Medicine warehouse storage management method

Abstract

The invention discloses a medicine warehouse storage management method, which comprises the steps of collecting multi-dimensional parameters such as temperature, humidity, illumination and air components of a storage environment through environment monitoring equipment, and constructing a time-space synchronous fusion database; meanwhile, digital feature tags are set based on physical and chemical attributes, importance levels and component stability of the drugs, and a drug classification feature list is formed; the method comprises the following steps: performing correlation analysis on environment data and drug characteristics, establishing a drug-environment relationship list, and dynamically setting an environment threshold value and a storage time threshold value of each drug according to the drug-environment relationship list; environment parameters are monitored in real time according to a threshold value, a storage strategy is automatically adjusted when a monitoring value exceeds a limit, and medicine stock optimization based on epidemic case prediction is supported; according to the method, the medicine quality problem caused by static setting of a traditional storage environment is solved, refined and self-adaptive management is achieved, and the medicine storage safety and the medicine effect stability are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of pharmaceutical warehousing technology, and in particular to a method for managing pharmaceutical warehouse storage. Background Technology

[0002] In the current field of pharmaceutical warehousing management, especially for conventional warehouses (non-cold storage, non-temperature-controlled storage), the commonly used environmental monitoring methods mainly rely on static threshold alarms for basic environmental parameters such as temperature and humidity. These methods typically deploy discrete sensors for single-point monitoring, triggering alarms or activating simple control devices (such as air conditioners and dehumidifiers) when parameters exceed preset fixed limits. However, this method has significant limitations: firstly, it lacks the ability to comprehensively and synchronously collect and analyze multiple environmental parameters (such as specific spectral illumination, microbial load, and air composition), making it difficult to reflect the complex impact of the environment on drug quality; secondly, the management strategy is static and universal, unable to provide differentiated and refined risk control based on the unique physicochemical properties of different drugs within the warehouse. This leads to a high risk of chemical degradation or microbial contamination of drugs in conventional warehouses with variable climates, especially for high-value or highly sensitive drugs, due to unreasonable environmental fluctuations.

[0003] Furthermore, existing technologies have introduced storage zoning concepts based on drug classification, such as placing photosensitive drugs in light-protected areas. However, such zoning is often fixed and rudimentary, relying on manual experience for setting and unable to adaptively adjust based on real-time dynamic environmental data and the dynamic state of the drugs. When localized, instantaneous abnormal fluctuations occur in the warehouse environment (such as sudden changes in temperature and humidity in a certain area due to equipment failure, or abnormally increased light due to open doors and windows), or when different risk types (such as the risk of microbial growth and the risk of photodegradation) overlap and conflict in the same area, static zoning and single control strategies often fail or hinder each other. Summary of the Invention

[0004] This application provides a dynamic warehouse management method for pharmaceutical storage, which can perceive complex environmental risks in real time, accurately match drug characteristics, and intelligently coordinate multi-objective control strategies to solve the contradictions and risks caused by the above-mentioned static and fragmented management.

[0005] This application provides a method for pharmaceutical warehouse storage management, including: S1. Deploy integrated sensor network nodes within the warehouse. Each node synchronously collects multi-source data from its location and adds timestamps and three-dimensional spatial location labels to all collected data to build a spatiotemporally synchronized fusion database. S2 establishes sensitivity labels for each drug stored in the warehouse in two dimensions: photosensitivity and microbial sensitivity, forming a digital profile of each drug with dual characteristics of sensitive bacteria and photosensitivity. S3, based on the fusion database and digital images, calculates the microbial contamination risk index and photodegradation risk index of medicines in each storage location under the current environment, and identifies risk conflict areas where both high microbial contamination risk and high photodegradation risk exist simultaneously. S4: For the identified risk conflict areas, calculate their fusion risk index, and dynamically divide different warehouse functional areas based on the real-time fusion risk index of each area in the entire warehouse. S5 generates the optimal collaborative control strategy based on different warehouse functional areas.

