General material inventory control method and system based on robust stochastic optimization
Through a robust stochastic optimization method, combined with LSTM network and exponential smoothing method, the safety inventory of general materials is predicted, which solves the prediction accuracy and efficiency problems of enterprises when setting up safe inventory, and achieves lower inventory costs and better High production stability.
Patent Information
- Application Number
- CN202411137659.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-19
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-08-19
AI Technical Summary
When enterprises set up safe inventory of general materials, they lack scientific and effective mathematical models, which leads to the inability to accurately predict demand changes, leading to the problems of material shortage and inventory backlog, and lack of an effective early warning mechanism, which affects production stability and delivery timeliness. .
Using a robust random optimization method, by obtaining historical data, verifying the data characteristics of material demand and supply lead time, using random optimization or robust optimization model, combining LSTM network and exponential smoothing method, predicting the security of future decision cycles Inventory quantity, consider minimizing costs and constraints, and optimize inventory decisions.
Improves the accuracy and efficiency of safety inventory forecasts, reduces material shortage losses and inventory costs, enhances production efficiency and system reliability, can cope with changes in uncertainties, reduce risks, and is suitable for dynamic customer demand and supplier delays .
Smart Images

Figure CN119027028A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of inventory management, and in particular to a general material inventory control method and system based on robust stochastic optimization. Background Art
[0002] The statements in this section merely mention background art related to the present invention and do not necessarily constitute prior art.
[0003] In order to prevent uncertain factors, such as demand fluctuations, supplier lead time extensions, etc., and to ensure the smooth implementation of production plans, companies need to maintain an appropriate amount of safety stock outside of demand plans for general materials during the procurement cycle. Once there is a large fluctuation in demand or a delay or interruption in supplier supply, the safety stock can be relied on to maintain the normal operation of production activities for a certain period of time. The more safety stock there is, the more guaranteed production and delivery will be, but the more inventory funds will be occupied, and the company itself will have to bear the storage costs of the safety stock; the less safety stock there is, the less inventory funds the company will occupy, but the risks of production stability and delivery timeliness will increase. Therefore, for any company that uses safety stock, setting an appropriate safety stock value is a difficult decision-making problem.
[0004] At present, the safety stock setting standards for general materials in enterprises are mostly based on the staff's rough estimate of quantity based on historical material usage experience, and the production scheduling mode of timely replenishment after consumption is adopted. The production is based on experience and there is a lack of scientific and effective mathematical model calculations. It is impossible to adjust based on the predicted changes in customer demand, which often results in a shortage of required materials and a backlog of temporarily unneeded materials. When customer demand changes, there is a lack of effective early warning mechanism, and the staff is busy dealing with the current problems. There are dual risks of sluggish materials and finished products and failure to meet customer delivery requirements. Summary of the invention
[0005] In order to address the deficiencies in the prior art, the present invention provides a general material inventory control method, system, electronic device, computer-readable storage medium and computer program product based on robust stochastic optimization, which combines stochastic optimization with robust optimization, predicts safety inventory according to the data characteristics of material demand and delivery lead time, and improves prediction accuracy.
[0006] In a first aspect, the present invention provides a general material inventory control method based on robust stochastic optimization;
[0007] A general material inventory control method based on robust stochastic optimization, comprising:
[0008] Obtain material demand and supply lead time in the historical period and conduct statistical analysis to verify whether the data characteristics of material demand and supply lead time conform to the normal distribution;
[0009] According to the verification results, the material demand and supply lead time in the historical time period are processed through stochastic optimization or safety stock robust optimization model to obtain the safety stock for future decision cycles.
[0010] In some implementations, if the data characteristics of material demand and supply lead time conform to normal distribution, the material demand and supply lead time in the historical time period are processed by random optimization, specifically:
[0011] Based on the average and standard deviation of material demand in the historical time period and the average and standard deviation of supply lead time, combined with the service level coefficient, the safety stock volume for the future decision cycle is obtained.
[0012] In some implementations, the safety stock amount for future decision cycles is expressed as:
[0013]
[0014] Where SS is the safety stock of the future decision cycle, z is the service level coefficient, μ L is the average value of the delivery lead time in the historical period, σ D is the standard deviation of supply and demand in the historical period, μ D is the average value of supply and demand in the historical period, σ L is the standard deviation of the supply lead time during the historical period.
