A multi-module dynamic collaborative type goods preparation demand calculation method

Abstract

The application discloses a multi-module dynamic collaborative type goods preparation demand calculation method, comprising the following steps: a data acquisition module collects sales forecast, current inventory and future arrival plan data, and generates a plurality of week times of net demand sequences; a parameter dynamic configuration module adjusts key parameters for goods preparation demand calculation, and the key parameters are changed and synchronized to a goods preparation demand collaborative calculation module; the goods preparation demand collaborative calculation module receives the net demand sequences and the key parameters, and performs the following steps: aligning the effective period of each transportation mode with the net demand sequence; determining the priority order of the transportation mode based on the transportation priority dynamic rule, distributing the demand interval for each transportation mode by using the interval locking algorithm, and calculating the goods preparation quantity; based on the total sum of the goods preparation quantity and the net demand of the corresponding week time, calculating the demand gap rate, and triggering reverse adjustment when the demand gap rate exceeds the set threshold; and synchronizing the calculation result to a business visualization system. The application realizes the automatic and accurate calculation of goods preparation demand.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of material requirement planning, and in particular to a multi-module dynamic collaborative inventory demand calculation method. BACKGROUND

[0002] In the practice of MRP (Material Requirement Planning) inventory management, enterprises need to handle multi-dimensional data such as sales forecast, inventory level, transportation scheme, etc. However, the existing technology has the following core defects:

[0003] 1. Module data fragmentation, serious time sequence dislocation: In the existing scheme, the function modules such as demand calculation, transportation time efficiency configuration, and safety stock management are independent, and the data needs to be manually transcribed and transmitted. For example, the demand calculation module outputs "monthly demand total", while the transportation module needs "weekly transportation plan", and time sequence dislocation (such as misallocating "3rd week demand" to "4th week transportation") is easy to occur in the manual conversion process, resulting in early or late inventory, according to industry statistics, the error rate is as high as 18%-25%.

[0004] 2. Poor adaptability of key parameters due to static solidification: Transportation time efficiency (head L / T), safety stock and other parameters are stored as fixed values and cannot be dynamically adjusted according to actual scenarios. For example: large goods continue to use the transportation time efficiency of standard parts (actually need an additional 2-3 weeks), or safety stock is not expanded with seasonal demand peaks, resulting in a 30% or more increase in out-of-stock rate during peak season and a 20% increase in inventory cost during off-season.

[0005] 3. Disconnection between transportation priority and demand allocation: Transportation mode priority is only used as a static sorting (such as "air transport > fast ship > slow ship"), and is not linked with demand cycle. High-cost transportation modes (such as air transport) are often used to cover long-term demand (accounting for more than 40%), resulting in high transportation costs; while low-cost transportation modes (such as slow ships) are used for short-term demand, causing more than 35% of the risk of out-of-stock.

[0006] 4. Lack of full-process collaboration, response lag: The full process from "demand update → parameter adjustment → inventory recalculation" needs to be triggered manually, with a response cycle of 24-48 hours. In the scenario of cross-border e-commerce peak season and other demand fluctuations, the lagging inventory plan will miss the replenishment window, according to research, such losses account for 5%-8% of the annual revenue of enterprises.

[0007] The essence of the above problems is the lack of dynamic collaboration mechanism between modules, and the existing technology can only optimize single function points, and cannot form a closed loop of "data-parameters-decision", which needs to be broken through by deep coupling of multiple modules. SUMMARY

[0008] In view of this, the purpose of the present application is to propose a multi-module dynamic collaborative inventory demand calculation method, which is applied to cross-border e-commerce, manufacturing and other scenarios that need to comprehensively handle the complex relationship of "demand fluctuation-transportation time limit-inventory buffer-cost optimization", and realizes the automatic and accurate calculation of inventory demand through a multi-module dynamic collaboration mechanism, solving the problems of module isolation, response lag, and cost and time limit imbalance in traditional solutions.

[0009] In order to achieve the above technical purpose, the technical scheme adopted by the present application is:

[0010] The present application provides a multi-module dynamic collaborative inventory demand calculation method, which realizes real-time data transmission and logic linkage between the data acquisition module, the parameter dynamic configuration module and the inventory demand collaborative calculation module through the data bus, and the method comprises the following steps:

[0011] Step 1, the data acquisition module automatically collects sales forecast data, current inventory data and future arrival plan data every week, generates a net demand sequence of multiple weeks with a time dimension unified as weeks based on the collected data, and triggers a net demand sequence recalculation when the inventory fluctuation exceeds a set threshold;

[0012] Step 2, the parameter dynamic configuration module configures and dynamically adjusts the key parameters for inventory demand calculation, including the head transportation time limit matrix, safety stock and transportation priority dynamic rules, and all key parameter changes are synchronized in real time to the inventory demand collaborative calculation module through the data bus;

[0013] Step 3, the inventory demand collaborative calculation module receives the net demand sequence and key parameters, and performs the following three-layer collaborative logic to calculate the inventory demand:

[0014] Step 31, align the effective period of each transportation mode with the net demand sequence;

[0015] Step 32, determine the priority order of the transportation mode based on the transportation priority dynamic rule, allocate demand intervals for each transportation mode in priority order using the interval locking algorithm, and calculate the inventory based on the net demand amount of the allocated demand interval;

[0016] Step 33, based on the calculated total inventory and the net demand of the corresponding week, calculate the demand gap rate of each week, and trigger temporary reverse adjustment of the key parameters when the demand gap rate exceeds a set threshold;

[0017] Step 4, the inventory demand collaborative calculation module synchronizes the calculation results to the business visualization system through the data bus.

[0018] Further, the data collection module in step 1 automatically collects sales forecast data, current inventory data and future arrival plan data every week, generates a future multiple-week net demand sequence with a time dimension unified as weeks based on the collected data, and specifically includes:

[0019] Step 11, the data collection module automatically collects sales forecast data, current inventory data and future arrival plan data every week;

[0020] a1) For the sales forecast data, a mechanism combining a fusion algorithm model and manual correction is used to generate;

[0021] The fusion algorithm model fuses a time series model and a machine learning model, the time series model is used to capture sales trends and periodicity, and the machine learning model is used to fit nonlinear factors, and the output results are weighted according to a preset weight ratio to obtain an initial prediction value;

[0022] Providing a visual interface allows business personnel to modify the initial prediction value based on market experience, the corrected initial prediction value automatically covers the initial prediction value before correction, and the correction record is saved;

[0023] a2) For the current inventory data, it includes in-stock inventory synchronized in real time from the warehouse management system and in-transit inventory with expected arrival time synchronized from the transportation management system;

[0024] a3) For the future arrival plan data, it includes predicted arrival data of placed purchase orders collected from the purchase management system;

[0025] Step 12, set a unified time reference, define the starting and ending time points of each week through a weekly boundary calibration algorithm, and add a standardized week timestamp to all collected data;

[0026] Step 13, according to the formula: D n = P n - (I0+ΣT k ) +ΣA m Calculate the net demand D n of the nth week in the future; Where P n is the sales forecast demand of the nth week, I0 is the current available inventory, ΣT k is the sum of all in-transit inventory with an expected arrival time no later than the nth week, k is the in-transit inventory batch number, ΣA m is the sum of all allocated inventory that has been locked by sales orders but not shipped and is expected to be occupied no later than the nth week, and m is the allocation record number;

[0027] Step 14, the calculated net demand sequence is synchronized in real time to the stock preparation demand collaborative calculation module through a data bus.