[0006] Preferably, the process of forming a digital profile of each drug with dual biosensitivity and photosensitivity specifically includes: analyzing the drug's photosensitivity based on its chemical composition, physical form, and photolysis characteristics; determining its sensitive wavelength range; establishing its photodegradation kinetic model; and determining its maximum allowable cumulative light exposure; assessing the drug's microbial contamination sensitivity based on its formulation, packaging materials, and preservative information; determining its microbial sensitivity level and susceptible microbial types; and structurally integrating the photosensitivity parameters with the quantification results of microbial sensitivity to form a digital profile of the drug. Preferably, the risk conflict area includes: extracting real-time environmental monitoring data of each storage location from a spatiotemporally synchronized fusion database, and retrieving the digital feature files of the drugs stored in the corresponding storage location from a drug feature label library; calculating the photodegradation risk index of drugs in each storage location based on the light data in the extracted environmental monitoring data and the light sensitivity parameters in the digital feature files; calculating the microbial contamination risk index of drugs in each storage location based on the microbial load, temperature, and humidity data in the extracted environmental monitoring data, and the microbial sensitivity parameters in the digital feature files; setting a high threshold for photodegradation risk and a high threshold for microbial contamination risk, and identifying storage locations that simultaneously exceed both high thresholds as risk conflict areas.

[0007] Preferably, step S2, which involves forming a digital profile of each drug with dual biosensitive and photosensitive features, further includes: S21. Input the initial physicochemical properties and accelerated stability test data of the drug into the digital profile to form the initial characteristic library of the drug. S22, extract the real-time environmental monitoring sequence of each storage location from the fusion database, associate it with the digital feature file of the corresponding drug in the storage location, and store it in time series to form a traceable dynamic database of drug-environment interaction; S23. Based on the environmental historical data sequence and the initial characteristics of the drug in the dynamic database, the drug active ingredient degradation rate prediction model and the drug preservative efficacy decay model are constructed by applying the principles of pharmacokinetics, and the predicted degradation curve of the drug active ingredient and the real-time decay status of its preservative efficacy are output. S24. Based on the predicted degradation curve and the real-time decay status of the preservative efficacy, calculate the decay index reflecting the current stability status of the drug; dynamically adjust the maximum allowable cumulative light exposure of the drug based on the predicted degradation curve; and dynamically adjust its comprehensive microbial susceptibility index by combining the real-time microbial load and the predicted decay status of the preservative efficacy. S25, recalculate the photodegradation and microbial contamination risk index of drugs in each storage location, reclassify the risk level based on the new risk index, and automatically generate and execute a storage location dynamic migration instruction when the drug risk level is upgraded, transferring the drug to a functional area that better matches its current stability state.

[0008] Preferably, the step of outputting the predicted degradation curve of the active pharmaceutical ingredient and its real-time decay status of preservative efficacy specifically includes: establishing a degradation rate prediction model for the active pharmaceutical ingredient and a decay model for the preservative efficacy of the drug based on the environmental historical data sequence and the initial characteristic parameters of the drug in a dynamic database through parameter fitting; verifying the accuracy of the prediction model and the decay model using reserved monitoring data or sampling test data, and dynamically updating the model parameters according to the verification results; combining the parameter-updated model with real-time environmental monitoring data to generate and output the predicted degradation curve of the active pharmaceutical ingredient and its real-time decay status of preservative efficacy.

[0009] Preferably, step S24 specifically includes: determining the current stability state of the drug based on the predicted degradation curve and the real-time decay status; dynamically reducing the maximum permissible cumulative light exposure threshold of the drug based on the current stability state; and dynamically increasing the comprehensive microbial susceptibility index of the drug by combining real-time environmental monitoring data and the real-time decay status.

[0010] Preferably, in S22, storing the dynamic database of traceable drug-environment interactions in a time-series manner includes: S221. Before the first entry of a drug into the warehouse, expand its dynamic database for each drug and add chemical interaction characteristic records. Among them, the chemical interaction characteristic records include the volatile component spectrum (VOCs fingerprint) of the drug, chemical compatibility data with other common drugs, and physical stress tolerance threshold. S222, when planning to adjust the storage location of medicines or to perform inbound / outbound operations, based on the digital twin model of the warehouse and combined with the physical stress tolerance parameters in the chemical interaction characteristics, the planned movement path is simulated to calculate the physical environmental stress that the medicines will accumulate during the entire movement process. S223. Based on the calculated cumulative physical environmental stress and combined with the physical stability model of the drug, a stability decay prediction curve of the drug during this movement is constructed. S224, based on a computational fluid dynamics model, simulates in real time the airflow state in the warehouse and the diffusion path and concentration distribution of VOCs released from existing stocked drugs; when a drug is planned to be moved to the target storage location, it calls the VOCs fingerprint of the drug and its chemical compatibility data with existing drugs around the target storage location in real time, and calculates the potential cross-reaction risk index after its volatiles are mixed with VOCs in the surrounding environment.

