Supply chain management method and system for cloud storage

Abstract

The invention relates to the technical field of cloud storage management, in particular to a supply chain management method and system for cloud storage, and the system discloses a corresponding warehouse scene building module which is used for determining each entity warehouse for the cloud storage and building a warehouse supply chain simulation scene corresponding to each entity warehouse; the warehouse supply chain management process generation module is used for regularly generating a warehouse supply chain management process and acquiring inventory management prediction data and supply chain order prediction data corresponding to each entity warehouse; the management recommendation landing scheme generation module is used for sequentially generating management recommendation landing schemes of all the entity warehouses according to the warehouse supply chain management process, and the change demand degree of the management recommendation landing schemes of all the entity warehouses in cloud storage is intelligently sensed through the regularly generated warehouse supply chain management process; the entity warehouse which needs to generate a new management recommendation landing scheme is intelligently identified, and the problem of resource distribution blindness in cloud warehouse management is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud warehousing management, and more specifically, to a supply chain management method and system for cloud warehousing. Background Art

[0002] In recent years, with the rapid development of cloud computing and big data technologies, the cloud warehousing management model has gradually become an important direction in the field of supply chain management. Traditional cloud warehousing management systems mostly adopt static resource allocation and passive response mechanisms, relying on manual experience for inventory scheduling and order processing, and it is difficult to cope with the dynamically changing supply chain demands. Especially in the scenario of multi-entity warehouse collaboration, due to the lack of in-depth analysis of the personalized operation requirements of warehouses, problems such as uneven resource allocation, inventory backlog or shortage, and order response delay often occur, resulting in low overall efficiency of the supply chain.

[0003] In the prior art, some solutions attempt to optimize warehousing management through data collection and visualization technologies, but there are still significant defects: insufficient solution adaptability: most systems adopt a "one-size-fits-all" management strategy and do not adapt personalized solutions according to the operation characteristics of different entity warehouses (such as geographical location, category characteristics, logistics network), resulting in resource waste; lack of dynamic optimization ability: there is a lack of a multi-dimensional evaluation mechanism for the implementation effect of management solutions, and it is impossible to actively identify the risk of solution failure and trigger the optimization process in a timely manner.

[0004] In view of the above problems, the present invention proposes a supply chain management method and system for cloud warehousing to achieve the transformation from "experience-driven" to "data intelligent-driven". Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a supply chain management method and system for cloud warehousing.

[0006] To achieve the above purpose, the present invention provides the following technical solutions: A supply chain management method for cloud warehousing, the steps are as follows: Step 1: Determine each entity warehouse targeted by the cloud warehousing; Step 2: Build a warehouse supply chain simulation scenario corresponding to each entity warehouse; Step 3: Regularly generate a warehouse supply chain management process; Step 4: Generate a management recommended implementation plan for each entity warehouse.

[0007] Furthermore, a supply chain management system for cloud warehousing includes A corresponding warehouse scenario building module, which determines each entity warehouse targeted by the cloud warehousing and builds a warehouse supply chain simulation scenario corresponding to each entity warehouse; Warehouse supply chain management process generation module, which regularly generates a warehouse supply chain management process, and obtains the inventory management forecast data and supply chain order forecast data corresponding to each physical warehouse; Management recommendation implementation plan generation module, which sequentially generates the management recommendation implementation plans for each physical warehouse according to the warehouse supply chain management process.

[0008] Furthermore, the steps for regularly generating the warehouse supply chain management process are as follows: regularly obtain the change values of the supply chain management plans for each physical warehouse, sort all physical warehouses in descending order of the values of the supply chain management plan change values, and determine the warehouse supply chain management process according to the sorting.

[0009] Furthermore, the steps for regularly obtaining the change value of the supply chain management plan for a physical warehouse are as follows: select a physical warehouse, obtain the management recommendation implementation plans for the previous consecutive p cycles of this physical warehouse, obtain the implementation differences between the management recommendation implementation plans for every two adjacent cycles, sum up all the implementation differences and calculate the average value, and calculate the change value of the supply chain management plan for this physical warehouse.