[0028] Further, the step 1 triggers the re-calculation of the net demand sequence when the inventory fluctuation exceeds the set threshold, specifically including:

[0029] The data acquisition module is connected with the inventory management system through the application program interface of the inventory management system, and the fluctuation threshold of the inventory fluctuation range is set;

[0030] The inventory management system calculates the fluctuation range of each inventory item in real time, and the calculation formula of the fluctuation range is: fluctuation range = |current inventory-last inventory collection| / last inventory collection*100%;

[0031] When the fluctuation range of a certain inventory item is monitored in real time to exceed the fluctuation threshold, the data acquisition module is automatically triggered to re-calculate all related net demand items according to the formula: n = P n - (I0+ΣT k ) +ΣA m

[0032] The re-calculated net demand sequence is synchronized to the backup demand collaborative calculation module through the data bus.

[0033] Further, the step 2 uses the head transportation time efficiency matrix to dynamically correct the deviation rate of the actual time efficiency and the basic time efficiency value, specifically including:

[0034] Step 211, the head transportation time efficiency matrix configures the basic time efficiency T according to the four dimensions of destination warehouse, transportation mode, cargo specification and cargo source, and the configuration method supports manual input or automatic synchronization through the logistics merchant system interface;

[0035] Step 212, record the historical transportation time efficiency of each four-dimensional combination in the past set period, and calculate the average value as the actual time efficiency;

[0036] Step 213, compare the actual time efficiency with the basic time efficiency value every week, and calculate the deviation rate, deviation rate = |actual time efficiency-basic time efficiency T| / basic time efficiency T*100%;

[0037] Step 214, when the deviation rate continuously reaches the set number of weeks and exceeds the set proportion, the basic time efficiency T of the four-dimensional combination is automatically updated to the corrected basic time efficiency T', and the corrected basic time efficiency T' = basic time efficiency T+ (actual time efficiency-basic time efficiency T)*K, wherein K is a smoothing coefficient configured according to user demand.

[0038] Further, the step 2 uses the safety inventory to link the fluctuation rate of the net demand sequence, specifically including:

[0039] ​Step 221, the safety stock is supported to be configured in double levels of SKU dimension and the category dimension to which it belongs, wherein the priority of the SKU dimension configuration is higher than that of the category dimension, when the SKU dimension and the category dimension exist at the same time and the values are different, the SKU dimension configuration is adopted; when the SKU dimension is not configured, the highest level category dimension of the category to which it belongs is automatically matched;

[0040] Step 222, the demand fluctuation rate between consecutive weeks is calculated based on the net demand sequence, and the fluctuation rate of the nth week = |D n - D n-1 | / D n-1 ×100%, wherein D n is the net demand of the nth week, D n-1 is the net demand of the (n-1)th week; for the fluctuation rate of the first week, the average value of the historical same period fluctuation rate is replaced;

[0041] Step 223, the trigger threshold of the fluctuation rate and the continuous trigger times threshold are set; when the week fluctuation rate of a certain inventory item is detected to exceed the trigger threshold for the first time, the counter starts to accumulate;

[0042] Step 224, if the number of times that the fluctuation rate continuously exceeds the trigger threshold reaches the continuous trigger times threshold, the safety stock S of the inventory item is automatically adjusted to S' = S × α, wherein S' is the adjusted safety stock, and α is an adjustment coefficient configured according to user demand;

[0043] Step 225, after the adjustment is completed, the counter is automatically reset.

[0044] Further, the transportation priority dynamic rule in step 2 is used in linkage with the net demand sequence, specifically including:

[0045] Step 231, at least two basic templates are provided: a cost priority template and a time efficiency priority template;

[0046] The default transportation mode sequence of the cost priority template is slow ship, fast ship and air transportation in turn, and a cost threshold constraint is built in;

[0047] The default transportation mode sequence of the time efficiency priority template is air transportation, fast ship and slow ship in turn, and a cost upper limit constraint is built in;

[0048] Step 232, a user is supported to set a custom trigger condition based on one or more of the magnitude of net demand, time node and cargo attribute;

[0049] Step 233, the core parameters in the custom trigger condition include the set net demand and seasonal time node, which are directly obtained from the net demand sequence or its derivative data; when the net demand sequence update causes the custom trigger condition to be met, the priority of the corresponding transportation mode of the used basic template is automatically switched.

[0050] Further, the step 31 specifically includes:

[0051] By matching the week timestamp, the basic time limit and the safety stock are respectively bound to each week of the net demand sequence, the effective period of each transportation mode covering future demand is calculated, the starting point of the effective period is the current week plus the sum of the corresponding basic time limit and safety stock; the end point of the effective period is the last week covered by the net demand sequence;

[0052] The calculation method of the effective period is: for any transportation mode, the period of future demand covered between the starting point of the effective period and the end point of the effective period is the effective period.

[0053] Further, the step 32 specifically includes:

[0054] Step 321, according to the currently effective transportation priority dynamic rule, all available transportation modes are sorted from high to low;

[0055] Step 322, in the order of priority from high to low, the interval locking algorithm is used to calculate the demand interval of each transportation mode:

[0056] b1) for the transportation mode with the highest priority, the demand interval is from the starting point of the effective period of its own to the end point of the effective period;

[0057] b2) for any transportation mode with a priority lower than the highest, an emergency demand period is set, the effective period of each transportation mode covering future demand is calculated, and it is judged whether the effective period is partially or entirely within the emergency demand period, if not, the demand interval is from the starting point of the effective period of its own to the end point of the effective period; if yes, the starting point of the demand interval is the starting point of the effective period of its own, and the end point of the demand interval is the starting point of the effective period of the transportation mode with a priority one step higher than its own minus one week; if the calculated starting point of the demand interval is greater than the end point of the demand interval, the transportation mode has no allocable demand interval this time;

[0058] Step 323, each transportation mode only allocates the net demand amount in the corresponding week of the net demand sequence within its demand interval as its stock quantity;

[0059] Stock quantity calculation formula: the nth week stock quantity E of a transportation mode n= Dstart of effective period of n + n - 1

[0060] Step 324, when the headway transportation time limit or safety stock of any transportation mode is dynamically adjusted, resulting in a change in its effective period, steps 321 to 323 are automatically re-executed, and the allocable demand interval and the reserve quantity of all related transportation modes are recalculated and updated.