[0015] In some implementations, if the data characteristics of material demand and supply lead time do not conform to normal distribution, the material demand and supply lead time in the historical time period are processed by the safety stock robust optimization model, specifically including:
[0016] Based on the material demand and supply lead time in the historical time period, obtain the material demand forecast and supply lead time forecast for the future decision cycle;
[0017] With the goal of minimizing the ordering cost, inventory holding cost, extension cost and shortage cost of safety stock, an uncertainty set is introduced. Under the constraints of urgent ordering, delivery delay and maximum material inventory level, the predicted value of material demand and the predicted value of supply lead time in the future decision cycle are used as input for iterative calculation to determine the safety stock in the future decision cycle.
[0018] In some embodiments, the method of obtaining the predicted material demand value and the predicted supply lead time for a future decision cycle based on the material demand quantity and the supply lead time in a historical time period is specifically as follows: the material demand quantity and the supply lead time in the historical time period are processed by a trained LSTM network or exponential smoothing method to obtain the predicted material demand value and the predicted supply lead time for the future decision cycle.
[0019] In some embodiments, the uncertainty set is the uncertain fluctuation range of material demand and supply lead time in the future decision cycle, which is determined based on the difference between the material demand and supply lead time in the historical time period and the corresponding material demand forecast value and supply lead time forecast value at the same time.
[0020] In a second aspect, the present invention provides a general material inventory control system based on robust stochastic optimization;
[0021] A general material inventory control system based on robust stochastic optimization, comprising:
[0022] The normal distribution verification module is configured to: obtain the material demand and supply lead time in the historical time period and perform statistical analysis to verify whether the data characteristics of the material demand and supply lead time conform to the normal distribution;
[0023] The general material inventory control module is configured to: process the material demand and supply lead time in the historical time period through stochastic optimization or safety stock robust optimization model according to the verification result to obtain the safety stock in the future decision cycle.
[0024] In a third aspect, the present invention provides an electronic device;
[0025] An electronic device comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above-mentioned general material inventory control method based on robust stochastic optimization.
[0026] In a fourth aspect, the present invention provides a computer-readable storage medium;
[0027] A computer-readable storage medium stores a computer program / instruction thereon, which, when executed by a processor, implements the steps of the general material inventory control method based on robust stochastic optimization as claimed in the claim.
[0028] In a fifth aspect, the present invention provides a computer program product;
[0029] A computer program product comprises a computer program / instruction, which implements the steps of the above-mentioned general material inventory control method based on robust stochastic optimization when executed by a processor.
[0030] Compared with the prior art, the present invention has the following beneficial effects:
[0031] 1. The technical solution provided by the present invention selects the corresponding safety stock prediction method according to the data characteristics of material demand and supply lead time. On the one hand, the prediction is made according to the data development characteristics, so as to improve the accuracy of the safety stock prediction, reduce the enterprise inventory cost, reduce the material shortage loss caused by uncertainty, and improve the production efficiency; on the other hand, there is no need to predict the material demand and supply lead time of all general materials in the future decision cycle, and the safety stock can be calculated directly according to the data distribution characteristics, thereby improving the efficiency of the safety stock prediction.
[0032] 2. The technical solution provided by the present invention uses random optimization to process the material demand and supply lead time in a certain historical period for material data that conforms to the normal distribution, and extracts its data distribution characteristics to achieve safety inventory decisions. The data processing process is simple and has strong real-time performance.
[0033] 3. The technical solution provided by the present invention adopts a robust optimization modeling method for material data whose material demand and supply lead time do not conform to the normal distribution within a certain historical period, combines demand and lead time forecasts with inventory decisions, and reduces the negative impact caused by uncertainty by considering the worst case scenario, thereby reducing risks. It can cope with various uncertainties and changes, so that decisions are still effective in the face of an unknown future, improve the reliability and flexibility of the system, and is usually easy to understand and implement. It does not require complex mathematical models or calculations, is more in line with the actual situation of the enterprise, and reflects the true status of the enterprise's inventory development. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.
[0035] Figure 1 A schematic diagram of a process flow provided by an embodiment of the present invention;
[0036] Figure 2 A schematic diagram of LSTM model parameters provided in an embodiment of the present invention;
[0037] Figure 3 A schematic diagram of the fitting effect of LSTM model training data provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0038] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.
[0039] It should be noted that the terms used herein are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that the terms "include" and "have" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0040] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.