[0010] Furthermore, the steps for obtaining the implementation difference between the management recommendation implementation plans for two adjacent cycles are as follows: label the management recommendation implementation plans for two adjacent cycles as management recommendation implementation plan M and management recommendation implementation plan N respectively, obtain the strategy parameters of management recommendation implementation plan M and the strategy parameters of management recommendation implementation plan N, synchronously import the strategy parameters of management recommendation implementation plan M and the strategy parameters of management recommendation implementation plan N into the implementation comparison model, and the implementation comparison model exports the implementation difference value.

[0011] Furthermore, the steps for regularly generating the inventory management forecast data and supply chain order forecast data corresponding to a physical warehouse are as follows: select a physical warehouse, regularly collect the inventory management data and supply chain order data corresponding to this physical warehouse, and collect the inventory management data and supply chain order data corresponding to the previous consecutive i cycles of this physical warehouse. Combine the i + 1 inventory management data into a time series set in a time series manner, combine the i + 1 supply chain order data into a time series set in a time series manner, import the time series set corresponding to the inventory management data into the inventory management data prediction model of this physical warehouse, and the inventory management data prediction model exports the inventory management forecast data. Import the time series set corresponding to the supply chain order data into the supply chain order data prediction model of this physical warehouse, and the supply chain order data prediction model exports the supply chain order forecast data.

[0012] Further, the steps for generating the implementation recommendation plan for the physical warehouse are as follows: Select a physical warehouse, import the inventory management prediction data and supply chain order prediction data corresponding to the physical warehouse into the warehouse supply chain simulation scenario, control the warehouse supply chain simulation scenario to conduct multiple simulations, generate a warehouse supply chain management plan for each simulation, obtain the plan implementation evaluation value of each warehouse supply chain management plan, set the plan implementation evaluation threshold. When the plan implementation evaluation value of the warehouse supply chain management plan is greater than or equal to the plan implementation evaluation threshold, mark the corresponding warehouse supply chain management plan as a management feasible plan. Further, obtain the comprehensive data fluctuation implementation value of each management feasible plan, and mark the management feasible plan with the largest comprehensive data fluctuation implementation value as the management recommendation implementation plan.

[0013] Further, the steps for generating the plan implementation evaluation value of the warehouse supply chain management plan are as follows: The warehouse supply chain simulation scenario conducts simulations based on a warehouse supply chain management plan. When all supply chain orders in the warehouse supply chain simulation scenario are processed, collect various supply chain management indicators of the warehouse supply chain simulation scenario, combine the various supply chain management indicators into a supply chain management indicator set in the form of a set, import the supply chain management indicator set into the plan implementation evaluation model, and the plan implementation evaluation model exports a plan implementation evaluation value.

[0014] Further, the steps for obtaining the comprehensive data fluctuation implementation value of the management feasible plan are as follows: Obtain the inventory management prediction data and supply chain order prediction data corresponding to the physical warehouse, determine the various detailed data included in the inventory management prediction data and supply chain order prediction data, determine the data fluctuation range of each detailed data, and then randomly generate multiple inventory management prediction fluctuation data and multiple supply chain order prediction fluctuation data. Randomly combine the multiple inventory management prediction fluctuation data and multiple supply chain order prediction fluctuation data to generate multiple random data groups. Obtain the plan implementation evaluation value of each random data group, sum up and calculate the average value of the plan implementation evaluation values of all random data groups to calculate the average random implementation evaluation value. Compare the plan implementation evaluation values of all random data groups pairwise, calculate the absolute difference between the plan implementation evaluation values of the two compared random data groups to calculate the random implementation evaluation volatility value, sum up and calculate the average value of all random implementation evaluation volatility values to calculate the average random implementation evaluation volatility value, and calculate the ratio of the average random implementation evaluation value to the average random implementation evaluation volatility value to calculate the comprehensive data fluctuation implementation value of this management feasible plan.

