An inventory management method and system for emergency supplies

Abstract

This application discloses a method and system for emergency supplies inventory management, belonging to the field of inventory technology. The method includes: real-time collection of multi-dimensional data on emergency supplies, wherein the multi-dimensional data includes at least inventory data, demand data, supply data, and cost data; construction of an indicator system covering demand risk indicators, supply risk indicators, and inventory risk indicators; use Monte Carlo simulation and machine learning algorithms for risk assessment and dynamic early warning; establishment of a bi-objective optimization model with the objectives of minimizing total cost and minimizing the risk index; solution using an improved multi-objective particle swarm optimization algorithm to generate a Pareto optimal solution set; and integration of the analytic hierarchy process (AHP), the TOPSIS method, and a fuzzy multi-criteria decision-making method to select the optimal inventory strategy from the Pareto optimal solution set. This application, by constructing a data-driven intelligent decision-making scheme, achieves dynamic optimization of emergency supplies inventory strategies, effectively balances costs and risks, and improves the scientific nature and effectiveness of emergency response.

Description

Technical Field

[0001] This application relates to the field of inventory management technology, and in particular to a method and system for inventory management of emergency supplies. Background Technology

[0002] With the frequent occurrence of emergencies such as natural disasters and public health incidents, the scientific reserve and efficient allocation of emergency supplies have become crucial to ensuring social security and stability.

[0003] Existing emergency supplies inventory management suffers from several shortcomings: First, unbalanced objectives: Traditional management models often focus solely on reducing inventory costs or mitigating stockout risks, failing to achieve synergistic optimization of costs and risks, resulting in a lack of globally optimal inventory strategies. For example, simply pursuing low costs may lead to insufficient safety stock, hindering timely response during emergencies; while excessive inventory may reduce risk, it can cause capital tied up and material waste. Second, unscientific decision-making: Relying on manual experience or simple rules to formulate inventory strategies makes it difficult to quantify complex, multi-dimensional indicators. For instance, when assessing supplier reliability, the inability to fully integrate delivery history data and transportation risks leads to inappropriate supplier selection. Third, weak dynamic response capabilities: Faced with the suddenness and uncertainty of emergencies, existing systems lack real-time data-driven dynamic adjustment mechanisms, failing to optimize inventory strategies promptly based on risk changes, thus impacting emergency response efficiency.

[0004] Current technologies, such as single-objective optimization algorithms and simple weighted multi-objective methods, cannot meet the complex needs of emergency material inventory management. There is an urgent need for an innovative system and method to solve these problems. Summary of the Invention

[0005] This application provides an inventory management method and system for emergency supplies, which is used to solve at least one of the above-mentioned technical problems.

[0006] The technical solution adopted in this application is as follows: On the one hand, this application provides a method for managing the inventory of emergency supplies. The method includes: real-time collection of multi-dimensional data on emergency supplies, wherein the multi-dimensional data includes at least inventory data, demand data, supply data, and cost data; construction of an indicator system covering demand risk indicators, supply risk indicators, and inventory risk indicators; and use Monte Carlo simulation and machine learning algorithms for risk assessment and dynamic early warning; establishment of a bi-objective optimization model with the objectives of minimizing total cost and minimizing risk index; solution using an improved multi-objective particle swarm optimization algorithm to generate a Pareto optimal solution set; and integration of the analytic hierarchy process, the TOPSIS method, and the fuzzy multi-criteria decision-making method to select the optimal inventory strategy from the Pareto optimal solution set.

[0007] In one possible implementation of this application, the multidimensional data includes: the inventory data includes at least the type, quantity, storage location, shelf life, and inventory status of the materials; the demand data includes at least historical disaster data, population distribution, medical institution demand, and community demand; the supply data includes at least supplier capacity, historical supply records, real-time traffic conditions of transportation routes, and transportation vehicle status; and the cost data includes at least the purchase price, warehousing costs, transportation costs, and stockout loss costs.

[0008] In one possible implementation of this application, after obtaining the multidimensional data of the emergency supplies, the method further includes: preprocessing the multidimensional data, including: cleaning the multidimensional data to remove duplicate data, erroneous data and irrelevant data; formatting the cleaned multidimensional data to unify the data format and units of the multidimensional data; and standardizing the formatted multidimensional data.

[0009] In one possible implementation of this application, the indicator system includes at least the following: the demand risk indicators include demand forecast error rate, sudden demand fluctuation coefficient, and demand spatiotemporal distribution imbalance; the supply risk indicators include at least the supplier delivery delay probability, transportation route interruption risk value, and supplier credit rating; and the inventory risk indicators include at least the material expiration rate, inventory turnover rate, and safety stock deviation.

