over the past three years, it found that when the consistency coefficient of the difference direction exceeded 0.6, the probability of the material difference direction shifting increased significantly. At the same time, if the absolute value of the gradual trend change rate exceeded 0.15, it usually meant that the material difference was increasing or decreasing period by period, with a clear trend. Therefore, these two values ​​were used as the final threshold.

[0063] Step S42: When the consistency coefficient of the difference direction is greater than 0.6 and the absolute value of the gradual trend change rate is greater than 0.15, it is determined that the material has a systematic deviation and enters the subsequent revision process; when either condition is not met, it is determined that the material does not have a systematic deviation and the processing flow of the material ends. In one embodiment, after obtaining the consistency coefficient of the direction of difference and the rate of change of the gradual trend, a dual-condition joint judgment is performed in sequence; if and only if the consistency coefficient of the direction of difference is strictly greater than 0.6 and the absolute value of the rate of change of the gradual trend is strictly greater than 0.15, the current material is marked as "having a systematic deviation" and automatically enters the calculation process of BOM revision; otherwise, the system determines that the material does not have obvious directional deviation or trend deviation, thereby ending the processing flow of the material and waiting for the data of the next statistical period to participate in the analysis again.

[0064] For example, if a material has a consistency coefficient of 0.73 and an asymptotic trend change rate of 0.22 over eight consecutive statistical periods, the system meets both conditions and determines that the material has a systematic deviation and enters the revision process. If the consistency coefficient is 0.81 but the asymptotic trend change rate is only 0.12, the trend change rate does not reach the threshold and the system will not enter the revision process.

[0065] Step S43: For materials with systematic deviations, calculate the BOM revision coefficient based on the average value of the difference rate time series, and generate the BOM revision recommendation value.

[0066] In one embodiment, when the system confirms a systematic deviation in a material, it automatically enters the BOM revision factor calculation stage. First, it iterates through the time series of difference rates and performs an arithmetic mean on all difference rates to obtain the average difference rate. Then, it determines the confidence adjustment factor based on the number of statistical periods and multiplies the average difference rate by the confidence adjustment factor to obtain the weighted difference rate. Next, it calculates the standard deviation of the difference rate for the series and determines whether the data fluctuation coefficient exceeds 0.3. If it does, it multiplies the weighted difference rate by a fluctuation suppression factor of 0.8 to obtain the revised difference rate; otherwise, it directly uses the weighted difference rate as the revised difference rate. Finally, it uses the negative of the revised difference rate as the preliminary revision factor and performs a limit processing with an absolute value upper limit of ±0.5 to obtain the final BOM revision factor. This factor is added to 1 and multiplied by the original BOM unit quota to obtain the preliminary revised quota. Then, this quota is processed according to the material measurement accuracy to produce the BOM revision suggestion value.

[0067] For example, if the average variance rate of a certain material is 0.12 and the confidence adjustment coefficient is 0.7, then the weighted variance rate is 0.084. If the data fluctuation coefficient is 0.25 (<0.3), the system obtains a revised variance rate of 0.084. Taking its opposite number gives an initial revision coefficient of -0.084 (not exceeding ±0.5). Substituting this coefficient into the original quota calculation will generate the BOM revision recommendation value.

[0068] Furthermore, step S43 includes the following steps: For materials with systematic deviations, iterate through all inventory difference rates in the difference rate time series, calculate the arithmetic mean, and denot it as the average difference rate; In one embodiment, for materials determined to have systematic deviations, the system first iterates through its corresponding difference rate time series, sequentially reading the inventory difference rate data for each statistical period and performing a summation operation during the reading process. After all difference rates have been included in the calculation, the sum of the difference rates is divided by the number of statistical periods to obtain the arithmetic mean difference rate of the material within the selected time range. This average difference rate reflects the overall offset direction and magnitude of the inventory deviation of the material and is an important gain factor for subsequent calculation of the BOM revision recommendation value.

