Same-level inventory adjustment method based on steady-state inventory theory

Abstract

The same-level inventory adjustment method based on the steady-state inventory theory has the following specific steps: S1. Calculate the existing inventory allocation to obtain the steady-state coefficient SF of the multi-point inventory network; S2. Determine whether the calculated steady-state coefficient SF exceeds the steady-state threshold STV. If so, the algorithm ends; if not, proceed to step S3; S3. Check the number of times the compensation strategy has been implemented. If it exceeds the set number N, place an order with the supplier; if it does not exceed N, proceed to step S4; S4. Calculate the corresponding compensation strategy through the allocation of the existing supply network inventory and adjust the inventory; S5. Calculate according to the allocated supply network inventory allocation to obtain the steady-state coefficient SF of the multi-point inventory network, and then proceed to step S2. This application can improve the guarantee success rate of the entire supply network.

Description

Technical Field

[0001] The present invention belongs to the technical fields of supply network theory, inventory demand analysis, and logistics distribution, and in particular relates to a same-level inventory adjustment method based on steady-state inventory theory. Background Art

[0002] With the rapid development of information technology, logistics supply points can leverage various information-based tools and platforms to share and communicate inventory information, enabling the exchange and transfer of inventory materials, a process known as inventory adjustment. Adjustment can significantly reduce inventory costs, minimize inventory overstocks, and mitigate supply failures in the face of unexpected orders. The essence of inventory adjustment is to transform static inventory into dynamic inventory, shifting traditional inventory management concepts to information-based management. While inventory "moves" in motion, adjustment enables management agencies to make timely adjustments based on changes in material consumption and the consumption environment. From a global perspective, inventory adjustment aims to achieve an overall balance between total inflows and outflows, but individual supply points may experience both inventory overstocks and stock-outs. Structurally, supply points form a network. Supply points are the "nodes" of this network, and "steady state" means that within this network structure, each node's inventory is minimized from stock-outs. Adjustment also considers each supply point's historical availability to ensure that stock-outs are avoided after adjustments are made. While different adjustment algorithms may meet supply point demand after adjustments, there are differences in whether these adjustments will create instability in future supply.

[0003] "Steady-state inventory adjustment" means that the location and elements of the supply nodes or inventory points are determined, and the total inventory remains unchanged. Through the reasonable and effective allocation of inventory ratios, the supply network can achieve the best or better risk resistance, that is, it can respond in time in the event of sudden large orders, so that the probability of out-of-stock is minimized or within an acceptable range, the lead time of customer orders is shortened, and the customer satisfaction rate reaches the highest or relatively high adjustment method.

[0004] The issue of inventory adjustments actually requires certain prerequisites: first, inventory availability; second, a discrepancy between forecasted and actual demand; and third, the existence of an algorithm or mechanism to coordinate arrangements between supply points. The efficiency and results of the adjustment algorithm significantly impact the actual effectiveness of inventory adjustments.

[0005] Demand uncertainty is a key factor in our ability to effectively forecast demand. When demand is difficult to predict, forecast accuracy is low. The discrepancy between actual demand and forecasts is large, requiring a larger safety stock to mitigate the discrepancy.

[0006] Inventory control within a supply network management environment is currently a key component of supply network management. Key inventory control strategies within supply networks include vendor-managed inventory, joint inventory control, and multi-echelon inventory control. The same-level inventory adjustment method, based on steady-state inventory theory, falls under this category of joint inventory control. This method targets same-level inventory systems within a supply network, meaning the adjustment targets are within the same support level within the supply chain. Summary of the Invention

[0007] To determine whether the inventory allocation in the supply system is safe and stable, and if not, to determine how to use a fast algorithm to calculate a feasible inventory allocation strategy within an acceptable time frame, this paper proposes a same-level inventory adjustment method based on steady-state inventory theory. The specific scheme is as follows:

[0008] The same-level inventory adjustment method based on the steady-state inventory theory has the following specific steps:

[0009] S1. Calculate the existing inventory allocation to obtain the steady-state coefficient SF of the multi-point inventory network;

[0010] S2, determine whether the calculated steady-state coefficient SF exceeds the steady-state threshold STV, if yes, the algorithm ends, if not, enter step S3;

[0011] S3. Check the number of times the compensation strategy has been implemented. If it exceeds the set number N, place an order with the supplier. If it does not exceed N, go to step S4;

[0012] S4. Calculate the corresponding replenishment strategy through the distribution of existing supply network inventory and carry out inventory deployment;

[0013] S5. Calculate based on the allocated supply network inventory allocation to obtain the steady-state coefficient SF of the multi-point inventory network, and then proceed to step S2.