[0010] In one possible implementation of this application, the transportation route disruption risk value in the supply risk index is calculated by fusing supplier delivery data, real-time transportation route information, weather data, and GIS data using a Bayesian network.

[0011] In one possible implementation of this application, after obtaining the bi-objective optimization model, the method further includes: determining the risk index weights and risk index values ​​through an improved entropy weight-analytic hierarchy process; setting multiple constraints; and using a multi-objective particle swarm optimization algorithm that introduces dynamic inertia weights and adaptive learning factors to solve for the model parameters and generate a Pareto optimal solution set.

[0012] In one possible implementation of this application, selecting the optimal inventory strategy from the Pareto optimal solution set specifically includes: constructing a judgment matrix to determine the subjective weights of the decision criteria; using the TOPSIS method, based on the indicator data under the constructed indicator system, calculating the similarity between each scheme in the Pareto optimal solution set and the ideal solution, and performing preliminary ranking of the schemes in the Pareto optimal solution set; and using triangular fuzzy quantitative fuzzy indicators to calculate the final score of each scheme and select the optimal inventory strategy.

[0013] In one possible implementation of this application, after selecting the optimal inventory strategy, the method further includes: issuing the optimal inventory strategy to the execution system and monitoring the execution process in real time; collecting execution data and optimizing the parameters of the bi-objective optimization model through a reinforcement learning algorithm.

[0014] In one possible implementation of this application, the optimal inventory strategy is issued to the execution system, specifically including: transmitting the optimal inventory strategy to the logistics management system and the warehouse management system to automatically execute purchase order generation, material allocation and inventory counting operations; then, establishing a real-time monitoring mechanism to collect real-time data of the optimal inventory strategy during execution; and dynamically adjusting the parameters of the bi-objective optimization model using a reinforcement learning algorithm based on historical execution data and the real-time data.

[0015] On the other hand, this application also provides an emergency supplies inventory management system, the system comprising: a data acquisition module for real-time acquisition of multidimensional data on emergency supplies, the multidimensional data including at least inventory data, demand data, supply data, and cost data; a risk assessment module for constructing an indicator system covering demand risk indicators, supply risk indicators, and inventory risk indicators, and using Monte Carlo simulation and machine learning algorithms for risk assessment and dynamic early warning; a dual-objective optimization module for establishing a dual-objective optimization model with the objectives of minimizing total cost and minimizing risk index, solving it using an improved multi-objective particle swarm optimization algorithm, and generating a Pareto optimal solution set; and a decision-making module for integrating the analytic hierarchy process, the TOPSIS method, and the fuzzy multi-criteria decision-making method to select the optimal inventory strategy from the Pareto optimal solution set.

[0016] The emergency supplies inventory management method and system provided in this application have the following beneficial effects: This application overcomes the limitations of traditional single-objective management by constructing a cost-risk dual-objective collaborative optimization model, achieving innovative collaborative optimization and dynamic balance between cost and risk in inventory strategies, thereby improving the overall efficiency of emergency material management. It enhances scientific decision-making capabilities by integrating multiple multi-criteria decision-making methods, combining subjective experience with objective data, effectively handling complex indicators and fuzzy information, improving the scientific rigor and reliability of decisions, and providing precise strategic support for emergency material inventory management. It strengthens dynamic adaptive capabilities by dynamically adjusting and optimizing the dual-objective optimization model based on real-time data and reinforcement learning algorithms, enabling rapid response to demand changes and supply chain risks brought about by emergencies, significantly improving emergency response efficiency and material support capabilities. Furthermore, it expands broad applicability; the system and methods can flexibly adjust parameters and indicator systems according to different types of emergency materials, such as medical supplies and disaster relief supplies, and different application scenarios, such as urban and rural emergency management, exhibiting strong versatility and scalability. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 A flowchart of an emergency supplies inventory management method provided in this application; Figure 2 This application provides an architecture diagram for an emergency supplies inventory management system. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0019] This application provides an intelligent solution for emergency supplies inventory management, proposing a cost-risk dual-objective collaborative optimization model and a multi-criteria decision-making method. The solution constructs a comprehensive sensing system through the Internet of Things (IoT), RFID, sensor networks, and GIS to collect real-time data on emergency supplies inventory, demand, supply chain, and costs. After data cleaning and standardization preprocessing, a high-dimensional data foundation is built. Based on this, an innovative three-dimensional assessment index system encompassing demand risk, supply risk, and inventory risk is constructed. Monte Carlo simulation and machine learning algorithms, such as LSTM neural networks, are used to achieve dynamic risk assessment and intelligent early warning. The core of this solution lies in establishing a dual-objective collaborative optimization model aimed at minimizing total cost and risk index. An improved multi-objective particle swarm optimization algorithm is introduced to generate a Pareto optimal solution set. Simultaneously, the analytic hierarchy process (AHP), the Topology-Topology Solution Ranking (TOPSIS) approximation method, and a fuzzy multi-criteria decision-making method are integrated. Through subjective weight construction, solution closeness calculation, and fuzzy index quantification, the optimal inventory strategy is selected from the Pareto solution set. At the execution level, the system automatically drives the logistics and warehousing systems to execute decisions. Based on reinforcement learning algorithms, the execution data is deeply mined to dynamically optimize system model parameters. Ultimately, a closed-loop management system is formed, encompassing "data collection - risk assessment - decision optimization - execution feedback - model improvement," achieving intelligent and scientific management of emergency material inventory.