[0069] For example, if the inventory difference rates of a certain material for six consecutive statistical periods are 0.10, 0.12, 0.09, 0.15, 0.11 and 0.13 respectively, the system will sum them up to get 0.70, and divide by 6 to get an average difference rate of approximately 0.1167.

[0070] The total number of statistical periods included in the statistical difference rate time series; When the total number of statistical periods is less than six, the confidence adjustment factor is set to 0.5; when the total number of statistical periods is greater than or equal to six and less than or equal to twelve, the confidence adjustment factor is set to 0.7; when the total number of statistical periods is greater than twelve, the confidence adjustment factor is set to 1.0.

[0071] In one embodiment, different confidence adjustment coefficients are selected based on different ranges of the total number of statistical periods to scale the average difference rate when generating subsequent BOM revision recommendations. Specifically, when the total number of statistical periods is less than six, the statistical stability is weak due to the small sample size; therefore, the system sets the confidence adjustment coefficient to 0.5 to reduce the impact of bias caused by insufficient samples on subsequent revision results. When the number of statistical periods is greater than or equal to six and less than or equal to twelve, the data volume reaches a moderate level; at this time, the system sets the confidence adjustment coefficient to 0.7 to balance the revision strength and data stability. When the number of statistical periods is greater than twelve, the sample data volume is sufficient and the period coverage is relatively complete; therefore, the confidence adjustment coefficient is set to 1.0, indicating complete trust in the average difference rate's reflection of systematic bias.

[0072] For example, if a material has recorded the difference rate for 14 periods, the system will set the confidence adjustment factor to 1.0, so that the average difference rate will not be attenuated.

[0073] See Figure 3This displays the inventory variance rate curves for a specific material over multiple statistical periods, including two trend lines for data before and after revision. The horizontal axis represents the statistical period number, and the vertical axis represents the inventory variance rate, indicating the degree of deviation between the actual and theoretical inventory in each period. Figure 3 As shown, the inventory discrepancy rate gradually increased between statistical periods 1 and 7, reaching approximately 0.16 in period 7. This indicates that the BOM quota for this material was consistently too high or too low before the revision, causing the theoretical inventory to deviate continuously from the actual inventory. The average discrepancy rate before the revision was approximately 0.116, which can be used as the baseline value before the revision. The red dashed line indicates the BOM revision point (between period 7 and period 8). After the system performed the revision and the quota was updated by ERP, the discrepancy rate decreased significantly starting from period 8, immediately falling back to approximately 0.05, and continued to decrease and stabilize in subsequent periods. The discrepancy rate was approximately 0.02 in period 12, and the average discrepancy rate after the revision was approximately 0.022, showing a significant improvement compared to before the revision.

[0074] Figure 3 The data shows the trend of inventory discrepancy rate before and after the BOM revision, indicating that the revised quotas are closer to actual consumption levels. The average discrepancy rate before and after the revision can be used to further calculate the improvement, evaluate the effectiveness of the revision, and create a record of the revision effect, providing data for subsequent optimization.

[0075] Furthermore, step S43 also includes the following steps: The product of the average difference rate and the confidence adjustment coefficient is used as the weighted difference rate; In one embodiment, the system multiplies the previously obtained average difference rate by the corresponding confidence adjustment coefficient to generate a weighted difference rate. This weighted difference rate combines the overall deviation of the difference rate with the reliability of the data sample size.

[0076] For example, if the average variance rate of a certain material is 0.12 and the confidence adjustment coefficient is 0.7, then the weighted variance rate is calculated as 0.12 × 0.7 = 0.084. This processing method suppresses the weighted variance rate when the sample period is short, while allowing the weighted variance rate to more comprehensively reflect the long-term deviation trend when the period is long.

[0077] Calculate the standard deviation of the difference rate between each inventory variance rate and the corresponding weighted variance rate; In one embodiment, to assess the dispersion of the difference rate relative to the weighted difference rate for each statistical period, the difference between each inventory difference rate and the weighted difference rate is calculated individually, and the standard deviation of these differences is calculated to obtain the standard deviation of the difference rate. This standard deviation can quantify the degree of fluctuation of the difference rate with the period, providing a basis for determining whether fluctuation suppression of the weighted difference rate is necessary.