[0041] Embodiment 1
[0042] The existing safety inventory forecast of general materials cannot be applied to uncertain factors such as dynamically changing customer demand and customer forecast quantity, and the forecast accuracy is low; therefore, the present invention provides a general material inventory control method based on robust stochastic optimization, which organically combines historical data analysis, long-short term memory network prediction, exponential smoothing prediction, stochastic optimization and robust optimization to achieve safety inventory control.
[0043] Next, combine Figure 1-Figure 3 , a general material inventory control method based on robust stochastic optimization disclosed in this embodiment is described in detail. The general material inventory control method based on robust stochastic optimization includes the following steps:
[0044] S1. Obtain the material demand and supply lead time in the historical time period and perform statistical analysis to verify whether the data characteristics of the material demand and supply lead time conform to the normal distribution; if so, execute S2; if not, execute S3.
[0045] In this embodiment, the material demand and supply lead time in the historical time period are obtained by analyzing and processing the real historical inventory data of the enterprise.
[0046] Among them, historical inventory data includes the daily input quantity, return quantity, purchase acceptance quantity, purchase acceptance time, purchase order issuance time, purchase order net value, purchase quantity, fixed cost of warehouse management, order amount corresponding to the material and delay penalty ratio, and maximum inventory quantity of each material; the specific calculation method is:
[0047] Material demand = input quantity - return quantity;
[0048] Delivery lead time = purchase receipt time - purchase order issuance time;
[0049] Unit ordering cost = net purchase order value / purchase quantity;
[0050] Unit inventory holding cost = fixed cost of unit material warehouse management + unit ordering cost;
[0051] Unit emergency order cost = 2 × unit order cost;
[0052] Unit delay penalty cost = order amount × delay penalty ratio.
[0053] For example, the existing normal distribution test algorithm is run in Python to perform a normal distribution test, and the normal distribution test coefficient p is output; if p≤0.05, the material demand or delivery lead time of the material conforms to the normal distribution, and the standard deviation σ of the material demand D is calculated. D , the average value of material demand D μ D , the average value of the delivery lead time L μ L and the standard deviation σ of the delivery lead time L L , execute S2; if p>0.05, the material demand or delivery lead time of the material does not conform to the normal distribution, execute S3.
[0054] S2. Process the material demand and supply lead time in the historical time period through random optimization to obtain the safety stock in the future decision cycle.
[0055] If the data characteristics of material demand and delivery lead time conform to normal distribution, accurate prediction of safety stock can be achieved based on its data distribution characteristics.
[0056] As an implementation method, the safety stock of the future decision cycle is obtained based on the average value and standard deviation of the material demand in the historical time period and the average value and standard deviation of the supply lead time, combined with the service level coefficient, as shown below:
[0057]
[0058] Where SS is the safety stock of the future decision cycle, z is the service level coefficient, μ L is the average value of the delivery lead time in the historical period, σ D is the standard deviation of supply and demand in the historical period, μ D is the average value of supply and demand in the historical period, σ L is the standard deviation of the supply lead time during the historical period.
[0059] In this embodiment, the service level corresponds to the out-of-stock rate, which expresses the extent to which the fluctuations in demand and supply lead time can be met, and is determined by the decision maker. If the service level is set to 95%, it means that the set safety stock has a 95% probability of meeting the fluctuations in demand and supply lead time. The service level coefficient z can be obtained by looking up the "Correspondence Table of Service Level and Safety Factor Z Value of Normal Distribution". If the service level is 95%, the corresponding z value is 1.645.
[0060] S3. Use the safety stock robust optimization model to process the material demand and supply lead time in the historical time period to obtain the safety stock in the future decision cycle. Specifically include:
[0061] S301. Obtain material demand forecast values and supply lead time forecast values for future decision cycles based on material demand quantities and supply lead time in a historical time period.
[0062] In some embodiments, a time threshold is set, and the material data information within the historical time period is used as training data to construct a training set and a test set. The LSTM network is trained using the training set, and the test set is input into the trained LSTM network, processed by the first exponential smoothing method, the second exponential smoothing method, and the third exponential smoothing method to obtain the corresponding predicted value, which is compared with the true value within the historical time period to obtain the error between the true value and the predicted value; the method with the smallest error is selected to predict the material demand and delivery lead time in the next decision cycle.