[0020] The method in this application will be described in detail below with reference to the accompanying drawings.

[0021] Figure 1 A flowchart of an emergency supplies inventory management method provided in this application is shown below. Figure 1 As shown, the emergency supplies inventory management method in this application includes at least the following steps: Step 101: Collect multi-dimensional data on emergency supplies in real time.

[0022] The emergency supplies inventory management method in this application first involves intelligent data collection and preprocessing. Relying on IoT devices, RFID tags, sensor networks, and GIS, a comprehensive data collection network is constructed to acquire multi-dimensional data on emergency supplies in real time. This includes inventory data (such as type, quantity, and storage location), demand data (such as historical disaster data and population distribution), supply data (such as supplier capacity and transportation route status), and cost data (such as purchase price and warehousing costs). The collected data is then cleaned, denoised, formatted, and standardized before being stored in a distributed database, laying a solid data foundation for subsequent analysis.

[0023] Specifically, among the multidimensional data collected above, inventory data mainly includes the types, quantities, storage locations, shelf lives, and inventory status of materials; demand data mainly includes historical disaster data, population distribution, demand from medical institutions, and community demand forecasts; supply chain data mainly includes supplier capacity, historical supply records, real-time traffic conditions of transportation routes, and the status of transportation vehicles; and cost data mainly includes purchase unit price, warehousing costs, transportation costs, and stockout loss costs.

[0024] Furthermore, data preprocessing of the collected multidimensional data refers to cleaning the multidimensional data of emergency supplies to remove duplicate, erroneous, and irrelevant data; formatting the cleaned multidimensional data of emergency supplies to unify the data format and units; and standardizing the formatted multidimensional data of emergency supplies to provide accurate data support for subsequent analysis.

[0025] Step 102: Construct an indicator system covering demand risk indicators, supply risk indicators, and inventory risk indicators, and use Monte Carlo simulation and machine learning algorithms for risk assessment and dynamic early warning.

[0026] In one possible implementation of this application, a scientific and comprehensive risk assessment indicator system is constructed, encompassing three dimensions: demand, supply, and inventory. Demand risk indicators include demand forecasting error rate and sudden demand fluctuation coefficient; supply risk indicators utilize a Bayesian network to integrate supplier delivery data, real-time transportation information, and GIS data to calculate supplier delivery delay probability and transportation route disruption risk values; inventory risk indicators cover material expiration rate and inventory turnover rate. A Monte Carlo simulation algorithm combined with an LSTM neural network is used to dynamically predict future risks, and red, yellow, and blue three-level warning thresholds are set, providing timely warnings through multiple channels such as SMS and email. In one example, the transportation route disruption risk value is evaluated using a Bayesian network based on weather data, traffic control information, and GIS data.

[0027] Step 103: Establish a bi-objective optimization model with the objectives of minimizing total cost and minimizing risk index, and solve it using an improved multi-objective particle swarm optimization algorithm to generate the Pareto optimal solution set.

[0028] In one possible implementation of this application, a dual-objective collaborative optimization model is established, with the core objectives of minimizing total cost and minimizing risk index. The model parameters include cost elements such as procurement cost and warehousing cost, as well as risk index weights and values ​​determined through an improved entropy-weighted analytic hierarchy process (AHP). Simultaneously, multiple constraints are set, such as inventory capacity limits and material shelf-life constraints. An improved multi-objective particle swarm optimization algorithm incorporating dynamic inertia weights and adaptive learning factors is used to solve the model, generating a Pareto optimal solution set.