[0015] Further, the steps for obtaining the implementation evaluation value of the random data group are as follows: Import the management feasible solution, the inventory management prediction fluctuation data of the random data group, and the supply chain order prediction fluctuation data into the warehouse supply chain simulation scenario. When all supply chain orders in the warehouse supply chain simulation scenario are processed according to the management feasible solution, the implementation evaluation value of the random data group can be obtained.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The system and method of the present invention specifically construct the warehouse supply chain simulation scenarios of each physical warehouse in the cloud warehouse, regularly predict the inventory management data and supply chain order data of each physical warehouse by combining time series analysis means, analyze the warehouse supply chain management solutions that adapt to the prediction data of the physical warehouse in cooperation with the warehouse supply chain simulation scenarios, and further analyze multiple warehouse supply chain management solutions that adapt to the physical warehouse, analyze various inventory management and supply chain order situations that the physical warehouse may encounter, select the warehouse supply chain management solution with the best implementation effect in various situations, and through the regularly generated warehouse supply chain management process, intelligently sense the change demand degree of the management recommended implementation solutions of each physical warehouse in the cloud warehouse, intelligently identify the physical warehouses that need to generate new management recommended implementation solutions, solve the problem of blindness in resource allocation in cloud warehouse management, and realize the leapfrog upgrade of the supply chain management of the cloud warehouse from "passive response" to "active optimization". BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is the principle block diagram of the system of the present invention; Figure 2 is the flowchart for generating the management recommended implementation solution; Figure 3 is the flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] Embodiment 1, as Figures 1 to 2 , a supply chain management system for a cloud warehouse, including a corresponding warehouse scenario construction module, a warehouse supply chain management process generation module, and a management recommended implementation solution generation module.

[0019] The corresponding warehouse scenario construction module determines each physical warehouse targeted by the cloud warehouse and constructs the warehouse supply chain simulation scenario corresponding to each physical warehouse.

[0020] The steps to build the warehouse supply chain simulation scenario corresponding to the physical warehouse are as follows: Select a physical warehouse and collect all the basic warehouse data of this physical warehouse (the basic warehouse data includes but is not limited to warehouse layout data, equipment parameters, personnel efficiency, etc. The warehouse layout data can be further refined into the area of the sorting area, the location of the labeling machine, the width of the passage, etc. The equipment parameters can be further refined into the processing efficiency of the labeling machine, the speed of the AGV, the capacity of the shelves, etc. The personnel efficiency can be further refined into the processing speed of the picker / packer, etc. The data source of the warehouse layout data is the CAD drawing, the data source of the equipment parameters is the equipment manual + actual measurement, and the data source of the personnel efficiency is the analysis of the operation records). First, create the physical model of the physical warehouse. The creation steps include drawing the layout and defining the resources. Drawing the layout: Import the CAD drawing into the simulation software, divide the storage area, picking area, sorting area, packaging area, and temporary outbound storage area, and mark the coordinates of each area (such as the starting coordinates of the picking area (10, 10) and the ending coordinates (50, 30)); Drag and drop modules such as "sorting machine", "labeling machine", "AGV charging pile", etc., and set the physical dimensions (such as the space occupied by the labeling machine is 2m × 1m); Defining resources: Personnel: Create resource pools for "pickers" and "packers", and set the quantity (such as 20 pickers), working hours (8 hours per shift, including a 15-minute break); Equipment: Set the number of labeling machines (such as 2 sets), processing time (1.2 seconds per order), the number of AGVs (30 sets), and load capacity (50 kg per set); Model the order processing process, including picking, sorting, packaging, and outbound processes, and then perform single-link calibration and full-link verification. Finally, build the warehouse supply chain simulation scenario corresponding to the physical warehouse. This scenario can simulate the supply chain order processing process of the physical warehouse and needs to be updated at any time according to the actual configuration of the physical warehouse (such as updating the layout area, the number of personnel).

[0021] The warehouse supply chain management process generation module regularly generates a warehouse supply chain management process (in the warehouse supply chain management process, all physical warehouses are sorted in a certain order), and obtains the inventory management prediction data and supply chain order prediction data corresponding to each physical warehouse.