[0014] Specifically, the steps for calculating the steady-state coefficient SF in step S1 are:

[0015] S11. Calculate the stability coefficient SFI of the four main factors affecting the stability of the supply network, which are the inventory levels in each supply node. i , the stability coefficient SFS related to the mutual assistance between supply nodes i , the stability coefficient SFO related to the demand situation in the area guaranteed by each supply node i , Stability factor SFP related to the loss cost of out-of-stock i , quantify and calculate these four factors;

[0016] S12. Through the above quantification, the stability factor SFI of the multi-point steady-state inventory network of a certain supply node is i、SFS i 、SFO i 、SFP i Calculate separately and after the calculation is completed, the total stability factor SF of the supply node is obtained by the following formula i :

[0017] SF i =θ×SFI i +τ×SFS i +κ×SFO i +λ×SFP i

[0018] where θ is SFI i The weight of i The weight of i The weight of SFP; λ is the weight of SFP i The weight of

[0019] S13. Stability factor SF of each supply node i After the calculation is completed, the mean is calculated, which is the steady-state coefficient SF of the multi-point steady-state inventory network of the supply network. The formula is as follows:

[0020]

[0021] where θ is SFI i The weight of i The weight of i The weight of i The weight of , where θ+τ+κ+λ=1.

[0022] Specifically, the stability coefficient SFI related to the inventory level in the supply node in step S1 is i The calculation formula is as follows:

[0023]

[0024] Among them I i is the inventory of the supply node; SS i is the safety inventory of the supply node; α is the inventory stability related factor, and α is less than 1.

[0025] Specifically, the stability coefficient SFS related to the mutual assistance between supply nodes i The calculation formula is as follows

[0026]

[0027] Where: P i The success rate of supplying nodes to the other party;j The inventory of the node supplying the other party; d ij is the transportation cost between supply nodes;

[0028] The transportation cost between supply nodes includes the economic cost of transportation and the time cost of transportation, d ij The calculation formula is:

[0029] d ij =A(T ij )γ+B(C ij ) δ

[0030] Among them: A is the weighting factor of time cost; B is the weighting factor of economic cost; T ij is the transportation time between two supply nodes; γ is the correlation factor of transportation time; C ij The transportation cost between two supply nodes; δ is the correlation factor of transportation cost;

[0031] d ij Substitute into the stability coefficient SFS related to the mutual assistance between supply nodes i The formula is obtained from

[0032]

[0033] Specifically, the stability factor SFO related to the demand situation in the area guaranteed by each supply node i The calculation formula is as follows

[0034] SFO i =1-(OP i ) E

[0035] Among them OP i is the probability of demand exceeding the safety stock of the supply node; ε is the risk-related factor of the order.

[0036] Specifically, the stability coefficient SP related to the loss cost of out-of-stock i The calculation formula is as follows

[0037] SFP i =1-(SP i ) ∈

[0038] Among them SP i is the risk level of out-of-stock at the supply node; ∈ is the penalty cost related factor of out-of-stock;

[0039] When the optimal inventory level I is calculated i and expected penalty costs EPC i(Expectation Punishment Cost), calculate the loss cost SP of out-of-stock i The formula is:

[0040]

[0041] Among them, EPC i is the expected penalty fee of the supply node; η is the penalty fee related factor;

[0042] Substitute the supply node risk calculation formula into the penalty cost stability factor SFP i In the formula, we get SFP i formula:

[0043]

[0044] When the expected penalty cost EPC is 0, SFP i is 1.