[0011] Preferably, the calculation of the cumulative physical environmental stress that the drug will experience throughout the entire movement process specifically includes: Based on the digital twin model of the warehouse, the planned drug movement path is extracted and discretized into multiple continuous path segments; the physical environmental stress sources of each path segment are parameterized and defined; based on the physical stress tolerance parameters of the drug, the physical environmental stress level borne by the drug in each path segment is calculated; the various physical environmental stresses that the drug will bear along the entire movement path are cumulatively analyzed, and a physical stress over-limit risk warning is generated based on the comparison results between the cumulative physical environmental stress and the physical stress tolerance threshold of the drug.

[0012] Preferably, the construction of the drug's stability degradation prediction curve during this migration process includes: Based on the calculated cumulative physical environmental stress and the physical stability model parameters of the drug, a stability decay prediction curve for the drug during this movement is constructed. According to the stability decay prediction curve, the predicted stability index of the drug at each moment during the movement is calculated. The predicted stability index is compared with the preset stability safety threshold, and a drug movement stability risk warning is triggered when the predicted stability index is lower than the safety threshold.

[0013] Preferably, the potential cross-reaction risk index specifically includes: integrating real-time environmental parameters of the warehouse with the VOCs release characteristics of all stored drugs to construct a spatial source term field reflecting the VOCs release situation of the entire warehouse; based on the spatial source term field and environmental boundary conditions, obtaining the real-time three-dimensional concentration distribution of each VOCs component in the warehouse through computational fluid dynamics model; when a drug is planned to be moved to the target storage location, it is set as a temporary release source in the model, simulating its volatile diffusion and obtaining its predicted concentration distribution in the target area; extracting the background concentration distribution of each VOCs component around the target area from the background concentration field; obtaining the chemical compatibility data between the target drug and existing drugs in the surrounding area, and calculating the comprehensive cross-reaction risk index caused by the introduction of the target drug based on the predicted concentration distribution, background concentration distribution, and chemical compatibility data; comparing the comprehensive cross-reaction risk index with a preset dynamic safety threshold, and generating a cross-contamination risk warning if the threshold is exceeded.

[0014] One or more technical solutions provided in this application have at least the following technical effects or advantages: By intelligently sensing and dynamically matching the storage environment with the characteristics of pharmaceuticals, the problem of declining pharmaceutical quality caused by unreasonable environmental settings is effectively solved. Its technological effects include significantly improving the quality of pharmaceutical storage and ensuring drug efficacy and storage safety. Advantages include achieving adaptive environmental control, reducing manual intervention costs, improving storage efficiency, and being particularly suitable for the refined, low-cost management of conventional warehouses in climate-changing environments.

[0015] By introducing a dynamic drug stability degradation prediction and adaptive control mechanism, the level of intelligence in warehouse management has been significantly improved. Its technical effectiveness lies in its ability to predict the quality degradation trend of drugs in the storage environment in real time and dynamically adjust environmental thresholds and storage strategies, thereby effectively avoiding drug deterioration caused by unsuitable environments. Advantages include significantly improving drug storage quality and safety, reducing the cost of manual intervention, enhancing adaptability to seasonal changes and drug condition fluctuations, improving warehousing efficiency through optimized resource allocation, and ensuring the stability of drugs throughout their entire lifecycle.

[0016] By introducing chemical interaction analysis and physical stress movement simulation of pharmaceuticals, this technology overcomes the limitations of traditional warehouse management that only focuses on static environmental parameters. Its technological advantage lies in its ability to proactively predict and mitigate the risks of chemical contamination caused by cross-contamination of volatile substances between pharmaceuticals, as well as the risks of physical damage during handling and transportation. Its advantages are reflected in achieving a leap from environmental monitoring to the coordinated prevention and control of components and stress. Through digital twins and fluid dynamics simulation, potential risks are accurately quantified in three-dimensional space, allowing for early warnings and optimized decisions before warehouse location adjustments are implemented. This significantly improves the quality and safety assurance level of pharmaceuticals throughout the entire warehousing and circulation process, achieving truly proactive, preventative, and refined management. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a pharmaceutical warehouse storage management method according to an embodiment of the present invention. Detailed Implementation

[0018] To facilitate understanding of the present invention, a more complete description of this application will be given below with reference to the accompanying drawings, which illustrate preferred embodiments of the invention. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to enable a more thorough and complete understanding of the disclosure of the present invention.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to limit the invention; the term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0020] Example 1: Figure 1 This is a flowchart illustrating a pharmaceutical warehouse storage management method according to an embodiment of the present invention.