[0078] For example, if a material has three inventory variance rates: 0.10, 0.15, and 0.09, and the weighted variance rate is 0.084, the system will calculate the deviation values ​​of (0.10–0.084), (0.15–0.084), and (0.09–0.084), and calculate the standard deviation based on these deviation values.

[0079] The data fluctuation coefficient is obtained by dividing the standard deviation of the variance rate by the absolute value of the average variance rate. In one embodiment, the standard deviation of the variance rate is divided by the absolute value of the average variance rate to obtain the data volatility coefficient. This data volatility coefficient can more clearly reflect the proportion of the volatility amplitude relative to the deviation trend itself, and is used to determine the stability of the variance rate time series.

[0080] For example, if the standard deviation of the variance rate is 0.03 and the absolute value of the average variance rate is 0.12, then the data fluctuation coefficient is 0.03 ÷ 0.12 = 0.25.

[0081] If the data volatility coefficient is greater than 0.3, the weighted difference rate is multiplied by the volatility suppression factor of 0.8 to obtain the revised difference rate; otherwise, the weighted difference rate is used as the revised difference rate. In one embodiment, the system determines whether to suppress the weighted difference rate based on the data volatility coefficient. When the data volatility coefficient is greater than 0.3, it indicates that the difference rate fluctuates significantly with the period, which may cause the weighted difference rate to deviate from reality when directly used for revision. In this case, the system multiplies the weighted difference rate by a volatility suppression factor of 0.8 to generate the revision difference rate, thereby reducing the risk of deviation in revision results under high-volatility data environments. If the data volatility coefficient is less than or equal to 0.3, the system directly uses the weighted difference rate as the revision difference rate.

[0082] For example, if the weighted variance rate is 0.084 and the data volatility coefficient is 0.35, then the revised variance rate is 0.084 × 0.8 = 0.0672; if the volatility coefficient is 0.20, then the revised variance rate is directly taken as 0.084.

[0083] The BOM revision coefficient is calculated using the revision difference rate to generate the BOM revision recommendation value.

[0084] In one embodiment, the aforementioned revision difference rate is used as a core parameter and applied to the current BOM unit quota calculation model to form the BOM revision coefficient. Generally, the BOM revision coefficient can be expressed as 1 plus the revision difference rate, so that the revision result can reflect the positive or negative direction of the deviation and the revision magnitude. After generating the BOM revision coefficient, the system multiplies it by the original BOM unit quota to obtain a new BOM revision suggestion value, which serves as an important basis for subsequent manual review or automatic system comparison.

[0085] For example, if the original BOM unit quota for a certain material is 5.0 and the revision difference rate is 0.0672, then the BOM revision coefficient is 1.0672, and the corresponding BOM revision suggestion value is 5.0 × 1.0672 = 5.336.

[0086] Furthermore, the BOM revision coefficient is calculated using the revision difference rate, and the generated BOM revision recommendation values ​​include: The negative of the revision difference rate is taken as the initial revision coefficient; In one embodiment, after obtaining the revision difference rate, its sign is reversed to form an initial revision factor for adjusting the BOM unit quota direction. Since higher actual inventory usually means lower BOM quotas, and lower inventory usually means higher quotas, a revision direction opposite to the deviation direction is needed to achieve accurate correction.

[0087] For example, if the revision difference rate of a certain material is 0.0672, then the initial revision coefficient is -0.0672; if the revision difference rate is -0.05, then the initial revision coefficient is 0.05.