[0045] Specifically, the steps for obtaining the compensation strategy in step S4 are as follows:

[0046] S41. Calculate the steady-state coefficient SF of each supply node i and the overall steady-state coefficient SF of the entire inventory network;

[0047] S42. Find the supply node i with the worst stability coefficient and the supply nodes adjacent to this node;

[0048] S43. Find a node j with the highest steady-state coefficient among these adjacent supply nodes, and set the steady-state coefficient SF of this supply node to j The steady-state coefficient SF of the target set by the system adj For comparison; if the coefficient SF of the supply node with the highest steady-state coefficient j No SF adj If the stability coefficient of the highest supply node is SF, then it means that there is no suitable supply node around the supply node i for the time being, so return to step S42 to find the supply node with the second worst stability coefficient. j Higher than SF adj , inventory adjustment is performed between these two supply nodes to increase the stability of the total supply network; the adjusted inventory quantity needs to be obtained by the following method: Calculate the inventory increase ΔI required for inventory i to reach ρSF j The maximum output inventory ΔI required for the j inventory to meet the steady-state requirements of ρSF j , compare the absolute values ​​of the two, the inventory allocation amount is the smaller value of the two, where ρ is a parameter between SF adj and between SF.

[0049] The beneficial effects of the present invention are:

[0050] (1) This application aims to address the current situation where demand forecasts cannot accurately meet the generation of inventory plans, and inventory backlogs and inventory shortages coexist. First, the adjustment cost is determined, and then the steady-state coefficient of a single supply node is calculated. On this basis, preventive adjustments are made to improve the guarantee success rate of the entire supply network.

[0051] (2) The steady-state coefficient of each supply node can be calculated across the entire network. First, the node with the worst steady-state coefficient and its adjacent nodes that can be adjusted are found. Then, adjustments are made according to the provided algorithm to complete the compensatory adjustment of the entire network. This can effectively reduce the inventory holding rate of the supply network and improve the supply success rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a flow chart of the same-level inventory adjustment method based on steady-state inventory theory proposed by the present invention.

[0053] Figure 2 Flowchart for remedial strategy. DETAILED DESCRIPTION

[0054] like Figure 1-2 As shown in the figure, the same-level inventory adjustment method based on the steady-state inventory theory includes the adjustment of an online node. The specific steps are as follows:

[0055] S1. Calculate the existing inventory allocation to obtain the steady-state coefficient SF of the multi-point inventory network;

[0056] S2. Determine the calculated steady-state coefficient SF. If it exceeds the steady-state threshold STV (when the stability coefficient reaches a certain value, this value is recorded as the stationary threshold value STV), it can be considered that the inventory allocation is relatively safe and stable. In reality, there is not only one inventory allocation method whose calculated stability coefficient SF exceeds the stability threshold STV. There may be many inventory allocation methods that meet the requirements of a multi-point steady-state inventory network. STV is a parameter that describes the stability of the inventory network), then it means that the supply network already belongs to a multi-point steady-state inventory network, and the algorithm ends. If not, proceed to step S3.

[0057] S3. Check the number of times the compensation strategy has been implemented. If it exceeds a certain number N (this number is determined by the enterprise based on the scale of the problem, that is, the number of nodes n in the supply network), then it means that no matter how many times the supply network is adjusted, it is basically impossible to achieve a multi-point steady-state inventory network. The subsequent compensation strategy will be meaningless. It is recommended to place an order with the supplier. If it does not exceed a certain number N, then it means that there is still the possibility of inventory adjustment, and go to step S4.

[0058] S4. Calculate the corresponding compensation strategy through the allocation of existing supply network inventory and make inventory adjustments.

[0059] S5. Calculate based on the allocated supply network inventory allocation and proceed to step S2 to correct the steady-state coefficient SF of the multi-point inventory network.

[0060] The steps for calculating the steady-state coefficient SF in step S1 are:

[0061] S11. The four main factors affecting the stability of the supply network are the inventory level SFI in each supply node. i , mutual assistance between supply nodes SFS i (often reflected in whether or not to participate in adjustments), the demand situation in the areas guaranteed by each supply node SFO i , loss cost of out-of-stock SFP i , quantify and calculate these four factors, set relevant weighting factors according to the actual situation, and calculate the steady-state coefficient SF of the entire equipment supply network.