[0061] Further, the step 33 specifically includes:

[0062] Step 331, define the demand gap rate of the nth week = (D n -∑E n ) / D n × 100%, wherein D n is the net demand of the nth week, and ∑E n is the sum of the reserve quantities of all transportation modes in the nth week;

[0063] Step 332, perform a full-cycle demand gap rate check automatically every interval time, and trigger a full-cycle demand gap rate check immediately when the net demand sequence or key parameters are updated;

[0064] Step 333, if the check finds that the demand gap rate of any future week exceeds the set threshold, perform a reverse adjustment strategy in a preset order, and the reverse adjustment strategy includes:

[0065] c1) preferentially temporarily increase the effective coverage range of the scenario-based high-priority transportation mode: select the transportation mode with sufficient current capacity and an interval of ≤5 weeks between the start of the effective period and the gap period, and temporarily advance the start of the effective period of the transportation mode by 1-2 weeks;

[0066] c2) if the demand gap rate does not meet the standard after step c1 is performed, temporarily shorten the basic time limit of the scenario-based low-priority transportation mode: select the transportation mode with the lowest absolute priority, temporarily reduce the value of the basic time limit of the transportation mode by a predetermined percentage, and advance the start of the effective period of the transportation mode by 1-2 weeks;

[0067] c3) if the demand gap rate exceeds the limit threshold, temporarily increase the priority of air transportation;

[0068] Step 334, the parameter changes generated after the reverse adjustment strategy is executed are only used for the current round of reserve demand calculation.

[0069] Further, the step 4 specifically includes:

[0070] Step 41, the reserve demand coordination calculation module synchronizes the calculation results including demand gap rate analysis and adjustment suggestions to the business visualization system through the data bus for display after completing the calculation;

[0071] Step 42, the response cycle from data update to calculation result output is controlled within a set time.

[0072] With the technical scheme, the application has the beneficial effects that:

[0073] 1. The timing dislocation problem caused by module data fragmentation is solved, the demand gap rate is reduced through timestamp alignment mechanism and dynamic verification, the time dimension of demand, time limit and inventory is unified, the accuracy of goods preparation is improved, the timing dislocation problem caused by manual transcription is completely solved, and the effect is remarkable in the cross-border e-commerce multi-warehouse goods preparation scene.

[0074] 2. The problem of static solidification of key parameters is solved, the dynamic adjustment rules of head transportation time limit matrix, safety stock and transportation priority are constructed, the automatic switching of multi-scenarios such as off-season and peak season, standard parts / large parts is supported, without secondary development, different industry demands such as cross-border e-commerce, 3C manufacturing and fresh cold chain can be adapted, and the scene adaptability is significantly enhanced.

[0075] 3. The problem of disconnection between transportation priority and demand allocation is solved, the linkage mechanism of priority and demand cycle is established, the transportation resource allocation is optimized, the proportion of high-cost transportation mode is reduced, and it is only used to cover short-term urgent demand, and the annual transportation cost of enterprises is reduced.

[0076] 4. The problem of lack of full-process cooperation is solved, the automatic triggering and data synchronization mechanism between design modules is designed, the response cycle is shortened, the response speed is improved, and the risk of out-of-stock in peak season is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0077] In order to more clearly illustrate the technical scheme in the embodiments of the application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0078] Figure 1 is an execution flowchart of a multi-module dynamic collaborative goods preparation demand calculation method provided by the embodiment of the application.

[0079] Figure 2 is a logical architecture diagram between the data acquisition module, the parameter dynamic configuration module and the goods preparation demand collaborative calculation module provided by the embodiment of the application. DETAILED DESCRIPTION

[0080] The application will be described in further detail below with reference to the drawings and embodiments. It is particularly pointed out that the following embodiments are only for illustration of the application, but do not limit the scope of the application. Similarly, the following embodiments are only part of the embodiments of the application, not all embodiments, and all other embodiments obtained by those of ordinary skill in the art without creative labor are within the scope of protection of the application.

[0081] The application adopts a "coordination center + distributed module + data bus" architecture, and the core includes three modules: a data acquisition module, a parameter dynamic configuration module, and a spare parts demand coordination calculation module. Cross-module data transmission and logic linkage are realized through a real-time data bus, and the architecture logic is as shown in Figure 2 .

[0082] Please refer to Figure 1 , a multi-module dynamic coordination type spare parts demand calculation method of the application realizes real-time data transmission and logic linkage between the data acquisition module, the parameter dynamic configuration module, and the spare parts demand coordination calculation module through a data bus. The method includes the following steps:

[0083] Step 1: The data acquisition module automatically collects sales forecast data, current inventory data, and future arrival plan data every week, generates a net demand sequence of multiple future weeks with a time dimension unified as a week based on the collected data, and triggers a net demand sequence recalculation when inventory fluctuation exceeds a set threshold. The data acquisition module as a dynamic data source breaks through the traditional static collection mode, generates a time-consistent weekly net demand sequence, and captures inventory fluctuations in real time, providing an accurate data source for coordination calculation.

[0084] In this embodiment, the data acquisition module in step 1 automatically collects sales forecast data, current inventory data, and future arrival plan data every week, generates a net demand sequence of multiple future weeks with a time dimension unified as a week based on the collected data, and specifically includes:

[0085] Step 11: The data acquisition module automatically collects sales forecast data, current inventory data, and future arrival plan data every week.

[0086] a1) For the sales forecast data, a mechanism combining a fusion algorithm model and manual correction is used to generate;

[0087] The fusion algorithm model fuses a time series model (ARIMA) and a machine learning model (LightGBM). The time series model is used to capture sales trends and periodicity, and the machine learning model is used to fit non-linear factors (such as promotion, season, etc.). The output results are weighted according to a preset weight ratio (such as a weight ratio of 7:3 for the output results of the time series model and the machine learning model) to obtain an initial prediction value.

[0088] It provides a visual interface that allows business personnel to modify the initial forecast based on market experience (such as sudden promotions and supply chain risks). The modified initial forecast automatically overwrites the original initial forecast and saves the modification record.

[0089] a2) The current inventory data includes in-stock inventory synchronized in real time from the warehouse management system and in-transit inventory synchronized from the transportation management system with an estimated arrival time.

[0090] a3) The future delivery plan data includes the expected delivery data of purchase orders that have been placed, collected from the procurement management system;

[0091] Step 12: Set a unified time base (the time unit is strictly unified to "week"). Define the start and end times of each week through the week boundary calibration algorithm (e.g., the start time of each week is Monday 00:00 and the end time is Sunday 24:00 to eliminate cross-system time zone deviation). Add a standardized week timestamp to all collected data (format: "year - W week", such as "week 1 = 2025-W40", to ensure that the time dimension of all data is unified to "week").