[0022] The steps to regularly generate the warehouse supply chain management process are as follows: Regularly obtain the change values of the supply chain management plans of each physical warehouse, sort all the physical warehouses in descending order of the numerical values of the supply chain management plan change values, and determine the warehouse supply chain management process according to the sorting.

[0023] The steps to regularly obtain the change value of the supply chain management plan of the physical warehouse are as follows: Select a physical warehouse, obtain the implemented management recommended plans of this physical warehouse in the previous consecutive p cycles, obtain the plan implementation differences between every two adjacent cycles of the management recommended plans, sum up all the plan implementation differences and calculate the average value, and calculate the change value of the supply chain management plan of this physical warehouse.

[0024] The steps to obtain the implementation difference value of the implementation plans of management recommendations for two adjacent cycles are as follows: Label the implementation plans of management recommendations for two adjacent cycles as Implementation Plan of Management Recommendation M and Implementation Plan of Management Recommendation N respectively. Obtain the strategy parameters of Implementation Plan of Management Recommendation M and the strategy parameters of Implementation Plan of Management Recommendation N, and synchronously import the strategy parameters of Implementation Plan of Management Recommendation M and the strategy parameters of Implementation Plan of Management Recommendation N into the implementation comparison model. The implementation comparison model exports the implementation difference value.

[0025] The construction method of the implementation comparison model: Build a deep learning model, collect multiple groups of plan strategy parameters. Each group of plan strategy parameters contains the strategy parameters of two implementation plans of management recommendations. Use the groups of plan strategy parameters as basic data to carry out the training work on the built deep learning model. In this process, assign an implementation difference value to each group of plan strategy parameters. The value range of the implementation difference value is set between 0 and 50. The size of the implementation difference value has a clear meaning. The larger the value, the greater the difference between the strategy parameters of the two implementation plans of management recommendations in the group of plan strategy parameters, that is, it is difficult to apply the implementation plan of management recommendations in the previous cycle to the next cycle. If applied, it may cause greater management risks. Then divide multiple groups of plan strategy parameters into a training set, a validation set and a test set according to a specific ratio. The specific division ratio is determined as 60%:20%:20%. First, use the training set to train the deep learning model repeatedly. During the training process, use the validation set to timely verify the performance of the model in the training stage. According to the verification results, adjust the parameters of the model in time and optimize the structure of the model to make it develop in a more accurate and stable direction. Finally, the implementation comparison model is built.

[0026] The steps for regularly generating inventory management prediction data and supply chain order prediction data corresponding to a physical warehouse are as follows: Select a physical warehouse, and regularly collect the inventory management data and supply chain order data corresponding to this physical warehouse (the cycle duration corresponding to the regular period is adjusted and set according to the management refinement requirements of cloud warehousing. The inventory management data specifically includes inventory level, inventory turnover rate, replenishment order execution rate, etc., and the supply chain order data specifically includes order response duration, order volume, on-time delivery rate, mis-delivery rate, order cancellation rate, etc.). Also collect the inventory management data and supply chain order data corresponding to the previous i consecutive cycles of this physical warehouse. Combine the i + 1 inventory management data into a time series set in a time series manner, and combine the i + 1 supply chain order data into a time series set in a time series manner. Import the time series set corresponding to the inventory management data into the inventory management data prediction model corresponding to this physical warehouse, and the inventory management data prediction model exports the inventory management prediction data. Import the time series set corresponding to the supply chain order data into the supply chain order data prediction model corresponding to this physical warehouse, and the supply chain order data prediction model exports the supply chain order prediction data.

[0027] Each physical warehouse contains an inventory management data prediction model and a supply chain order data prediction model. Both the inventory management data prediction model and the supply chain order data prediction model are built based on the LSTM model. Hereinafter, Warehouse No. 1 will be randomly defined, and the brief building steps of the inventory management data prediction model and the supply chain order data prediction model corresponding to Warehouse No. 1 will be disclosed.