[0063] If the error of the LSTM network is less than that of the exponential smoothing method, the material demand and supply lead time in the historical time period are processed through the trained LSTM network, and the demand forecast value at a certain time point in the future decision cycle is output to obtain the material demand forecast value and supply lead time forecast value of the future decision cycle.
[0064] Among them, training the LSTM network specifically includes:
[0065] First, a time threshold is set, and the material data information within the time period is used as training data. The data is preprocessed, such as data cleaning (processing missing values or outliers), data normalization (normalization or standardization), and possible feature engineering (such as calculating the moving average or seasonal characteristics of demand); second, the time series data is divided into a series of fixed lengths to create a data set for training and prediction; third, an LSTM model is constructed. The model architecture defined in this embodiment is detailed in Figure 2, using Adam as the optimizer and mean square error (MSE) as the loss function; fourth, train, evaluate, and adjust the model, evaluate the model performance based on the loss function, and adjust the model parameters and architecture; fifth, input future time series batches into the model and output the predicted data; sixth, denormalize the predicted data and deploy the model to the production environment.
[0066] At the same time, the error between the predicted value and the true value obtained when training the LSTM network is removed, and the fluctuation range of the error is set as the fluctuation range of the demand and lead time uncertainty in the next decision cycle, that is, the upper and lower limits of the uncertainty set.
[0067] By setting an uncertain set, a certain amount of flexibility can be provided for the company's inventory decisions, so that the company will not suffer large losses due to fluctuations in demand.
[0068] If the error of the LSTM network is greater than that of the exponential smoothing method, the material demand and supply lead time in the historical time period are processed by the exponential smoothing method to obtain the material demand forecast value and supply lead time forecast value for the future decision cycle.
[0069] Among them, the test set is input into the code of the first exponential smoothing method, the second exponential smoothing method and the third exponential smoothing method programmed in Python for processing, and the corresponding prediction results MSE are obtained respectively.
[0070] Firstly, the MSE and seasonal trend charts of the forecast results obtained by the first exponential smoothing method, the second exponential smoothing method and the third exponential smoothing method were compared and judged respectively. It was observed that both the demand and the lead time had seasonal trends, and the exponential smoothing method with the smallest MSE was selected as the forecasting method. Secondly, the MSE of the forecast results under different α were compared respectively, and the α value with the smallest MSE was selected as the model parameter.
[0071] In this embodiment, the triple exponential smoothing method is selected.
[0072] S302. Taking the predicted value of material demand and the predicted value of supply lead time in the future decision cycle as input, and taking minimizing the ordering cost, inventory holding cost, extension cost and shortage cost of safety stock as the goal, an uncertainty set is introduced. Under the constraints of urgent ordering, delivery delay and maximum material inventory level, the safety stock amount in the future decision cycle is determined by iterative calculation through the safety stock robust optimization model.
[0073] For example, let Indicates that in the tth decision cycle, when the demand is Supplier delivery lead time is Safety stock is set to The total cost function at the time of decision-making is used to set the mathematical model of the safety stock in the next decision cycle using robust optimization:
[0074]
[0075] In the formula, the total cost includes the ordering cost of safety stock, inventory holding cost, emergency ordering cost caused by actual demand exceeding forecast demand and consuming safety stock, and delay penalty cost caused by supplier delayed delivery.
[0076] According to the above definition, the model can be transformed into:
[0077]
[0078] Where: is the average unit ordering cost of material i in the tth period; is the average unit inventory holding cost of material i in the tth period; is the average unit emergency ordering cost after material i is out of stock in the tth period; is the average penalty cost per unit after material i is delayed in the tth period; Forecast demand for material i in period t; The average daily consumption of material i in the tth period; The predicted lead time for material i in period t; is the maximum inventory of material i; and All are intermediate variables.
[0079] The second and third formulas indicate that emergency ordering will be triggered only if actual demand exceeds forecast demand and consumes all safety stock (i.e., emergency ordering constraint). The fourth and fifth formulas indicate that emergency ordering will be triggered only if supplier delivery is delayed. When the total consumption in a day exceeds the safety stock, the delay penalty (i.e., delivery delay constraint) is triggered, where: It can be obtained based on the material requirement plan. The sixth formula means that the safety stock cannot exceed the maximum inventory level of the material and cannot be less than 0 (that is, the maximum inventory level constraint of the material).