[0092] Step 13: According to the formula: D n = P n - (I0+ΣT k ) +ΣA m Calculate the net demand D for the next nth week. n (Unit: piece); where P n Let I0 be the sales forecast demand (units) for week n, and I0 be the current available inventory (units). ΣT k ΣA represents the total in-transit inventory with an estimated arrival time no later than week n, where k is the batch number of the in-transit inventory. m This represents the total allocated inventory that has been locked by sales orders but has not yet been shipped, and is expected to be held for no later than week n, where m is the allocation record number;

[0093] Step 14: Synchronize the calculated net demand sequence to the inventory demand collaborative calculation module in real time via the data bus.

[0094] In this embodiment, step 1, which triggers a recalculation of the net demand sequence when inventory fluctuations exceed a set threshold, specifically includes:

[0095] Connect the data acquisition module to the inventory management system through the application programming interface (API) of the inventory management system, and set the fluctuation threshold of inventory fluctuation (default 5%).

[0096] The inventory management system calculates the fluctuation amplitude of each inventory item in real time, and the calculation formula of the fluctuation amplitude is: fluctuation amplitude = |current inventory - last collected inventory| / last collected inventory x 100%;

[0097] When the fluctuation amplitude of a certain inventory item is monitored in real time and exceeds the fluctuation threshold, the data acquisition module is automatically triggered to recalculate the net demand sequence according to the formula: D n = P n - (I0+ΣT k ) +ΣA m Recalculate all related net demand items;

[0098] The recalculated net demand sequence is synchronized to the replenishment demand collaborative calculation module through the data bus. When the data bus is transmitted, the week time stamp format of the replenishment demand collaborative calculation module data is automatically checked. If there is non-integral week data (such as 3-day inventory change), it is converted into weekly equivalent data (3 days = 3 / 7 ≈ 0.43 weeks) through the "non-integral week data intelligent conversion algorithm", ensuring consistency with other modules in time dimension.

[0099] Step 2, the parameter dynamic configuration module configures and dynamically adjusts the key parameters for replenishment demand calculation, including head transportation time efficiency matrix, safety stock, and transportation priority dynamic rules. All key parameter changes are synchronized to the replenishment demand collaborative calculation module in real time through the data bus. When the parameters are changed, the replenishment demand collaborative calculation module is automatically triggered to recalculate (response delay < 30 seconds), ensuring that the parameters are dynamically adapted to the demand; among them, the synchronization through the data bus adopts the event-driven mechanism: "event-driven + intelligent judgment of influence range" mode. When the data acquisition module or the parameter dynamic configuration module changes data, an event carrying the change content and the influence range is automatically generated (such as carrying parameter type, modification range, etc.); the recalculation range: only the part of the replenishment demand collaborative calculation module corresponding to the influence range is recalculated (such as the modification of the time efficiency of a certain transportation mode in a certain warehouse, only the replenishment demand related to the transportation mode in the warehouse is recalculated), instead of global recalculation, to improve response efficiency;

[0100] The parameter dynamic configuration module serves as a linkage parameter pool, builds a multi-dimensional dynamic parameter system, realizes intelligent adjustment of head transportation time efficiency, safety stock, and transportation priority, and is linked in real time with the net demand sequence.

[0101] In the embodiment, the headhaul transportation time matrix (headhaul generally refers to the first segment of logistics transportation from the supplier or production site to the destination warehouse. The transportation time refers to the time required from shipment to receipt. The matrix represents a multi-dimensional time configuration table for dynamically calculating transportation time according to different transportation condition combinations. The headhaul transportation time matrix supports dynamic adjustment of transportation time according to transportation conditions, improving the accuracy and adaptability of the inventory plan) is used to dynamically correct the deviation rate of the actual time from the base time value, avoiding transportation plan deviation caused by using static values, specifically including:

[0102] Step 211, the headhaul transportation time matrix configures the base time T according to the destination warehouse, transportation mode, cargo specification, and cargo source, and the configuration method supports manual input (users fill in the base time value according to four-dimensional combinations on the system interface) or automatic synchronization through the logistics company system interface (interface with the logistics company system API, automatically obtain real-time transportation time data and fill in); the two configuration methods can be switched between each other.

[0103] Step 212, record the historical transportation time under each four-dimensional combination in the past setting period (preferably the past 4 weeks), and calculate the average value as the actual time to ensure data stability;

[0104] Step 213, compare the actual time with the base time T every week, calculate the deviation rate, and the deviation rate = |actual time-base time T| / base time T×100%;

[0105] Step 214, when the deviation rate continuously reaches the set number of weeks and exceeds the set proportion (for example, the deviation rate exceeds 10% for 2 consecutive weeks), automatically update the base time T of the four-dimensional combination to the corrected base time T', and the corrected base time T' = base time T+ (actual time-base time T)×K, where K is a smoothing coefficient configured according to user requirements, which can be configured as K=0.8, to avoid excessive adjustment of time (for example, the actual time of "large + slow ship + A warehouse + B location cargo source" is 2 weeks longer than the configured value, and the system automatically increases the base time T value of this combination by 1.6 weeks).

[0106] In the embodiment, the safety stock in step 2 is used to adjust the volatility of the net demand sequence, specifically including:

[0107] Step 221, the safety stock supports the double-level configuration of the SKU (SKU is the smallest unit of goods in warehouse and logistics management for unique identification and management. Each SKU represents a specific combination of product attributes) dimension and the category dimension to which it belongs (deeply bound with demand fluctuation linkage rules), wherein the dimension priority rule: the priority of the SKU dimension configuration is higher than that of the category dimension. When the SKU dimension and the category dimension exist at the same time and the values are different, the SKU dimension configuration is adopted; when there is no SKU dimension configured, the highest level category dimension of the category to which the SKU belongs is automatically matched (for example, if the SKU belongs to a "turn signal", the "turn signal" category configuration is matched first, otherwise the "automobile accessories" category is matched, and the trace is continued upwards);

[0108] The demand fluctuation linkage rule is as follows:

[0109] Step 222, the demand fluctuation rate between consecutive weeks is calculated based on the net demand sequence, and the fluctuation rate of the nth week = |D n - D n-1 | / D n-1 ×100%, wherein D n is the net demand of the nth week, and D n-1 is the net demand of the (n-1)th week; for the fluctuation rate of the first week, the historical average fluctuation rate of the same period is used instead (for example, the average fluctuation rate of the first week of the same season in the past three years is used);

[0110] Step 223, set the trigger threshold of the fluctuation rate (which can be configured as 20%) and the continuous trigger times threshold (which can be configured as 3 times); maintain an independent counter, and when the weekly fluctuation rate of a certain inventory item is detected to exceed the trigger threshold for the first time, the counter starts to accumulate;

[0111] Step 224, if the number of times that the fluctuation rate continuously exceeds the trigger threshold reaches the continuous trigger times threshold indicated by the counter, the safety stock S (unit: pieces) of the inventory item is automatically adjusted to S' = S×α, wherein S' is the adjusted safety stock (unit: pieces), and α is an adjustment coefficient configured according to user demand, which can be configured as α=1.2;

[0112] Step 225, after the adjustment is completed, the counter is automatically reset. For example, the counter starts to accumulate when the fluctuation rate is detected to exceed 20% (which can be configured) for the first time, and adjustment (S'=S×1.2, 1.2 is a configurable adjustment coefficient) is performed when the fluctuation is triggered continuously for 3 times. After the adjustment, the counter is automatically reset; if the fluctuation continues in the 4th week, it needs to be accumulated for 3 times again to trigger the next adjustment, so as to avoid frequent fluctuations.