[0021] like Figure 1 As shown, a method for managing drug warehouse storage includes the following steps: S1. Deploy integrated sensor network nodes within the warehouse. Each node synchronously collects multi-source data from its location and adds timestamps and three-dimensional spatial location labels to all collected data to build a spatiotemporally synchronized fusion database.

[0022] The multi-source data includes microbial data (using equipment such as ATP bioluminescence analyzers to monitor the microbial load on the air and drug surface in real time (expressed as total colony count or relative light unit value)), optical data (using spectral sensors to simultaneously collect the light intensity and spectral distribution of ultraviolet (UVA, UVB), visible and infrared bands at the same location), and basic physical environment data (simultaneously recording the temperature and humidity at that point).

[0023] S2 establishes sensitivity labels for each drug stored in the warehouse in two dimensions: photosensitivity and microbial sensitivity, forming a digital profile of each drug with dual characteristics of sensitive bacteria and photosensitivity.

[0024] Among them, photosensitivity labels are based on the photolysis experimental data or literature of the drug, which clearly define its sensitivity coefficient, degradation kinetic parameters and the threshold of cumulative light exposure that it can tolerate to different wavelengths of light (especially ultraviolet light).

[0025] Microbial susceptibility labeling assesses a drug's susceptibility to microbial contamination based on the nutritional value of its ingredients, the antimicrobial properties of its packaging materials, and the drug's own preservative capabilities. The susceptibility is categorized into high, medium, and low levels, and the specific types of microorganisms the drug is susceptible to are noted.

[0026] Specifically, detailed information on each drug is obtained, including its chemical composition, physical form, packaging materials, and relevant photolysis experimental research reports or publicly available literature. Simultaneously, information on the drug's formulation, excipients, and preservatives is collected for microbial susceptibility assessment.

[0027] Based on the collected information, the photolysis characteristics of the drug are analyzed. By resolving its ultraviolet-visible absorption spectrum, the wavelength range of its main sensitive ultraviolet region is determined, and its photosensitivity coefficient is determined within this range. This coefficient can be obtained from the molar extinction coefficient at a specific wavelength (especially at the maximum absorption wavelength). Characterized by this, its relationship with absorbance A is described by the Lambert-Beer law: Where c is the concentration and l is the optical path length. A photodegradation kinetic model of the drug is then established. First-order reaction kinetics can typically be used to describe the relationship between the concentration C of the active ingredient and time t and light intensity I, with the differential form being: C is the concentration of the active ingredient in the drug at time t, I is the light intensity, k is the photodegradation rate constant, and α and β are empirical constants (for many drugs, β≈1, i.e., a pseudo-first-order reaction). Its integral form (when β=1) is: ,in, This refers to the initial concentration. Accelerated photostability testing can determine the maximum cumulative light exposure a drug can withstand before a significant decrease in quality occurs (e.g., a 5% decrease in the main component content). Its calculation formula is ,in The average light intensity, This is the maximum safe exposure time.

[0028] Assess the drug's potential to support microbial growth and its ability to resist contamination. Calculate its nutritional score. It is based on the weight percentage of various nutrients that support microbial growth (such as carbohydrates, proteins, amino acids, etc.) in the drug components. and its preset growth-promoting weight coefficient The weighted summation is calculated using the following formula: The antimicrobial efficacy index of its packaging materials was then evaluated. This index is typically assigned an empirical value between 0 and 1 based on the antibacterial properties of the packaging material (such as whether it contains antibacterial agents, surface characteristics, etc.). Simultaneously, the preservative efficacy factor of the drug itself is determined. This factor is scored on a scale of 0 to 1 based on the type, concentration, and known potency of preservatives in the prescription. Finally, the comprehensive microbial susceptibility index is calculated by weighting the results of the three assessments. Its calculation formula is ,in , , The weighting coefficients for each factor are given, and their sum is 1. Based on the calculated M_index value range, the microbial susceptibility of the drug is divided into high, medium, and low levels. This application sets... >0.7 indicates high sensitivity, 0.3< ≤0.7 is considered moderately sensitive. ≤0.3 indicates low sensitivity, and its composition is associated with specific microbial types (such as fungi, bacteria, etc.) that it is susceptible to.