[0088] Determine if the absolute value of the preliminary revision coefficient exceeds 0.5. If it exceeds 0.5, limit the preliminary revision coefficient to ±0.5 to obtain the BOM revision coefficient; if it does not exceed 0.5, use the preliminary revision coefficient as the BOM revision coefficient. In one embodiment, a magnitude judgment is performed on the initial revision coefficient to ensure that the generated BOM revision result is not over-adjusted due to statistical fluctuations or abnormal deviations. When the absolute value of the initial revision coefficient is greater than 0.5, the system truncates it to ±0.5; if the absolute value is less than or equal to 0.5, it is directly used as the BOM revision coefficient. This magnitude limiting logic can effectively avoid abnormal expansion of the difference rate due to abnormal inventory records, large-scale scrapping, or special operating conditions, keeping the BOM revision within a controllable range.

[0089] For example, if the initial revision coefficient is -0.0672, the lower limit determination does not trigger truncation, and the BOM revision coefficient remains -0.0672; if an analysis yields an initial revision coefficient of 0.73, the system will limit it to 0.5 as the final BOM revision coefficient.

[0090] It should be noted that the upper limit of ±0.5 comes from the long-term experience of manufacturing companies and is intended to ensure that the BOM adjustment range does not exceed ±50% and to maintain the stability of process parameters.

[0091] Multiply the unit price in the BOM by one and the sum of the BOM revision factor to obtain the preliminary revised price; In one embodiment, the current BOM unit quota for the material is combined with a revision factor to calculate the preliminary revised quota. Specifically, 1 is added to the BOM revision factor to form a revision ratio factor, which is then multiplied by the existing BOM unit quota to obtain the preliminary revised quota. This method ensures that the revised result maintains a strictly linear relationship with the revision magnitude, achieving rigorous proportional correction.

[0092] For example, if the original BOM unit quota for a material is 5.0 and the BOM revision factor is -0.0672, then the revision ratio factor is 1-0.0672=0.9328, and the corresponding preliminary revised quota is 5.0×0.9328=4.664.

[0093] The preliminary revised quotas are processed for precision based on the material's unit of measurement attribute to obtain the BOM revision recommendation value.

[0094] In one embodiment, after the initial revised quota calculation is completed, precision adjustments are performed based on the material's unit of measurement attribute to ensure that the revised BOM quota meets the measurement specifications acceptable to actual production and the ERP system. Different materials have different precision requirements for their units of measurement. For example, weight-based materials may require three decimal places, liquid materials require two decimal places, and bulk materials require rounding. According to the preset unit of measurement precision rules in the material master data, the initial revised quota is rounded or rounded down / up to obtain the final BOM revision recommendation value.

[0095] For example, if the initial revised quota is 4.664 and the measurement unit of the material requires two decimal places, the system will process the value as 4.66; if the material is a whole piece, the system will round 4.664 to 5.

[0096] Of particular importance, the precision processing includes: When the unit of measurement is "piece", "set", "unit", or "set", the preliminary revised quota is rounded up; when the unit of measurement is "kg", "g", or "t", the preliminary revised quota is retained to three decimal places; when the unit of measurement is "m", "cm", or "mm", the preliminary revised quota is retained to two decimal places; when the unit of measurement is "L" or "mL", the preliminary revised quota is retained to two decimal places. Furthermore, step S5 includes the following steps: Step S51: After verifying the BOM revision suggestion value, update the BOM unit quota of the material in the ERP system; In one embodiment, after the system generates a suggested BOM revision value and it undergoes manual review or automatic rule verification, the suggested revision value is synchronized to the ERP system, correspondingly updating the unit BOM quota for that material. The system typically performs the write operation through the ERP's standard or customized interface to ensure that the revised value takes effect immediately in modules such as production planning, material requisition, and cost accounting.

[0097] For example, if the suggested revision value for material A is 4.66 and its original quota is 5.00, after the approval, the system will update the corresponding BOM quota field for material A in the ERP to 4.66, so that all subsequent production material requisition logic will be executed based on the new quota.