[0113] In the embodiment, the transportation priority dynamic rule in step 2 is used to link with the net demand sequence, specifically including:

[0114] Step 231, provide at least two basic templates: cost priority template and time priority template;

[0115] The default transportation mode of the cost priority template is in the order of slow boat, fast boat and air transportation, and has a built-in cost threshold constraint (such as the longest time limit for slow boat is 45 days, and if it exceeds, it is automatically upgraded to fast boat), balancing cost and time;

[0116] The default transportation mode of the time priority template is in the order of air transportation, fast boat and slow boat, and has a built-in cost upper limit constraint (such as the cost of air transportation accounts for no more than 20% of the total transportation cost), to avoid cost out of control;

[0117] Step 232, support users to set custom trigger conditions based on one or more of net demand magnitude, time node and goods attributes; for example, "D1>100 pieces, enable time priority", "6-8 months of peak season automatically promote fast boat priority", "fragile goods default enable time priority";

[0118] Step 233, the core parameters in the custom trigger condition, including the set net demand and seasonal time node, are directly obtained from the net demand sequence or its derivative data; when the net demand sequence is updated to satisfy the custom trigger condition, the priority of the transportation mode used by the basic template is automatically switched.

[0119] Step 3, the inventory demand coordination calculation module (core coordination logic, function: as the core of coordination, based on the input of real-time data bus, calculate the inventory demand through "time alignment-priority allocation-dynamic verification" three-layer logic, realize the deep coupling of modules) receives the net demand sequence and key parameters, and executes the following three-layer coordination logic to calculate the inventory demand:

[0120] Step 31, align the effective period of each transportation mode with the net demand sequence;

[0121] In this embodiment, the step 31 specifically includes:

[0122] By matching the week timestamp, the basic time and safety stock are respectively bound to each week of the net demand sequence, and the effective period of each transportation mode covering future demand is calculated. The starting point of the effective period is the sum of the current week, the corresponding basic time and safety stock, that is, the starting point of the effective period = current week + basic time + safety stock; The end point of the effective period is the last week covered by the net demand sequence, for example: net demand sequence (D1~D 48 ), then the end point of the effective period = 48th week;

[0123] The calculation method of the effective period is: for any transportation mode, the period covered by the future demand between the start point of the effective period and the end point of the effective period as the effective period (for example, the current of the slow ship is the first week, the basic time limit is 12 weeks, and the safety stock is 2 weeks, so the start point of the effective period of the slow ship is 1+12+2=15 weeks, and the effective period is the demand of the 15th week and later).

[0124] Step 32, determine the priority order of the transportation mode based on the transportation priority dynamic rule, allocate the demand interval to each transportation mode in the priority order by using the interval locking algorithm, and calculate the inventory based on the net demand of the allocated demand interval; the demand interval is dynamically adjusted with the effective period of the highest priority transportation mode;

[0125] In this embodiment, the step 32 specifically comprises:

[0126] Step 321, according to the currently effective transportation priority dynamic rule, sort all available transportation modes from high to low (for example, the time limit priority template: air transport> fast ship> slow ship);

[0127] Step 322, in the order from high to low, calculate the demand interval of each transportation mode by using the interval locking algorithm:

[0128] b1) for the highest priority transportation mode (such as air transport), its demand interval is the start point to the end point of its own effective period (for example, the current week of air transport is the first week, the basic time limit T is 2 weeks, and the safety stock S is 1 week, so the start point of the effective period is the first 1+2+1 week=the fourth week, the end point of the effective period is the 48th week, the effective period is the fourth week~the 48th week, and the demand interval is locked);

[0129] b2) for any transport mode other than the highest priority, set the emergency demand period (e.g. weeks 4-14), calculate the effective period of each transport mode covering future demand, determine whether the effective period is partially or entirely within the emergency demand period, if not, its demand interval is the start of its own effective period to the end of the effective period; if yes, its demand interval starts at the start of its own effective period, and ends at the start of the effective period of the transport mode with the next higher priority minus 1 week, i.e. the demand interval of the non-highest priority transport mode = start of its own effective period ~ start of the effective period of the transport mode with the next higher priority - 1; if the start of the calculated demand interval is greater than the end of the demand interval, the transport mode has no allocable demand interval this time (e.g. the effective period of the fast ship is weeks 9-48 (of which weeks 9-14 are within the emergency demand period), and the effective period of the transport mode with the next higher priority, air, is weeks 4-48, then the fast ship has no allocable demand interval, i.e. weeks 9-3, and no demand is allocated; e.g. the effective period of the slow ship is weeks 15-48, which is not within the emergency demand period, and the slow ship has an allocable demand interval of weeks 15-48, which is locked);

[0130] Step 323, each transport mode only allocates the net demand of the corresponding week in the net demand sequence as its stock quantity within its demand interval;

[0131] Stock quantity calculation formula: the stock quantity of the nth week of a transport mode E n = D_{effective period start + n - 1} (only when the demand of this week is not covered by the transport mode with the higher priority).

[0132] Step 324, when the base lead time or safety stock of any transport mode is dynamically adjusted, resulting in a change in its effective period, automatically re-execute steps 321-323 to recalculate and update the demand interval and stock quantity of all related transport modes. When the effective period of a transport mode with higher priority is changed, the re-calculation of the allocable interval of the transport mode with lower priority is automatically triggered (e.g. the base lead time of air is modified to 3 weeks, and the effective period is changed to weeks 5-48, the demand interval of the fast ship is updated to weeks 9-4, and there is still no effective interval);

[0133] Specific example (cost priority template: slow ship > fast ship > air, emergency demand period is weeks 4-14):

[0134] The effective period of the slow ship is weeks 15-48, and the demand of weeks 15-48 is allocated (the transport mode with the highest priority, and its demand interval is its own effective period);

[0135] Fast ship effective period: 9th week-48th week, the allocable demand interval is 9th week-14th week, allocate the demand in this interval (non-highest priority transportation mode, its effective period 9th week-14th week is in the emergency demand period, its demand interval = the start point of its own effective period-the start point of the previous priority transportation mode effective period-1, i.e. 9th week-14th week);