[0080] Assume that the actual demand and actual lead time distribution belong to the interval uncertainty set defined as follows:
[0081] Definition 1. Actual demand is the uncertainty coefficient The value range is:
[0082]
[0083] In the formula, Represents the actual demand for material i, which is a random variable with unknown distribution; represents the corresponding predicted value; Indicates the corresponding fluctuation range; and Indicates the corresponding upper and lower limits of the fluctuation range; η j The role of is to adjust the conservatism of the robust optimization solution and 0≤η j ≤1,j=1,2,…,m.
[0084] Definition 2. Actual lead time is the uncertainty coefficient The value range is:
[0085]
[0086] In the formula, represents the actual lead time of material i, which is a random variable with unknown distribution; represents the corresponding predicted value; Indicates the corresponding fluctuation range; and Represents the corresponding upper and lower limits of the fluctuation range; ξ j The role of is to adjust the conservatism of the robust optimization solution and 0≤ξ j ≤1, j=1,2,…,m. In the actual operation process, for the convenience of calculation, η is set according to the need for conservatism. j and j Take η and ξ as constants.
[0087] According to the above basic model construction and interval uncertainty set definition, the interval uncertainty set is introduced into the objective function of the total cost of safety inventory, and we can get:
[0088]
[0089]
[0090] The upper and lower limits of demand and lead time fluctuations given in the forecast interval are brought into the model, and the conservatism of robust optimization is adjusted by setting the adjustment coefficient of robust optimization. Finally, Python is used to call the gurobi solver for solution.
[0091] Next, in order to verify the advancedness of the method described in this embodiment, this embodiment uses historical inventory data containing 300 common materials of the enterprise to conduct experiments. The data are as follows:
[0092] The data used includes the historical inventory data of 300 common materials of the company, including the daily material input, material return, purchase acceptance, purchase acceptance time, purchase order issuance time, purchase order net value, and purchase quantity of each material. The company also determined the fixed cost of warehouse management, the order amount corresponding to the material and the penalty ratio for delays, the maximum inventory and other data, and calculated the required data such as demand quantity, lead time and cost from 2022.01.01 to 2022.12.31.
[0093] The data from 2022.01.01 to 2022.10.31 are selected as the training set. In the LSTM model, the training fitting effect is detailed in the attached Figure 3 At the same time, it is observed that the MSE of the triple exponential smoothing prediction is the smallest, and when α1 = 0.1, the MSE is the smallest, and the demand forecast result is closest to the true value. When α2 = 0.4, the MSE is the smallest, and the lead time forecast result is closest to the true value. Therefore, the triple exponential smoothing method is used for prediction, and the parameters of the prediction model are set to 0.1 and 0.4 respectively.
[0094] In order to obtain the demand forecast values for November and December 2022, the forecast time step is set to 60 in the trained LSTM model and triple exponential smoothing model, and the forecast results are output and analyzed. At the same time, the fluctuation range of demand and lead time is further calculated based on the forecast results.
[0095] After solving the safety stock calculation model based on the combination of stochastic optimization and robust optimization, the total inventory of 300 kinds of materials in this example was reduced by 59.83% compared with before the improvement, and the total inventory holding cost was reduced by 1.26 million yuan.
[0096] Embodiment 2
[0097] This embodiment discloses a general material inventory control system based on robust stochastic optimization, including:
[0098] The normal distribution verification module is configured to: obtain the material demand and supply lead time in the historical time period and perform statistical analysis to verify whether the data characteristics of the material demand and supply lead time conform to the normal distribution;
[0099] The general material inventory control module is configured to: process the material demand and supply lead time in the historical time period through stochastic optimization or safety stock robust optimization model according to the verification result to obtain the safety stock in the future decision cycle.
[0100] It should be noted that the above normal distribution verification module and general material inventory control module correspond to the steps in Embodiment 1, and the examples and application scenarios implemented by the above modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above Embodiment 1. It should be noted that the above modules, as part of the system, can be executed in a computer system such as a set of computer executable instructions.
[0101] Embodiment 3
[0102] Embodiment 3 of the present invention provides an electronic device, including a memory and a processor, and computer instructions stored in the memory and executed on the processor. When the computer instructions are executed by the processor, the steps of the above-mentioned general material inventory control method based on robust stochastic optimization are completed.
[0103] Embodiment 4
[0104] Embodiment 4 of the present invention provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, the steps of the above-mentioned general material inventory control method based on robust stochastic optimization are completed.