[0098] Step S52: After the BOM revision is completed, collect the inventory difference rate for multiple statistical periods after the revision; In one embodiment, after the revised BOM value officially takes effect, the system will continue to calculate the inventory discrepancy rate of the material and collect data based on the actual inventory quantity in the WMS and the theoretical inventory quantity in the ERP for multiple subsequent statistical periods. By collecting data from multiple statistical periods, it is possible to assess whether the revised BOM unit quota has brought the inventory discrepancy rate to a reasonable range.

[0099] For example, if the statistical period is set to once a day, the system will collect data for, for example, 10 consecutive days after the revision and generate 10 difference rate data points for effect evaluation.

[0100] Step S53: Sum the revised inventory difference rates for multiple statistical periods and calculate the revised average difference rate; In one embodiment, after collecting difference rate data for multiple statistical periods following the revision, the system sequentially sums these difference rates and then divides them by the number of statistical periods to obtain the revised average difference rate. This average difference rate can serve as an important indicator of the revision effect, representing whether the deviation after revision has improved compared to before revision.

[0101] For example, if the revised difference rates are 0.03, 0.05, 0.01, and -0.02 respectively, the system will calculate (0.03+0.05+0.01-0.02)÷4=0.0175, resulting in a revised average difference rate of 0.0175.

[0102] Step S54: Take the absolute values ​​of the average difference rate before revision and the average difference rate after revision respectively, and subtract the two absolute values ​​to obtain the improvement amount; divide the improvement amount by the absolute value of the average difference rate before revision to obtain the improvement range of the difference rate. In one embodiment, to further quantify the revision effect, the system takes the absolute value of the average difference rate before revision and the absolute value of the average difference rate after revision, and calculates the difference between the two to obtain the improvement amount. Then, dividing the improvement amount by the absolute value of the average difference rate before revision yields the improvement margin of the difference rate, which is used to measure the relative improvement of the inventory difference rate after revision compared to before revision.

[0103] For example, if the average difference rate before revision is 0.12 and the average difference rate after revision is 0.0175, then the improvement is |0.12|-|0.0175|=0.1025; the improvement in the difference rate is 0.1025÷0.12≈0.854, that is, the improvement is 85.4%.

[0104] It should be noted that when the average difference rate before revision is very close to zero, the calculation results may be proportionally amplified. Therefore, the system usually only executes this calculation logic when there is a systematic bias.

[0105] Step S55: Determine the revision effect by the improvement range of the difference rate, and associate and store the material code, the unit quota of BOM before and after the revision, the average difference rate before and after the revision, the improvement range of the difference rate, and the revision effect to form a revision effect record.

[0106] In one embodiment, the system determines whether the revision has achieved the expected results based on the improvement in the difference rate; for example, an improvement exceeding 30% can be considered sufficient to determine if the revision is effective. The system then associates and stores the material code, the unit price in the pre-revision BOM, the unit price in the post-revision BOM, the average difference rate before revision, the average difference rate after revision, the improvement in the difference rate, and the final revision effect, generating a systematic record of the revision effect. This record can be used for long-term analysis of revision patterns, to assist in optimizing future revision strategies, and for review and quality control.

[0107] For example, if the improvement of a certain material reaches 85.4%, the system will determine it as "revision effective" and generate a complete record entry to be stored in the revision effect database.

[0108] See Figure 4 The present invention also provides a data-driven WMS warehouse optimization system 100 for executing the above-described data-driven WMS warehouse optimization method, wherein the data-driven WMS warehouse optimization system 100 includes: The difference calculation module 101 is used to obtain the actual inventory quantity of intermediate warehouse materials in the WMS system and the theoretical inventory quantity in the ERP system at the end of each preset statistical period, and calculate the inventory difference rate. The sequence construction module 102 is used to continuously collect inventory difference rates for multiple statistical periods to form a time series of difference rates; Trend analysis module 103 is used to calculate the consistency coefficient of the difference direction and the rate of change of the gradual trend of the difference rate time series; The deviation determination module 104 is used to determine that the material has a systematic deviation when both the consistency coefficient of the difference direction and the rate of change of the gradual trend exceed their respective thresholds; for materials with systematic deviations, the BOM revision coefficient is calculated based on the average value of the difference rate time series, and a BOM revision suggestion value is generated. The effect verification module 105 is used to verify the BOM revision suggestion value and complete the revision; collect the inventory difference rate of multiple statistical periods after the revision, calculate the average difference rate change before and after the revision, and determine the revision effect.