[0028] The building steps of the inventory management data prediction model corresponding to Warehouse No. 1 are as follows: Build an LSTM model, collect multiple time series sets corresponding to the inventory management data of Warehouse No. 1, and use the multiple time series sets corresponding to the inventory management data as the basic data to carry out the training work on the built LSTM model. In this process, assign an inventory management prediction data to each time series set corresponding to the inventory management data. The inventory management prediction data is the predicted inventory management data of Warehouse No. 1 in the next cycle. Then divide the multiple time series sets corresponding to the inventory management data into a training set, a validation set, and a test set according to a specific ratio. The specific division ratio is determined to be 60%:20%:20%. Training set (60%): Used for model parameter learning; Validation set (20%): Evaluate the model performance during training, assist in adjusting hyperparameters (such as learning rate, number of neurons in the hidden layer), and avoid overfitting; Test set (20%): After the model training is completed, use independent data that has not participated in the training to evaluate the prediction accuracy and ensure the generalization ability. Finally, the inventory management data prediction model corresponding to Warehouse No. 1 is built.

[0029] The steps for building the prediction model for the supply chain order data corresponding to the No. 1 physical warehouse are as follows: Build an LSTM model, collect multiple time series sets corresponding to the supply chain order data of the No. 1 physical warehouse, use the multiple time series sets corresponding to the supply chain order data as the basic data, and carry out the training work on the built LSTM model. In this process, assign a supply chain order prediction data to each time series set corresponding to the supply chain order data. The supply chain order prediction data is the predicted supply chain order data of the No. 1 physical warehouse in the next cycle. Then, divide the multiple time series sets corresponding to the supply chain order data into a training set, a validation set, and a test set according to a specific ratio. The specific division ratio is determined to be 60%:20%:20%. Training set (60%): Used for model parameter learning; Validation set (20%): Evaluate the model performance during training, assist in adjusting hyperparameters (such as learning rate, number of neurons in the hidden layer), and avoid overfitting; Test set (20%): After the model training is completed, use independent data that has not participated in the training to evaluate the prediction accuracy and ensure the generalization ability. Finally, the prediction model for the supply chain order data corresponding to the No. 1 physical warehouse is built.

[0030] The management recommendation implementation plan generation module generates the management recommendation implementation plans for each physical warehouse in sequence according to the warehouse supply chain management process.

[0031] The steps for generating the management recommendation implementation plan for the physical warehouse are as follows: Select a physical warehouse, import the inventory management prediction data and supply chain order prediction data corresponding to the physical warehouse into the warehouse supply chain simulation scenario (corresponding to the physical warehouse), control the warehouse supply chain simulation scenario to conduct multiple simulations, and generate a warehouse supply chain management plan for each simulation (there are differences in the corresponding policy parameters between different warehouse supply chain management plans, such as different safety inventory coefficients, different replenishment buffer lead times, different order allocation cost weights, different order allocation time efficiency weights. The above policy parameters are all adjustable variables, and different warehouse supply chain management plans are generated by dynamically modifying these policy parameters), obtain the plan implementation evaluation value of each warehouse supply chain management plan, set the plan implementation evaluation threshold (the plan implementation evaluation threshold is scientifically set by combining industry benchmarks and cost-benefit analysis). When the plan implementation evaluation value of the warehouse supply chain management plan is greater than or equal to the plan implementation evaluation threshold, mark the corresponding warehouse supply chain management plan as a management-feasible plan (do not mark in the case of less than). Further, obtain the comprehensive data fluctuation implementation value of each management-feasible plan, and mark the management-feasible plan with the largest comprehensive data fluctuation implementation value as the management recommendation implementation plan.