[0105] Embodiment 5
[0106] Embodiment 5 of the present invention provides a computer program product, including a computer program / instruction, which implements the steps of the above-mentioned general material inventory control method based on robust stochastic optimization when executed by a processor.
[0107] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0108] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0109] These computer program instructions can also be loaded onto a computer or other programmable data processing device, and a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0110] The description of each embodiment in the above embodiments has different emphases. For parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0111] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A general material inventory control method based on robust stochastic optimization, characterized in that: include: Obtain material demand and supply lead time in the historical period and conduct statistical analysis to verify whether the data characteristics of material demand and supply lead time conform to the normal distribution; According to the verification results, the material demand and supply lead time in the historical time period are processed through stochastic optimization or safety stock robust optimization model to obtain the safety stock for future decision cycles.
2. The general material inventory control method based on robust stochastic optimization according to claim 1, characterized in that: If the data characteristics of material demand and supply lead time conform to normal distribution, the material demand and supply lead time in the historical time period are processed through random optimization, specifically: Based on the average and standard deviation of material demand in the historical time period and the average and standard deviation of supply lead time, combined with the service level coefficient, the safety stock volume for the future decision cycle is obtained.
3. The general material inventory control method based on robust stochastic optimization as claimed in claim 2, characterized in that: The safety stock for future decision cycles is expressed as: Where SS is the safety stock of the future decision cycle, z is the service level coefficient, μ L is the average value of the delivery lead time in the historical period, σ D is the standard deviation of supply and demand in the historical period, μ D is the average value of supply and demand in the historical period, σ L is the standard deviation of the supply lead time during the historical period.
4. The general material inventory control method based on robust stochastic optimization according to claim 1, characterized in that: If the data characteristics of material demand and supply lead time do not conform to the normal distribution, the material demand and supply lead time in the historical time period are processed through the safety stock robust optimization model, including: Based on the material demand and supply lead time in the historical time period, obtain the material demand forecast and supply lead time forecast for the future decision cycle; With the goal of minimizing the ordering cost, inventory holding cost, extension cost and shortage cost of safety stock, an uncertainty set is introduced. Under the constraints of urgent ordering, delivery delay and maximum material inventory level, the predicted value of material demand and the predicted value of supply lead time in the future decision cycle are used as input for iterative calculation to determine the safety stock in the future decision cycle.
5. The general material inventory control method based on robust stochastic optimization according to claim 4, characterized in that: The method of obtaining the predicted material demand value and the predicted supply lead time of the future decision cycle based on the material demand quantity and the supply lead time in the historical time period is specifically as follows: the material demand quantity and the supply lead time in the historical time period are processed by a trained LSTM network or exponential smoothing method to obtain the predicted material demand value and the predicted supply lead time of the future decision cycle.
6. The general material inventory control method based on robust stochastic optimization according to claim 4, characterized in that: The uncertainty set is the uncertain fluctuation range of material demand and supply lead time in the future decision cycle, which is determined based on the difference between the material demand and supply lead time in the historical time period and the corresponding material demand forecast value and supply lead time forecast value at the same time.
7. A general material inventory control system based on robust stochastic optimization, characterized in that: include: The normal distribution verification module is configured to: obtain the material demand and supply lead time in the historical time period and perform statistical analysis to verify whether the data characteristics of the material demand and supply lead time conform to the normal distribution; The general material inventory control module is configured to: process the material demand and supply lead time in the historical time period through stochastic optimization or safety stock robust optimization model according to the verification result to obtain the safety stock in the future decision cycle.
8. An electronic device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the general material inventory control method based on robust stochastic optimization according to any one of claims 1 to 6.
9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instruction is executed by a processor, the steps of the general material inventory control method based on robust stochastic optimization described in any one of claims 1 to 6 are implemented.
10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the steps of the general material inventory control method based on robust stochastic optimization described in any one of claims 1 to 6 are implemented.
Citation Information
Patent Citations
Retail industry inventory optimization analysis method
CN110322203A
Inventory path optimization method and device based on distribution robustness, equipment and medium
CN117634716A
Robust inventory management in multi-stage inventory networks with demand shocks
US20130159045A1
Ally-Adversary Bimodal Resource Allocation Optimization
US20240135312A1
Cited By
Production resource management and control method and system based on data analysis
CN119477171A