[0109] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0110] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0111] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0112] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0113] Those skilled in the art will understand that all or part of the steps in the above facts and methods can be implemented by a program instructing related hardware. The program or the program described therein can be stored in a computer-readable storage medium. When the program is executed, it includes the following steps: at this time, the corresponding method steps are introduced. The storage medium can be ROM / RAM, magnetic disk, optical disk, etc.

[0114] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A data-driven WMS warehouse optimization method, characterized in that, Includes the following steps: Step S1: At the end of each preset statistical period, obtain the actual inventory quantity of intermediate warehouse materials in the WMS system and the theoretical inventory quantity in the ERP system, and calculate the inventory difference rate. Step S2: Collect inventory difference rates for multiple statistical periods continuously to form a time series of difference rates; Step S3: Calculate the consistency coefficient of the difference direction and the rate of change of the asymptotic trend of the difference rate time series; Step S4: When both the consistency coefficient of the difference direction and the rate of change of the gradual trend exceed their respective thresholds, it is determined that the material has a systematic deviation; for materials with systematic deviations, the BOM revision coefficient is calculated based on the average value of the difference rate time series, and a BOM revision recommendation value is generated. Step S5: Verify the BOM revision recommendation value and complete the revision. Collect the inventory difference rate for multiple statistical periods after the revision, calculate the average difference rate change before and after the revision, and determine the revision effect.

2. The data-driven WMS warehouse optimization method according to claim 1, characterized in that, The acquisition of theoretical and actual inventory quantities in step S1 includes: Obtain the actual inventory quantity of materials at the end of the statistical period from the WMS system; Obtain the beginning inventory quantity, incoming quantity, completed quantity, and BOM unit quota of materials from the ERP system within the statistical period; The theoretical inventory quantity is obtained by subtracting the sum of the products of all completed quantities and the corresponding BOM unit quota from the sum of the beginning inventory quantity and the quantity received.

3. The data-driven WMS warehouse optimization method according to claim 1, characterized in that, The calculation of the consistency coefficient of the difference direction in step S3 includes: For any two adjacent periods in the difference rate time series, subtract the inventory difference rate of the previous period from the inventory difference rate of the later period, and calculate the difference rate change series accordingly. Traverse the sequence of changes in difference rate, count the number of changes greater than zero (recorded as positive values), and count the number of changes less than zero (recorded as negative values). Calculate the difference between the number of positive values ​​and the number of negative values, take the absolute value of the difference, and divide the absolute value by the total number of elements in the difference rate change sequence to obtain the difference direction consistency coefficient.

4. The data-driven WMS warehouse optimization method according to claim 3, characterized in that, The calculation of the gradual trend change rate in step S3 includes: When the consistency coefficient of the difference direction is greater than 0.6, calculate the average rate of change of the difference rate change sequence; Extract the inventory difference rate of the first statistical period from the difference rate time series, and denote it as the initial difference rate; Dividing the average rate of change by the initial rate of difference yields the gradual trend rate of change.

5. The data-driven WMS warehouse optimization method according to claim 4, characterized in that, Step S4 includes the following steps: Step S41: When the consistency coefficient of the difference direction is greater than the corresponding judgment threshold and the absolute value of the gradual trend change rate is greater than the corresponding judgment threshold, it is determined that the material has a systematic deviation and enters the subsequent revision process; when either condition is not met, it is determined that the material does not have a systematic deviation and the processing flow of the material ends. Step S42: For materials with systematic deviations, calculate the BOM revision coefficient based on the average value of the difference rate time series, and generate the BOM revision recommendation value.