[0029] Specifically, in the above dual-objective collaborative optimization model, the model parameters are procurement cost, warehousing cost, transportation cost and stockout cost, risk index, risk indicator weight (which can be determined by the improved entropy weight-analytic hierarchy process), risk indicator value, etc. Meanwhile, the constraints include inventory capacity limit, material shelf life constraint, emergency response time constraint, such as requiring that the probability of key materials being delivered to the designated area within 48 hours is not less than 90%, and capital budget constraint.

[0030] Step 104: Integrating the Analytic Hierarchy Process (AHP), TOPSIS, and fuzzy multi-criteria decision-making methods, the optimal inventory strategy is selected from the Pareto optimal solution set.

[0031] Multi-criteria decision-making methods include the Analytic Hierarchy Process (AHP), which invites experts in emergency management and supply chain to construct a judgment matrix and determine the subjective weights of decision criteria such as cost, risk, and timeliness; the TOPSIS method, which calculates the closeness of each option to the ideal solution based on standardized cost, risk, and other relevant indicator data, and performs preliminary ranking of the options in the Pareto optimal solution set; and fuzzy multi-criteria decision-making such as fuzzy MCDM, which uses triangular fuzzy numbers to quantify fuzzy indicators such as "probability of disaster occurrence" and "urgency of materials," and calculates the final score of the options through fuzzy comprehensive evaluation to select the inventory strategy with the best overall performance. In one example, the optimal inventory strategy includes safety stock level, procurement cycle, supplier selection combination, and material allocation plan.

[0032] In summary, this step uses the analytic hierarchy process (AHP) to determine the subjective weights of each decision criterion; the TOPSIS method is used to initially rank the schemes in the Pareto optimal solution set; and a fuzzy multi-criteria decision-making method is used to process fuzzy indicators, calculate the final score of the schemes, and select the optimal inventory strategy.

[0033] In one possible implementation of this application, after obtaining the optimal inventory strategy, the inventory strategy is issued to the execution system, and the execution process is monitored in real time. Execution data is collected, and the model parameters are optimized using reinforcement learning algorithms to achieve continuous improvement of the decision-making method. Specifically, the final inventory strategy is transmitted to the logistics management system and the warehouse management system, automatically executing operations such as purchase order generation, material allocation, and inventory counting. A real-time monitoring mechanism is established to collect actual data during the execution of the inventory strategy (such as actual purchase cost, actual response time, and actual number of stockouts). Based on historical execution data and current real-time data, reinforcement learning algorithms are used to dynamically adjust and optimize the parameters of the risk assessment model and the bi-objective optimization model, forming a closed-loop management system of "data collection - risk assessment - decision optimization - execution feedback - model improvement".

[0034] Based on the same inventive concept, this application also provides an emergency supplies inventory management system, the architecture of which is as follows: Figure 2 As shown.

[0035] Figure 2 This application provides an architecture diagram for an emergency supplies inventory management system. (For example...) Figure 2As shown, the emergency supplies inventory management system 200 in this application specifically includes: a data acquisition module 201, which collects multi-dimensional data of emergency supplies in real time, including at least inventory data, demand data, supply data, and cost data; a risk assessment module 202, which constructs an indicator system covering demand risk indicators, supply risk indicators, and inventory risk indicators, and uses Monte Carlo simulation and machine learning algorithms for risk assessment and dynamic early warning; a dual-objective optimization module 203, which establishes a dual-objective optimization model with the objectives of minimizing total cost and minimizing risk index, and uses an improved multi-objective particle swarm optimization algorithm to solve it, generating a Pareto optimal solution set; and a decision module 204, which integrates the analytic hierarchy process, the TOPSIS method, and the fuzzy multi-criteria decision method to select the optimal inventory strategy from the Pareto optimal solution set.

[0036] In one possible implementation of this application, the emergency supplies inventory management system may further include an execution and feedback module 205, which issues the optimal inventory strategy to the execution system, monitors the execution process in real time, collects execution data, and optimizes the parameters of the bi-objective optimization model through a reinforcement learning algorithm.

[0037] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0038] The system and method provided in this application are one-to-one correspondences. Therefore, the system also has similar beneficial technical effects as its corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the system will not be repeated here.

[0039] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code.

[0040] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0041] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for managing the inventory of emergency supplies, characterized in that, The method includes: Real-time collection of multi-dimensional data on emergency supplies, including at least inventory data, demand data, supply data, and cost data; Construct an indicator system covering demand risk indicators, supply risk indicators, and inventory risk indicators, and use Monte Carlo simulation and machine learning algorithms for risk assessment and dynamic early warning; A bi-objective optimization model is established with the objectives of minimizing total cost and minimizing risk index. An improved multi-objective particle swarm optimization algorithm is used to solve the model and generate a Pareto optimal solution set. By integrating the Analytic Hierarchy Process (AHP), the TOPSIS method, and the fuzzy multi-criteria decision-making method, the optimal inventory strategy is selected from the Pareto optimal solution set.