[0032] The steps to obtain the comprehensive data fluctuation landing value of the management feasible solution are as follows: Obtain the inventory management prediction data corresponding to the entity warehouse and the supply chain order prediction data, and determine the various refined data included in the inventory management prediction data and the supply chain order prediction data (the inventory management prediction data includes various refined data such as inventory level, inventory turnover rate, replenishment order execution rate, etc., and the supply chain order prediction data includes various refined data such as order response duration, order volume, on-time delivery rate, mis-delivery rate, order cancellation rate, etc.), determine the data fluctuation range of each refined data (the data fluctuation range of each refined data is different and is determined comprehensively based on historical data statistical analysis, scenario extreme values, and industry experience parameters to ensure that the fluctuation range reflects both the historical law of the data and covers possible future extreme situations), and then randomly generate multiple inventory management prediction fluctuation data and multiple supply chain order prediction fluctuation data (each inventory management prediction fluctuation data is different, and each supply chain order prediction fluctuation data is also different. Each inventory management prediction fluctuation data is based on the inventory management prediction data and randomly adjusts each refined data within the data fluctuation range, and the corresponding inventory management prediction fluctuation data is generated after adjustment. The same applies to the supply chain order prediction fluctuation data), randomly combine the multiple inventory management prediction fluctuation data and the multiple supply chain order prediction fluctuation data to generate multiple random data groups (each random data group includes an inventory management prediction fluctuation data and a supply chain order prediction fluctuation data), obtain the solution landing evaluation value of each random data group, sum up and calculate the average value of the solution landing evaluation values of all random data groups to calculate the average random landing evaluation value, compare the solution landing evaluation values of all random data groups pairwise, calculate the absolute difference between the solution landing evaluation values of the two compared random data groups to calculate the random landing evaluation volatility value, sum up and calculate the average value of all random landing evaluation volatility values to calculate the average random landing evaluation volatility value, and calculate the ratio of the average random landing evaluation value to the average random landing evaluation volatility value to calculate the comprehensive data fluctuation landing value of this management feasible solution.

[0033] The steps to obtain the solution landing evaluation value of the random data group are as follows: Import the management feasible solution, the inventory management prediction fluctuation data and the supply chain order prediction fluctuation data of the random data group into the warehouse supply chain simulation scenario. When all supply chain orders in the warehouse supply chain simulation scenario are processed according to the management feasible solution, the solution landing evaluation value of the random data group can be obtained.

[0034] The steps for generating the implementation evaluation value of the warehouse supply chain management solution are as follows: The warehouse supply chain simulation scenario is simulated based on a warehouse supply chain management solution. When all supply chain orders in the warehouse supply chain simulation scenario are processed, collect various supply chain management indicators of the warehouse supply chain simulation scenario (supply chain management indicators include but are not limited to the following: inventory cost, order fulfillment rate, inventory turnover rate). Combine the various supply chain management indicators into a supply chain management indicator set in the form of a set, and import the supply chain management indicator set into the implementation evaluation model. The implementation evaluation model exports an implementation evaluation value (which is the implementation evaluation value of the warehouse supply chain management solution).

[0035] The steps for building the implementation evaluation model are as follows: Build a deep learning model, collect multiple supply chain management indicator sets, and each supply chain management indicator set belongs to a warehouse supply chain management solution. Use the supply chain management indicator sets as basic data to carry out training work on the built deep learning model. During this process, assign an implementation evaluation value to each supply chain management indicator set. The value range of the implementation evaluation value is set between 0 and 100, and the size of the implementation evaluation value has a clear meaning. The larger the value, the better the comprehensive performance of the warehouse supply chain management solution corresponding to the supply chain management indicator set in dimensions such as cost, efficiency, and service, and the more suitable it is for actual implementation. Then, divide the multiple supply chain management indicator sets into a training set, a validation set, and a test set according to a specific ratio. The specific division ratio is determined to be 70%:15%:15%. First, use the training set to repeatedly train the deep learning model. During the training process, use the validation set to timely verify the performance of the model during the training stage. According to the verification results, adjust the model parameters in a timely manner and optimize the model structure to make it develop in a more accurate and stable direction. Finally, the implementation evaluation model is built.