[0062] (1) Inventory levels in supply nodes

[0063] Each supply node has its own safety stock (SS). This parameter is calculated based on the demand forecast algorithm according to the previous supply history, which will not be described here. When the actual inventory of a supply node is greater than SS, the excess inventory can be used to participate in the adjustment range of the entire supply network. Obviously, this will lead to an increase in the steady-state coefficient SF of the entire supply network. Therefore, the stability coefficient SFI related to the inventory level of each supply node is i for:

[0064]

[0065] Among them: I i is the inventory of the supply node; SS i is the safety stock of the supply node; α is the inventory stability related factor;

[0066] Regarding α, an increase in inventory level will lead to an increase in the stability of the supply node and the entire supply network, but the increase in inventory level and the increase in stability are not linear. Assume that an increase in inventory level by 500 will increase SFI i Increase by 5, but after the inventory doubles to 1000, SFI i It will not double accordingly. According to daily experience, an increase in inventory will improve the stability of the supply network, but after the inventory increases to a certain level, the increase in stability is not so obvious, so the two are a nonlinear relationship, so α should be less than 1. As for the final α, it needs to be analyzed according to different companies, and the α settings of different companies will not be the same. In particular, when I i Much larger than SS i When SFI i If it is 1, it means that the stability factors related to the inventory of the supply node are relatively good.

[0067] (2) Mutual assistance between supply nodes

[0068] The degree of mutual assistance between supply nodes is sometimes very important. It determines whether a supply node can or is willing to enter the adjustable range. Because the different ranges may artificially restrict the range of adjustment. The factors that determine the degree of mutual assistance include: the transportation cost between supply nodes d ij , the inventory of the other party's supply node I i (Inventory), and the amount of inventory available to support IA i (Inventory Available), the success rate of shipment from the other supply node P i The calculation of the support of this supply node requires the calculation of all other supply nodes and the sum of them. So the support of each supply node is SFS i for:

[0069]

[0070] Where: P i The success rate of supplying nodes to the other party; j The inventory of the node supplying the other party; d ij is the transportation cost between supply nodes.

[0071] Because for the other supply node, if it needs to provide support to the node that is out of stock, it will first ship out the inventory used for support in this supply node. If this inventory is not enough to make up for the vacancy of the crisis supply node, it will also need to use the inventory in the safety stock of this supply node to make up for it when necessary.

[0072] The relationship between the transportation cost and support between supply nodes should also be nonlinear, and the greater the increase in cost, the faster the decrease in support, so β ​​should be greater than 1. And when the support is large, then SFS i If it is 1, it means that the supply node has good support.

[0073] And the transportation cost between supply nodes includes two parts: the economic cost of transportation and the time cost of transportation, so d ij The calculation formula is:

[0074] d ij =A(T ij ) γ +B(C ij ) δ

[0075] Among them: A is the weighting factor of time cost; B is the weighting factor of economic cost; T ij is the transportation time between two supply nodes; γ is the correlation factor of transportation time; C ij The transportation cost between two supply nodes; δ is the correlation factor of transportation cost.

[0076] You can see that in addition to T ij and C ij In addition to the parameters that can be calculated based on actual conditions, the remaining parameters are set according to the different characteristics of different enterprises. Each enterprise has a parameter configuration that suits the characteristics of its own supply network. Substitute the transportation cost formula into the support SFS of the supply node. i In the calculation formula, we can get the new mutual assistance degree SFS i formula:

[0077]

[0078] (3) Demand situation in the area served by the supply node

[0079] The demand situation factor includes two aspects: one is the probability of demand occurring (sometimes it may change suddenly due to emergencies or random events), and the other is the average demand for each demand. Two parameters are defined: the probability of demand occurring OP i (Order Probability) and the mathematical expectation of demand OE i (OrderExpectation). Therefore, the stability factor SFO related to the order occurrence of a certain supply node i for:

[0080] SFO i =1-(OP i) ε

[0081] Among them OP i is the probability of demand exceeding the safety stock of the supply node; ε is the risk-related factor of the order.

[0082] It can be seen that the greater the possibility of generating new demand exceeding the safety stock requirement, the worse the stability of the supply node at that point in the supply network. Therefore, it is necessary to increase the inventory of the supply node and the mutual assistance of the adjacent supply nodes to increase the stability of the supply node. And when the probability of generating new demand exceeding the safety stock of the supply node is the lowest 0, then SFO i If it is 1, it means that the stability factor of the supply node is relatively good.