6. The data-driven WMS warehouse optimization method according to claim 5, characterized in that, Step S42 includes the following steps: For materials with systematic deviations, iterate through all inventory difference rates in the difference rate time series and calculate the arithmetic mean, which is recorded as the average difference rate. The total number of statistical periods included in the statistical difference rate time series; When the total number of statistical periods is less than six, the confidence adjustment factor is set to 0.5; when the total number of statistical periods is greater than or equal to six and less than or equal to twelve, the confidence adjustment factor is set to 0.7; when the total number of statistical periods is greater than twelve, the confidence adjustment factor is set to 1.

0.

7. The data-driven WMS warehouse optimization method according to claim 6, characterized in that, Step S43 further includes the following steps: The product of the average difference rate and the confidence adjustment coefficient is used as the weighted difference rate; Calculate the standard deviation of the difference rate between each inventory variance rate and the corresponding weighted variance rate; The data fluctuation coefficient is obtained by dividing the standard deviation of the variance rate by the absolute value of the average variance rate. If the data volatility coefficient is greater than 0.3, the weighted difference rate is multiplied by the volatility suppression factor of 0.8 to obtain the revised difference rate; otherwise, the weighted difference rate is used as the revised difference rate. The BOM revision coefficient is calculated using the revision difference rate to generate the BOM revision recommendation value.

8. The data-driven WMS warehouse optimization method according to claim 7, characterized in that, The step of calculating the BOM revision coefficient using the revision difference rate and generating the BOM revision recommendation value includes: The negative of the revision difference rate is taken as the initial revision coefficient; Determine if the absolute value of the preliminary revision coefficient exceeds 0.

5. If it exceeds 0.5, limit the preliminary revision coefficient to ±0.5 to obtain the BOM revision coefficient; if it does not exceed 0.5, use the preliminary revision coefficient as the BOM revision coefficient. Multiply the unit price in the BOM by one and the sum of the BOM revision factor to obtain the preliminary revised price; The preliminary revised quotas are processed for precision based on the material's unit of measurement attribute to obtain the BOM revision recommendation value.

9. The data-driven WMS warehouse optimization method according to claim 7, characterized in that, Step S5 includes the following steps: Step S51: After verifying the BOM revision suggestion value, update the BOM unit quota of the material in the ERP system; Step S52: After the BOM revision is completed, collect the inventory difference rate for multiple statistical periods after the revision; Step S53: Sum the revised inventory difference rates for multiple statistical periods and calculate the revised average difference rate; Step S54: Take the absolute values ​​of the average difference rate before revision and the average difference rate after revision respectively, and subtract the two absolute values ​​to obtain the improvement amount; divide the improvement amount by the absolute value of the average difference rate before revision to obtain the improvement range of the difference rate. Step S55: Determine the revision effect by the improvement range of the difference rate, and associate and store the material code, the unit quota of BOM before and after the revision, the average difference rate before and after the revision, the improvement range of the difference rate, and the revision effect to form a revision effect record.

10. A data-driven WMS warehouse optimization system, applied to the data-driven WMS warehouse optimization method according to any one of claims 1-9, characterized in that, include: The discrepancy calculation module is used to obtain the actual inventory quantity of intermediate warehouse materials in the WMS system and the theoretical inventory quantity in the ERP system at the end of each preset statistical period, and to calculate the inventory discrepancy rate. The sequence construction module is used to continuously collect inventory difference rates for multiple statistical periods to form a time series of difference rates; The trend analysis module is used to calculate the coefficient of consistency of the direction of difference and the rate of change of the gradual trend in the time series of difference rates. The deviation determination module is used to determine that there is a systematic deviation in the material when both the consistency coefficient of the difference direction and the rate of change of the gradual trend exceed their respective thresholds; for materials with systematic deviations, the BOM revision coefficient is calculated based on the average value of the difference rate time series, and a BOM revision recommendation value is generated. The effect verification module is used to verify the BOM revision suggestions and complete the revision, collect the inventory difference rate for multiple statistical periods after the revision, calculate the average difference rate change before and after the revision, and determine the revision effect.

Images