[0036] The system specifically builds the warehouse supply chain simulation scenarios of each physical warehouse in the cloud warehouse, regularly combines time series analysis methods to predict the inventory management data and supply chain order data of each physical warehouse, analyzes the warehouse supply chain management solutions that adapt to the predicted data of the physical warehouse in cooperation with the warehouse supply chain simulation scenarios, and further analyzes multiple warehouse supply chain management solutions that adapt to the physical warehouse, analyzes various inventory management and supply chain order situations that the physical warehouse may encounter, selects the warehouse supply chain management solution with the best implementation effect in various situations, and intelligently senses the change demand degree of the recommended implementation solutions for the management of each physical warehouse in the cloud warehouse through the regularly generated warehouse supply chain management process.

[0037] Example 2, as Figure 3 a supply chain management method for cloud warehousing, the steps are as follows: Step 1: Determine each physical warehouse targeted by the cloud warehouse; Step 2: Build a warehouse supply chain simulation scenario corresponding to each physical warehouse; Step 3: Regularly generate the warehouse supply chain management process; Step 4: Generate a management recommendation implementation plan for each physical warehouse.

[0038] The above method intelligently identifies the physical warehouses that need to generate new management recommendation implementation plans, solves the problem of blindness in resource allocation in cloud warehouse management, and realizes the leapfrog upgrade of the supply chain management of cloud warehouses from "passive response" to "active optimization".

[0039] The above formulas are all dimensionless and take their numerical values for calculation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0040] Those of ordinary skill in the art can realize that, in combination with the units and algorithm steps of each example described in the embodiments disclosed in this article, they can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0041] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0042] In several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the displayed or discussed mutual coupling or direct coupling or communication connection can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be in an electrical, mechanical, or other form.

[0043] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A supply chain management method for cloud warehousing, characterized in that The steps are as follows: Step 1: Determine each physical warehouse targeted by the cloud warehouse; Step 2: Build a warehouse supply chain simulation scenario corresponding to each physical warehouse; Step 3: Regularly generate a warehouse supply chain management process; Step 4: Generate a management recommendation implementation plan for each physical warehouse.

2. A supply chain management system for cloud warehousing, which is applied to the supply chain management method for cloud warehousing described in claim 1, and is characterized in that, Including A corresponding warehouse scenario building module that determines each physical warehouse targeted by the cloud warehouse and builds a warehouse supply chain simulation scenario corresponding to each physical warehouse; A warehouse supply chain management process generation module that regularly generates a warehouse supply chain management process and obtains inventory management prediction data and supply chain order prediction data corresponding to each physical warehouse; A management recommendation implementation plan generation module that sequentially generates a management recommendation implementation plan for each physical warehouse according to the warehouse supply chain management process.

3. The supply chain management system for cloud warehousing according to claim 2, characterized in that, The steps for regularly generating the warehouse supply chain management process are as follows: Regularly obtain the change values of the supply chain management plans for each physical warehouse, sort all physical warehouses in descending order of the values of the supply chain management plan change values, and determine the warehouse supply chain management process according to the sorting.

4. The supply chain management system for cloud warehousing according to claim 3, wherein The steps for regularly obtaining the change value of the supply chain management plan for a physical warehouse are as follows: Select a physical warehouse, obtain the management recommendation implementation plans for the previous consecutive p cycles of this physical warehouse, obtain the plan implementation differences between every two adjacent cycles of the management recommendation implementation plans, sum up all the plan implementation differences and calculate the average value to calculate the change value of the supply chain management plan for this physical warehouse.

5. The supply chain management system for cloud warehousing according to claim 4, wherein The steps for obtaining the plan implementation difference between the management recommendation implementation plans of two adjacent cycles are as follows: Mark the management recommendation implementation plans of two adjacent cycles as management recommendation implementation plan M and management recommendation implementation plan N respectively, obtain the policy parameters of management recommendation implementation plan M and the policy parameters of management recommendation implementation plan N, synchronously import the policy parameters of management recommendation implementation plan M and the policy parameters of management recommendation implementation plan N into the plan implementation comparison model, and the plan implementation comparison model exports the plan implementation difference value.