[0136] Air transport effective period: 4th week-48th week, the allocable demand interval is 4th week-8th week, allocate the demand in this interval (non-highest priority transportation mode, its effective period 4th week-14th week is in the emergency demand period, its demand interval = the start point of its own effective period-the start point of the previous priority transportation mode effective period-1, i.e. 4th week-8th week);

[0137] Step 33, based on the calculated total stock quantity and the net demand corresponding to the week, calculate the demand gap rate of each week, and trigger temporary reverse adjustment of the key parameters when the demand gap rate exceeds the set threshold;

[0138] In this embodiment, the step 33 specifically comprises:

[0139] Step 331, define the demand gap rate of the nth week = (D n -ΣE n ) / D n ×100%, wherein D n is the net demand of the nth week, and ΣE n is the total stock quantity of all transportation modes in the nth week;

[0140] Step 332, automatically execute the whole-period demand gap rate verification once every interval set time (such as 15 minutes), and trigger the whole-period demand gap rate verification once immediately when the net demand sequence or the key parameters are updated;

[0141] Step 333, if the verification finds that the demand gap rate of any future week exceeds the set threshold, execute the reverse adjustment strategy in the preset order, and the reverse adjustment strategy comprises:

[0142] c1) preferentially temporarily increase the effective coverage range of the scenario-based high-priority transportation mode: screen the transportation modes (such as the slow ship, whose effective period start point is the 15th week, and the interval between the 10th week of the gap period and the start point is 5 weeks, which has adjustment feasibility), which have sufficient capacity and whose effective period start point and gap period interval ≤5 weeks, and temporarily advance the start point of its effective period by 1-2 weeks (such as the slow ship effective period start point from the 15th week to the 13th-14th week);

[0143] I. Judgment standard of sufficient capacity

[0144] Core definition: the current allocated inventory of the transportation mode does not exceed 80% of its maximum carrying capacity (20% reserved for flexibility to avoid overloading);

[0145] Key parameter calculation:

[0146] Maximum carrying capacity (referred to as "maximum capacity"): take the highest actual inventory of the transportation mode in the past 3 months × 1.2 (1.2 is the flexibility coefficient to adapt to short-term demand fluctuations);

[0147] Allocated inventory: the sum of the inventory allocated by the transportation mode in all valid demand intervals (e.g. the slow ship has allocated inventory for weeks 15-48, the sum is ΣD 15 ~D 48 );

[0148] Determination formula: allocated inventory ÷ maximum capacity ≤ 80% → determine that the capacity is sufficient;

[0149] Example verification: in the slow ship example, assume the highest weekly inventory of the slow ship in the past 3 months is 1000, the maximum capacity = 1000 × 1.2 = 1200; the slow ship has allocated a total of 800 inventory for weeks 15-48, 800 ÷ 1200 ≈ 66.7% ≤ 80% → determine that the capacity is sufficient.

[0150] II. Definition and calculation logic of gap period

[0151] Core definition: the gap period refers to the "first week when the demand gap rate exceeds the set threshold (default 5%) continuously", which is the core time node for triggering reverse adjustment;

[0152] Calculation premise: first calculate the gap rate of each week according to the formula: the demand gap rate of the nth week = (D n -ΣE n ) / D n × 100%, where D n is the net demand of the nth week, ΣE n is the total inventory of all transportation modes in the nth week;

[0153] Calculation steps:

[0154] Step 1: iterate through all future weeks (e.g. weeks 4-48) and calculate the gap rate of each week;

[0155] Step 2: filter the weeks with a gap rate > 5% (set threshold), if there are 2 or more consecutive weeks with a gap rate exceeding the threshold, take the first week as the "gap period";

[0156] Step 3: if only a single week has a gap rate exceeding the threshold (without continuity), that week is directly determined as the gap period;

[0157] Example verification:

[0158] Net demand D of the 10th week 10 = 1000 pieces, total sum of stock levels ΣE 10 = 920 pieces.

[0159] The logic of the "10th week of the gap cycle" is: the 10th week gap rate = (D 10 - ΣE 10 ) / D 10 ×100% = 8%>5%, and the consecutive gap rate of the 10th-11th week exceeds the standard → the first week of exceeding the standard (the 10th week) is determined as the gap cycle.

[0160] c2) If the demand gap rate does not meet the standard after performing step c1 (such as the gap rate >5%), temporarily shorten the base lead time of the scenario-based low-priority transportation mode: select the transportation mode with the lowest absolute priority (such as slow ship), temporarily adjust the value of its base lead time by a predetermined percentage (such as adjusting the value of the base lead time of the slow ship from 14 weeks to 10% to 12.6 weeks), and advance the starting point of its effective cycle by 1-2 weeks (such as from the 13th-14th week to the 12th-13th week), expand the effective coverage range to fill the gap;

[0161] c3) If the demand gap rate exceeds the limit threshold, temporarily increase the priority of air transportation (only when the demand gap rate >10% is enabled);

[0162] Priority definition (two categories, adaptive adjustment scenario):

[0163] ① Absolute priority (global fixed): air transportation (high) > fast ship (medium) > slow ship (low) (the faster the lead time and the higher the cost, the higher the priority);

[0164] ② Scenario-based priority (only used for gap correction):

[0165] Scenario-based high priority: transportation mode that can cover the gap and is not at full capacity (feasibility first);

[0166] Scenario-based low priority: transportation mode with the lowest absolute priority and the smallest time compression cost (cost first).

[0167] Step 334, the parameter changes generated after the execution of the reverse adjustment strategy are only used for the calculation of the stock demand of the current round, and do not modify the basic parameters permanently to avoid affecting subsequent normal calculations, and the reverse adjustment records are separately retained for tracing.

[0168] Step 4, the stock demand coordination calculation module synchronizes the calculation results to the business visualization system through the data bus.

[0169] In the present embodiment, the step 4 specifically comprises:

[0170] Step 41, after the calculation is completed, the calculation result containing the demand gap rate analysis and adjustment suggestion (such as "the 10th week gap rate is 8%, and it is suggested that the effective period starting point of the fast ship be advanced by 1 week") is synchronized to the business visualization system through the data bus for display;

[0171] Step 42, the response period from data update to calculation result output is controlled within a set time (such as within 15 minutes).

[0172] From the research and development practice, the module coordination mechanism of the present application has multiple technical barriers that are easily overlooked or difficult to achieve by non-technical personnel in the field:

[0173] The technical complexity of time sequence alignment In the prior art, the time units of demand, time efficiency and inventory are naturally heterogeneous (demand is in "month", time efficiency is in "day", and inventory is in "real time"). In research and development, problems such as "non-integer week data intelligent conversion" (such as 3-day inventory change converted to 0.43 weeks) and "cross-system timestamp synchronization" (solving different database time zone deviations) need to be broken through. For example, in order to ensure that "the nth week" is completely consistent in time range in each module, a "weekly boundary calibration algorithm" needs to be developed, which requires 6-8 weeks of iterative testing and is not a simple unit conversion.