2. The method for managing the inventory of emergency supplies according to claim 1, characterized in that, In the multidimensional data: The inventory data includes at least the type, quantity, storage location, shelf life, and inventory status of the materials; the demand data includes at least historical disaster data, population distribution, medical institution demand, and community demand; the supply data includes at least supplier capacity, historical supply records, real-time traffic conditions of transportation routes, and transportation vehicle status; and the cost data includes at least the purchase price, warehousing costs, transportation costs, and stockout loss costs.

3. The method for managing the inventory of emergency supplies according to claim 1, characterized in that, After obtaining the multidimensional data of the emergency supplies, the method further includes: Preprocessing the multidimensional data includes: Data cleaning is performed on the multidimensional data to remove duplicate, erroneous, and irrelevant data. The cleaned multidimensional data is formatted to unify the data format and units of the multidimensional data; The formatted multidimensional data is then standardized.

4. The method for managing the inventory of emergency supplies according to claim 1, characterized in that, In the aforementioned indicator system: The demand risk indicators include at least the demand forecast error rate, the coefficient of sudden demand fluctuation, and the degree of unevenness in the spatiotemporal distribution of demand; the supply risk indicators include at least the supplier delivery delay probability, the risk value of transportation route disruption, and the supplier credit rating. The inventory risk indicators include at least the material expiration rate, inventory turnover rate, and safety stock deviation.

5. The method for managing the inventory of emergency supplies according to claim 4, characterized in that, The transportation route disruption risk value in the supply risk indicators is calculated by integrating supplier delivery data, real-time transportation route information, weather data, and GIS data using a Bayesian network.

6. The method for managing the inventory of emergency supplies according to claim 1, characterized in that, After obtaining the bi-objective optimization model, the method further includes: Risk indicator weights and values ​​were determined using an improved entropy-weighted analytic hierarchy process. Multiple constraints are set, and a multi-objective particle swarm optimization algorithm with dynamic inertia weights and adaptive learning factors is used to solve the model parameters and generate the Pareto optimal solution set.

7. The method for managing the inventory of emergency supplies according to claim 1, characterized in that, Selecting the optimal inventory strategy from the Pareto optimal solution set specifically includes: Construct a judgment matrix to determine the subjective weights of the decision-making criteria; Using the TOPSIS method, based on the index data under the constructed index system, the similarity between each scheme in the Pareto optimal solution set and the ideal solution is calculated, and the schemes in the Pareto optimal solution set are initially ranked. Using triangular fuzzy quantitative fuzzy indicators, the final scores of each scheme are calculated, and the optimal inventory strategy is selected.

8. The method for managing the inventory of emergency supplies according to claim 1, characterized in that, After selecting the optimal inventory strategy, the method further includes: The optimal inventory strategy is issued to the execution system, and the execution process is monitored in real time. Collect execution data and optimize the parameters of the dual-objective optimization model using reinforcement learning algorithms.

9. The method for managing the inventory of emergency supplies according to claim 8, characterized in that, The optimal inventory strategy is issued to the execution system, specifically including: The optimal inventory strategy is transmitted to the logistics management system and warehouse management system to automatically execute purchase order generation, material allocation and inventory counting operations. Next, a real-time monitoring mechanism was established to collect real-time data on the execution of the optimal inventory strategy; Based on historical execution data and the real-time data, reinforcement learning algorithms are used to dynamically adjust the parameters of the bi-objective optimization model.

10. An emergency supplies inventory management system, characterized in that, The system includes: The data acquisition module collects multidimensional data on emergency supplies in real time, including at least inventory data, demand data, supply data, and cost data. The risk assessment module constructs an indicator system covering demand risk indicators, supply risk indicators, and inventory risk indicators, and uses Monte Carlo simulation and machine learning algorithms for risk assessment and dynamic early warning. The dual-objective optimization module establishes a dual-objective optimization model with the objectives of minimizing total cost and minimizing risk index, and uses an improved multi-objective particle swarm optimization algorithm to solve it, generating a Pareto optimal solution set; The decision-making module integrates the analytic hierarchy process, the TOPSIS method, and the fuzzy multi-criteria decision-making method to select the optimal inventory strategy from the Pareto optimal solution set.

Images