6. The supply chain management system for cloud warehousing according to claim 2, wherein The steps for regularly generating the inventory management prediction data and supply chain order prediction data corresponding to a physical warehouse are as follows: Select a physical warehouse, regularly collect the inventory management data and supply chain order data corresponding to this physical warehouse, and collect the inventory management data and supply chain order data corresponding to the previous consecutive i cycles of this physical warehouse. Combine the i + 1 inventory management data into a time series set in a time series manner, combine the i + 1 supply chain order data into a time series set in a time series manner, import the time series set corresponding to the inventory management data into the inventory management data prediction model corresponding to this physical warehouse, and the inventory management data prediction model exports the inventory management prediction data. Import the time series set corresponding to the supply chain order data into the supply chain order data prediction model corresponding to this physical warehouse, and the supply chain order data prediction model exports the supply chain order prediction data.

7. The supply chain management system for cloud warehousing according to claim 2, characterized in that, The generation steps of the implementation recommendation plan for the physical warehouse are as follows: Select a physical warehouse, import the inventory management prediction data and supply chain order prediction data corresponding to the physical warehouse into the warehouse supply chain simulation scenario, control the warehouse supply chain simulation scenario to conduct multiple simulations, generate a warehouse supply chain management plan for each simulation, obtain the plan implementation evaluation value of each warehouse supply chain management plan, set the plan implementation evaluation threshold. When the plan implementation evaluation value of the warehouse supply chain management plan is greater than or equal to the plan implementation evaluation threshold, mark the corresponding warehouse supply chain management plan as a management feasible plan. Further obtain the comprehensive data fluctuation implementation value of each management feasible plan, and mark the management feasible plan with the largest comprehensive data fluctuation implementation value as the management recommendation implementation plan.

8. The supply chain management system for cloud warehousing according to claim 7, wherein, The generation steps of the plan implementation evaluation value of the warehouse supply chain management plan are as follows: The warehouse supply chain simulation scenario conducts simulations based on a warehouse supply chain management plan. When all supply chain orders in the warehouse supply chain simulation scenario are processed, collect various supply chain management indicators of the warehouse supply chain simulation scenario, combine the various supply chain management indicators into a supply chain management indicator set in the form of a set, import the supply chain management indicator set into the plan implementation evaluation model, and the plan implementation evaluation model exports a plan implementation evaluation value.

9. The supply chain management system for cloud warehousing according to claim 7, wherein, The steps to obtain the comprehensive data fluctuation implementation value of the management feasible plan are as follows: Obtain the inventory management prediction data and supply chain order prediction data corresponding to the physical warehouse, determine the various detailed data included in the inventory management prediction data and the supply chain order prediction data, determine the data fluctuation range of each detailed data, and then randomly generate multiple inventory management prediction fluctuation data and multiple supply chain order prediction fluctuation data. Randomly combine the multiple inventory management prediction fluctuation data and the multiple supply chain order prediction fluctuation data to generate multiple random data groups. Obtain the plan implementation evaluation value of each random data group, sum up and take the average of the plan implementation evaluation values of all random data groups to calculate the average random implementation evaluation value. Compare the plan implementation evaluation values of all random data groups pairwise, calculate the absolute difference between the plan implementation evaluation values of the two compared random data groups to calculate the random implementation evaluation fluctuation value, sum up and take the average of all random implementation evaluation fluctuation values to calculate the average random implementation evaluation fluctuation value, and calculate the ratio of the average random implementation evaluation value to the average random implementation evaluation fluctuation value to calculate the comprehensive data fluctuation implementation value of this management feasible plan.

10. The supply chain management system for cloud warehousing according to claim 9, wherein, The steps to obtain the plan implementation evaluation value of the random data group are as follows: Import the management feasible plan, the inventory management prediction fluctuation data and the supply chain order prediction fluctuation data of the random data group into the warehouse supply chain simulation scenario. When all supply chain orders in the warehouse supply chain simulation scenario are processed according to the management feasible plan, the plan implementation evaluation value of the random data group can be obtained.

Images