[0174] Coupling logic of dynamic parameter correction The deviation correction of the head transportation time efficiency matrix needs to establish a ternary model of "historical data - configuration value - demand fluctuation" (rather than single-dimensional correction), and the fluctuation linkage of safety stock needs to balance "response sensitivity" and "adjustment stability" (avoiding frequent shocks). For example, when the demand fluctuation rate exceeds the threshold, it needs to be checked whether the fluctuation is caused by a sudden order (not a trend change) before deciding whether to adjust the safety stock. This logic needs to integrate statistics and business rules and cannot be directly derived through existing single parameter configuration functions.

[0175] Interval lock algorithm of priority allocation The rule that low-priority transportation only covers "pre-period gaps" needs to compare the effective periods of multiple transportation modes with the demand sequence in real time. In research and development, problems such as "interval overlap detection" and "gap quick positioning" need to be solved. For example, when the effective period of the slow ship is extended by 1 week due to parameter correction, the coverage interval of the fast ship needs to be automatically contracted to ensure no repeated allocation. This process involves the design of dynamic interval tree data structure, which has much higher complexity than static sorting.

[0176] In summary, the technical features of the present application require overcoming multiple technical obstacles such as time sequence alignment, dynamic correction, and interval allocation. The coordination logic is a creative solution formed after a large number of tests and iterations in the research and development process, and is not a simple addition of existing technologies.

[0177] The above merely illustrates some embodiments of the present application, and is not intended to limit the protection scope of the present application. Any equivalent device or equivalent process transformation, or direct or indirect application in other related technical fields, which is made according to the content of the present application, shall be included in the patent protection scope of the present application.

Claims

1. A multi-module dynamic collaborative inventory demand calculation method, characterized in that, Real-time data transmission and logic linkage are realized between the data acquisition module, the parameter dynamic configuration module and the stock preparation demand collaborative calculation module through a data bus, and the method comprises the following steps: Step 1: The data acquisition module automatically collects sales forecast data, current inventory data and future arrival plan data every week, generates a net demand sequence of multiple future weeks with the time dimension unified as weeks based on the collected data, and triggers the recalculation of the net demand sequence when the inventory fluctuation exceeds a set threshold; for the sales forecast data, a mechanism combining a fusion algorithm model and manual correction is used to generate the data; the fusion algorithm model fuses a time series model and a machine learning model, the time series model is used to capture sales trends and periodicity, and the machine learning model is used to fit nonlinear factors, and the initial prediction value is obtained by weighting the output results of the two models according to a preset weight ratio; Step 2: The parameter dynamic configuration module configures and dynamically adjusts key parameters for stock preparation demand calculation, including head transportation time efficiency matrix, safety stock and transportation priority dynamic rules, the head transportation time efficiency matrix is used to dynamically correct the deviation rate of actual time efficiency and basic time efficiency value, the safety stock is used to link and adjust the fluctuation rate of the net demand sequence, and the transportation priority dynamic rules are used to link with the net demand sequence; all key parameter changes are synchronized in real time to the stock preparation demand collaborative calculation module through the data bus; Step 3: The stock preparation demand collaborative calculation module receives the net demand sequence and key parameters, and calculates the stock preparation demand by executing the following three-layer collaborative logic: Step 31: Align the effective period of each transportation mode with the net demand sequence; Step 32: Determine the priority order of the transportation modes based on the transportation priority dynamic rules, allocate demand intervals to each transportation mode in priority order by using interval locking algorithm, and calculate the stock preparation amount based on the net demand amount of the allocated demand interval; b1) For the transportation mode with the highest priority, the demand interval is from the start point of the effective period of the transportation mode itself to the end point of the effective period; b2) For any transportation mode other than the highest priority, an emergency demand period is set, the effective period of each transportation mode covering future demand is calculated, it is judged whether the effective period is partially or entirely within the emergency demand period, if not, the demand interval is from the start point of the effective period of the transportation mode itself to the end point of the effective period; if yes, the start point of the demand interval is the start point of the effective period of the transportation mode itself, and the end point of the demand interval is the start point of the effective period of the transportation mode one week earlier than the transportation mode with higher priority; if the calculated start point of the demand interval is greater than the end point of the demand interval, the transportation mode has no allocatable demand interval this time; Step 33: Calculate the demand gap rate of each week based on the sum of the calculated stock preparation amounts and the net demand of the corresponding week, and trigger temporary reverse adjustment of the key parameters when the demand gap rate exceeds a set threshold; Step 4: The stock preparation demand collaborative calculation module synchronizes the calculation results to the business visualization system through the data bus.

2. The multi-module dynamic collaborative inventory demand calculation method of claim 1, wherein, The data collection module in step 1 automatically collects sales forecast data, current inventory data and future arrival plan data every week, generates a future multiple-week net demand sequence with a time dimension unified as weeks based on the collected data, and specifically includes: Step 11, the data collection module automatically collects sales forecast data, current inventory data and future arrival plan data every week; a1) for the sales forecast data, a visual interface is provided to allow business personnel to modify the initial forecast value based on market experience, the modified initial forecast value automatically covers the initial forecast value before correction, and the correction record is saved; a2) for the current inventory data, it includes in-stock inventory synchronized in real time from the warehouse management system and in-transit inventory synchronized from the transportation management system and having a predicted arrival time; a3) for the future arrival plan data, it includes predicted arrival data of the placed purchase order collected from the purchase management system; Step 12, set a unified time reference, define the starting and ending time points of each week through a weekly boundary calibration algorithm, and add a standardized week timestamp to all collected data; Step 13, calculate the net demand D n n of the future week according to the formula: n = P k - (I0+∑T m ) +∑A n ; wherein P n is the sales forecast demand of the nth week, I0 is the current available inventory, ∑T k is the sum of all in-transit inventory with an estimated arrival time no later than the nth week, k is the in-transit inventory batch number, ∑A m is the sum of all allocated inventory that has been locked by a sales order but not shipped and is estimated to be occupied no later than the nth week, m is the allocation record number. Step 14, the calculated net demand sequence is synchronized in real time to the stock preparation demand collaborative calculation module through the data bus.

3. The multi-module dynamic collaborative inventory requirement calculation method of claim 2, wherein, The net demand sequence recalculation is triggered when the inventory fluctuation exceeds the set threshold in step 1, specifically including: Through the application programming interface of the inventory management system, the data collection module is connected with the inventory management system, and the fluctuation threshold of the inventory fluctuation range is set; The inventory management system calculates the fluctuation range of each inventory item in real time, and the calculation formula of the fluctuation range is: fluctuation range = |current inventory-last collected inventory| / last collected inventory×100%; When the fluctuation amplitude of a certain inventory item is monitored in real time to exceed the fluctuation threshold, the data collection module is automatically triggered to recalculate all related net demand items according to the formula: D n = P n - (I0+ΣT k ) +ΣA m recompute all related net demand items; The recalculated net demand sequence is synchronized to the stock preparation demand collaborative calculation module through the data bus.