[0083] (4) Loss costs caused by out-of-stock

[0084] The risk cost of the consequences of out-of-stock situations. For example, some sensitive areas have a basically zero tolerance for out-of-stock situations, while some areas have a higher tolerance for out-of-stock situations. Therefore, different costs should be set according to the tasks undertaken. Although each supply point will try its best to avoid out-of-stock situations, if a certain degree of out-of-stock risk can be exchanged for a considerable reduction in inventory volume, or even a significant reduction in the inventory cost of the overall supply network, then this risk should be allowed to exist. At the same time, different out-of-stock loss costs PC i Punishment Cost will lead to different tolerable risk levels. That is, if the loss cost of a certain supply node is relatively high, then more inventory will be needed to reduce the probability of stockout. If the loss cost of the inventory is not high and the unit inventory holding cost of the supply node is relatively high, then even if there is a risk of stockout, the inventory can be reduced to reduce the total cost of the supply network. Therefore, we can calculate the stability factor SFP of the penalty cost based on this. i :

[0085] SFO i =1-(SP i ) ∈

[0086] Among them SP i is the risk level of out-of-stock at the supply node; ∈ is the penalty cost related factor of out-of-stock.

[0087] The higher the risk factor of out-of-stock, the worse the stability of the supply network. i For example, this is calculated by the enterprise based on the holding cost and penalty cost of the supply node, and SP iThe value range is from 0 to 1. The main idea is: according to a certain inventory level, the out-of-stock probability of the supply node is calculated, and this probability is multiplied by the penalty cost of out-of-stock to obtain the expected penalty cost under the inventory level, and then the holding cost of the inventory level is added to obtain the total cost of the supply node. According to the continuous change of inventory level, different total costs of supply nodes are obtained, and the supply node inventory level under the optimal total cost is obtained by comparison. Since this is not the main research of this article, detailed calculations are not performed. When the optimal inventory level I is calculated i and expected penalty costs EPC i (Expectation Punishment Cost), calculate the loss cost SP of out-of-stock i The formula is:

[0088]

[0089] Among them, EPC i is the expected penalty fee for the supply node; η is the penalty fee related factor.

[0090] The expected penalty fee of the supply node should be proportional to the risk level of the supply node. Therefore, this study uses the penalty fee correlation factor η to establish the relationship between the two. Substitute the supply node risk level calculation formula into the penalty cost stability factor SFP i In the formula, we can also get the new penalty cost stability factor SFP i formula:

[0091]

[0092] In particular, when the expected penalty cost EPC is 0, SFP i If it is 1, the penalty cost stability factor of the supply node is better.

[0093] S12. Through the above quantification, the stability factor SFI of the multi-point steady-state inventory network of a certain supply node is i 、SFS i 、SFO i 、SFP i Calculate separately. After the calculation is completed, the total stability factor SF of the supply node can be obtained by the following formula i :

[0094] SF i =θ×SFI i +τ×SFS i +κ×SFO i +λ×SFP i

[0095] where θ is SFIi The weight of i The weight of i The weight of i The weight of , where θ+τ+κ+λ=1;

[0096] S13. Stability factor SF of each supply node i After the calculation is completed, the mean is calculated, which is the steady-state coefficient SF of the multi-point steady-state inventory network of the supply network. The formula is as follows:

[0097]

[0098] It's worth noting that we can also calculate the stability coefficient for each inventory point within the inventory network. If the stability coefficient of any point is found to be lower than the average stability coefficient by a certain percentage μ, then this indicates that the supply node is in a critical state, and an immediate alarm will be issued. While the overall average stability coefficient indicates the stability of the supply network, the presence of individual inventory points in a critical state should prompt the company to implement strategic remediation measures to increase the stability of the entire supply network.

[0099] Based on the analysis and research of various elements of the multi-point steady-state warehouse network, we found that for a specific supply node, the order occurrence factor SFO in the area covered by its supply node is i and the penalty cost factors for out-of-stock situations SFP i It is certain and generally does not change suddenly over time. In other words, it is unlikely that there will be a significant change in demand in a certain area, unless some special events occur, such as sudden war or local conflict. Therefore, these two factors can be temporarily ignored.