4. The multi-module dynamic collaborative inventory requirement calculation method of claim 1, wherein, The head transportation time efficiency matrix in step 2 is used to dynamically correct the deviation rate of the actual time efficiency and the basic time efficiency value, specifically including: Step 211, the head transportation time efficiency matrix configures the basic time efficiency T according to the destination warehouse, transportation mode, cargo specification and cargo source, and the configuration method supports manual input or automatic synchronization through the logistics merchant system interface; Step 212, record the historical transportation time efficiency of each four-dimensional combination in the past set period, and calculate the average value as the actual time efficiency; Step 213, compare the actual time efficiency with the basic time efficiency value every week, calculate the deviation rate, deviation rate = |actual time efficiency-basic time efficiency T| / basic time efficiency T×100%; Step 214, when the deviation rate continuously reaches the set number of weeks and exceeds the set proportion, the basic time efficiency T of the four-dimensional combination is automatically updated to the corrected basic time efficiency T', corrected basic time efficiency T'=basic time efficiency T+ (actual time efficiency-basic time efficiency T)×K, wherein K is a smoothing coefficient configured according to user requirements.

5. The multi-module dynamic collaborative inventory requirement calculation method of claim 1, wherein, The safety stock in step 2 is used to link the fluctuation rate of the net demand sequence, specifically including: Step 221, the safety stock is supported to be configured in double levels of SKU dimension and the category dimension to which it belongs, wherein the priority of the SKU dimension configuration is higher than that of the category dimension, when the SKU dimension and the category dimension exist at the same time and the values are different, the SKU dimension configuration is adopted; when the SKU dimension is not configured, the highest level category dimension of the category to which it belongs is automatically matched; Step 222, calculate the demand fluctuation rate between consecutive weeks based on the net demand sequence, the fluctuation rate of the nth week = |D n -D n-1 | / D n-1 x 100%, wherein D n is the net demand of the nth week, D n-1 is the net demand of the n-1th week; for the fluctuation rate of the 1st week, the historical same period fluctuation rate average is replaced; Step 223, the trigger threshold of the fluctuation rate and the continuous trigger number threshold are set; when it is detected that the fluctuation rate of a certain inventory item exceeds the trigger threshold for the first time, the counter starts to accumulate; Step 224, if the number of times that the fluctuation rate continuously exceeds the trigger threshold reaches the continuous trigger number threshold, the safety stock S of the inventory item is automatically adjusted to S' = S x a, wherein S' is the adjusted safety stock, and a is an adjustment coefficient configured according to user demand; Step 225, after the adjustment is completed, the counter is automatically reset.

6. The multi-module dynamic collaborative inventory requirement calculation method of claim 1, wherein, The transportation priority dynamic rule in step 2 is used in linkage with the net demand sequence, and specifically includes: Step 231, at least two basic templates are provided: a cost priority template and a time efficiency priority template; The default transportation mode sequence of the cost priority template is slow ship, fast ship and air transportation in turn, and a cost threshold constraint is built in; The default transportation mode sequence of the time efficiency priority template is air transportation, fast ship and slow ship in turn, and a cost upper limit constraint is built in; Step 232, a user is supported to set a custom trigger condition based on one or more of the net demand magnitude, time node and cargo attribute; The core parameters in the custom trigger condition include the set net demand and seasonal time node, which are directly obtained from the net demand sequence or derived data thereof; when the net demand sequence is updated to cause the custom trigger condition to be met, the priority of the transportation mode used by the basic template is automatically switched.

7. The multi-module dynamic collaborative inventory requirement calculation method of claim 1, wherein, The step 31 specifically includes: The basic time efficiency and the safety stock are bound to each week of the net demand sequence respectively through week timestamp matching, the effective period of each transportation mode for covering future demand is calculated, the start point of the effective period is the sum of the current week and the sum of the corresponding basic time efficiency and safety stock; the end point of the effective period is the last week covered by the net demand sequence; The calculation method of the effective period is that, for any transportation mode, the period of the future demand covered between the start point of the effective period and the end point of the effective period is the effective period.

8. The multi-module dynamic collaborative inventory requirement calculation method of claim 1, wherein, The step 32 specifically includes: Step 321, according to the currently effective transportation priority dynamic rule, all available transportation modes are sorted from high to low; Step 322, in the order from high to low, the interval locking algorithm is used to calculate the demand interval for each transportation mode; Step 323, each transportation mode only allocates the net demand quantity of the corresponding week in the net demand sequence in the demand interval as its stock quantity. The formula for calculating the inventory quantity is: the inventory quantity E of the nth week of a certain transportation mode n = D_{start of the effective period + n - 1}; Step 324, when the headway transportation time limit or safety stock of any transportation mode is dynamically adjusted, resulting in a change in its effective period, automatically re-perform steps 321 to 323, recalculate and update the allocable demand interval and stock quantity of all related transportation modes.

9. The multi-module dynamic collaborative inventory needs calculation method of claim 1, wherein, The step 33 specifically includes: Step 331, define the demand gap rate of the nth week = (D n -∑E n ) / D n x 100%, wherein D n is the net demand of the nth week, and ∑E n is the total sum of the stock-up amount of all transportation modes in the nth week. Step 332, define the demand gap rate of the nth week = (D n -∑E n ) / D n x 100%, wherein D n is the net demand of the nth week, and ∑E n is the total sum of the stock-up amount of Step 332, automatically perform a full-cycle demand gap rate check every interval set time; when the net demand sequence or key parameters are updated, immediately trigger a full-cycle demand gap rate check; Step 333, if the check finds that the demand gap rate of any future week exceeds the set threshold, perform the reverse adjustment strategy in the preset order, the reverse adjustment strategy includes: c1) preferentially temporarily increase the effective coverage range of the scenario-based high-priority transportation mode: select the transportation mode with sufficient current capacity and an effective period start point and gap period interval ≤5 weeks, and temporarily advance the start point of its effective period by 1-2 weeks; c2) if the demand gap rate still does not meet the standard after step c1 is performed, temporarily shorten the basic time limit of the scenario-based low-priority transportation mode: select the transportation mode with the lowest absolute priority, temporarily lower the value of its basic time limit by a predetermined percentage, and advance the start point of its effective period by 1-2 weeks; c3) if the demand gap rate exceeds the limit threshold, temporarily increase the priority of air transportation; Step 334, the parameter changes generated after the reverse adjustment strategy is executed are only used for the current round of stock demand calculation.

10. The multi-module dynamic collaborative inventory needs calculation method of claim 1, wherein, The step 4 specifically includes: Step 41, the stock demand coordination calculation module synchronizes the calculation results including demand gap rate analysis and adjustment suggestions to the business visualization system through the data bus after completing the calculation for display; Step 42, control the response period from data update to calculation result output within the set time.

Images