[0100] For the mutual assistance factor SFS between supply nodes i Generally speaking, this depends on the inventory situation of the surrounding supply nodes and the distance between the supply nodes, and this is a combination of the support of all surrounding supply nodes. The inventory change of a surrounding supply node will not have a big impact on the stability coefficient of this inventory, so we can postpone considering this factor for the time being.

[0101] In fact, the most critical factor that determines the inventory stability coefficient of a certain supply point is the inventory of that supply node, which is also consistent with objective reality.

[0102] The steps for obtaining the compensation strategy in step S4 are as follows:

[0103] S41. Calculate the steady-state coefficient SF of each supply node i and the overall steady-state coefficient SF.

[0104] S42. Find the supply node i with the worst stability coefficient and the supply nodes adjacent to this point. The corresponding nodes are nodes that can provide mutual assistance with i.

[0105] S43. Find a node j with the highest steady-state coefficient among these adjacent supply nodes, and set the steady-state coefficient SF of this supply node to j The steady-state coefficient SF of the target set by the system adj Make comparisons;

[0106] If the coefficient of the supply node with the highest steady-state coefficient SF j No SF adj If it is high, it means that there is no suitable supply node around supply node i, so we go back to the second step and find the supply node with the second worst stability coefficient;

[0107] If the supply node with the highest steady-state coefficient SF j Higher than SF adj , then it means that the supply node is suitable for outputting inventory, so inventory adjustment can be performed between the two supply nodes to increase the stability of the overall supply network. The adjusted inventory quantity needs to be obtained by the following method: Calculate the inventory increase ΔI required for inventory i to reach ρSF j The maximum output inventory ΔI required for the j inventory to meet the steady-state requirements of ρSF j , compare the absolute values ​​of the two. The inventory adjustment amount is the smaller value of the two. So a one-time adjustment strategy is formed. The parameter ρ makes ρSF between SF adj Between SF and SF, by setting the target steady-state coefficient to ρSF, it is mainly to prevent the following problems: if the steady-state coefficient target point of the warehouse that accepts inventory adjustment is directly set to the mean steady-state coefficient SF, then after the adjustment is completed, the inventory of the adjacent warehouse has decreased, that is, the inventory of the output warehouse has changed, so the support that can be provided will also decrease accordingly, resulting in the warehouse that accepts the compensation adjustment still failing to reach the mean steady-state coefficient SF. Similarly, the warehouse that provides adjustment and replenishment will also get an increase in the other party's support after outputting the inventory, so if the target steady-state coefficient is set to SF adj After that, the final result will also be higher than SF adj .

[0108] This application generates a steady-state coefficient for each supply node and the entire network based on certain parameters. This allows for re-adjustment of inventory across the entire network based on preventive and compensatory adjustments. Preventive adjustments involve proactively replenishing inventory at supply nodes with low steady-state coefficients that are at risk of a stockout, based on an algorithm, before a stockout occurs. Compensatory adjustments involve comparing the steady-state coefficients of nodes around the stockout node that are eligible for adjustment using an adjustment algorithm. The node with the highest steady-state coefficient is then selected to allocate inventory to the stockout node.

[0109] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. The same-level inventory adjustment method based on the steady-state inventory theory is characterized by: The specific steps are: S1. Calculate the existing inventory allocation to obtain the steady-state coefficient SF of the multi-point inventory network; S2, determine whether the calculated steady-state coefficient SF exceeds the steady-state threshold STV, if yes, the algorithm ends, if not, enter step S3; S3. Check the number of times the compensation strategy has been implemented. If it exceeds the set number N, place an order with the supplier. If it does not exceed N, go to step S4; S4. Calculate the corresponding replenishment strategy through the distribution of existing supply network inventory and carry out inventory deployment; S5. Calculate the steady-state coefficient SF of the multi-point inventory network based on the allocated supply network inventory allocation, and then proceed to step S2; The steps for calculating the steady-state coefficient SF in step S1 are: S11. Calculate the stability coefficient SFI of the four main factors affecting the stability of the supply network, which are the inventory levels in each supply node. i , the stability coefficient SFS related to the mutual assistance between supply nodes i , the stability coefficient SFO related to the demand situation in the area guaranteed by each supply node i , Stability factor SFP related to the loss cost of out-of-stock i , quantify and calculate these four factors; S12. Through the above quantification, the stability factor SFI of the multi-point steady-state inventory network of a certain supply node is i 、SFS i 、SFO i 、SFP i Calculate separately and after the calculation is completed, the total stability factor SF of the supply node is obtained by the following formula i : SF i =θ×SFI i +τ×SFS i +κ×SFO i +λ×SFP i where θ is SFI i The weight of i The weight of i The weight of SFP; λ is the weight of SFP i The weight of S13. Stability factor SF of each supply node i After the calculation is completed, the mean is calculated, which is the steady-state coefficient SF of the multi-point steady-state inventory network of the supply network. The formula is as follows: where θ is SFI i The weight of i The weight of i The weight of i The weight of , where θ+τ+κ+2=1; The steps for obtaining the compensation strategy in step S4 are as follows: S41. Calculate the steady-state coefficient SF of each supply node i and the overall steady-state coefficient SF of the entire inventory network; S42. Find the supply node i with the worst stability coefficient and the supply nodes adjacent to this node; S43. Find a node j with the highest steady-state coefficient among these adjacent supply nodes, and set the steady-state coefficient SF of this supply node to j The steady-state coefficient SF of the target set by the system adj For comparison; if the coefficient SF of the supply node with the highest steady-state coefficient j No SF adj If it is high, it means that there is no suitable supply node around supply node i, so we return to step S42 to find the supply node with the second worst stability coefficient; If the highest supply node steady-state coefficient SF j Higher than SF adj , inventory adjustment is performed between these two supply nodes to increase the stability of the total supply network; the adjusted inventory quantity needs to be obtained by the following method: Calculate the inventory increase ΔI required for inventory i to reach ρSF j The maximum output inventory ΔI required for the j inventory to meet the steady-state requirements of ρSF j , compare the absolute values ​​of the two, the inventory allocation amount is the smaller value of the two, where ρ is a parameter between SF adj and between SF.

2. The same-level inventory adjustment method based on steady-state inventory theory according to claim 1 is characterized in that: The stability factor SFI related to the inventory level in the supply node in step S1 i The calculation formula is as follows: Among them I i is the inventory quantity of the supply node; SS i is the safety inventory of the supply node; α is the inventory stability related factor, and α is less than 1.

3. The same-level inventory adjustment method based on steady-state inventory theory according to claim 1 is characterized in that: Stability coefficient SFS related to the mutual assistance between supply nodes i The calculation formula is as follows Where: P i The success rate of supplying nodes to the other party; j The inventory of the node supplying the other party; d ij is the transportation cost between supply nodes; The transportation cost between supply nodes includes the economic cost of transportation and the time cost of transportation, d ij The calculation formula is: d ij =A(T ij ) γ +B(c ij ) δ Among them: A is the weighting factor of time cost; B is the weighting factor of economic cost; T ij is the transportation time between two supply nodes; γ is the correlation factor of transportation time; c ij The transportation cost between two supply nodes; δ is the correlation factor of transportation cost; d ij Substitute into the stability coefficient SFS related to the mutual assistance between supply nodes i The formula is obtained from 4. The same-level inventory adjustment method based on steady-state inventory theory according to claim 1 is characterized in that: Stability coefficient SFO related to the demand situation in the area guaranteed by each supply node i The calculation formula is as follows: After-school care i =1-(OP i ) s Among them OP i is the probability of demand exceeding the safety stock of the supply node; ε is the risk-related factor of the order.

5. The same-level inventory adjustment method based on steady-state inventory theory according to claim 1 is characterized in that: Stability coefficient SP related to the loss cost of out-of-stock i The calculation formula is as follows: SFP i =1-(SP i ) ∈ Among them SP i The risk level of stock-out at the supply node; ∈ is the penalty cost related factor of out-of-stock; When the optimal inventory level I is calculated i and expected penalty costs EPC i (Expectation Punishment Cost), calculate the loss cost SP of out-of-stock i The formula is: Among them, EPC i The expected penalty fee for this supply node; η is the penalty cost related factor; Substitute the supply node risk calculation formula into the penalty cost stability factor SFP i In the formula, we get SFP i formula: When the expected penalty cost EPC is 0, SFP i is 1.

Images