[0029] The calculated photosensitivity parameters (including sensitive wavelength range, photosensitivity coefficient, photodegradation kinetic model parameters, and maximum cumulative light exposure) are structurally integrated and correlated with the microbial susceptibility quantification results (including sensitivity level, comprehensive microbial susceptibility index, and a list of susceptible microbial types) to form a unique, standardized digital profile for each drug record. This complete digital profile is stored as core metadata in a dedicated drug feature tag library. When the drug's formulation or packaging changes, or new research data is released, a review process is triggered, and the corresponding analysis and calculations are re-executed to update the corresponding records in the tag library in a timely manner.

[0030] S3 calculates the microbial contamination risk index and photodegradation risk index of medicines in each storage location under the current environment based on the fused database and digital images, and identifies risk conflict areas where both high microbial contamination risk and high photodegradation risk exist simultaneously.

[0031] Specifically, from the spatiotemporal synchronous fusion database established in step S1, environmental monitoring values ​​at various specific locations within the warehouse at a specified evaluation time are extracted, including the microbial load, light intensity in each band, temperature, and humidity data at that point. From the drug feature label library constructed in step S2, a complete digital feature profile of the actual drug stored in each corresponding storage location is retrieved. The core of this profile includes the drug's photosensitivity parameters and microbial sensitivity parameters.

[0032] After obtaining the above data, the photodegradation risk index for each drug storage location was calculated. For each storage location, based on the monitored light intensity at different wavelengths, the total actual light exposure dose received by the drug within a specific assessment period was calculated. The calculated real-time exposure dose was compared with the maximum permissible tolerable light dose specified by the drug's own characteristics. By dividing the real-time dose by the maximum permissible dose, a quantified photodegradation risk index was obtained. This index directly reflects the risk level of photochemical degradation of the drug's active ingredient under the current light conditions; the higher the value, the greater the risk.

[0033] The microbial contamination risk index of drugs at each storage location is calculated in parallel. For each storage location, the monitored microbial load, ambient temperature, and humidity data are comprehensively analyzed. An assessment model is established to describe how temperature and humidity jointly affect the growth rate of microorganisms, thus converting the monitored temperature and humidity values ​​into a potential coefficient favorable to microbial growth. This coefficient is multiplied by the measured microbial load to obtain an assessment of the microbial growth potential of the current environment. This environmental growth potential is then multiplied by the inherent comprehensive microbial sensitivity index in the digital profile of the drug at that storage location to obtain the final microbial contamination risk index of the drug. This index quantifies the level of risk of microbial contamination of the drug under the current environmental conditions.

[0034] After calculating the two risk indices (photodegradation risk index and microbial contamination risk index) for all storage locations, the next step is to identify risk conflict areas. Two risk thresholds are pre-set (and limited according to specific application scenarios). Each storage location is examined individually to determine whether it is at high risk of photodegradation or high risk of microbial contamination. Locations with both photodegradation and microbial contamination risk indices exceeding their respective high-risk thresholds are selected. All storage locations meeting this dual high-risk condition are defined as risk conflict zones.

[0035] S4 calculates the fusion risk index for the identified risk conflict areas and dynamically divides different warehouse functional areas based on the real-time fusion risk index of each area in the entire warehouse.

[0036] Specifically, a comprehensive risk assessment is conducted on identified risk conflict areas. For each storage location marked as having a risk conflict, its photodegradation risk index and microbial contamination risk index are calculated comprehensively. During the calculation process, a weight is assigned to each risk, and the magnitude of the weight depends on the different emphases on the severity of the two risks in the management of the specific application scenario.

[0037] For example, if the current management strategy focuses more on the chemical stability of the drug, the risk of photodegradation will have a higher weight; if it focuses more on aseptic assurance levels, the risk of microbial contamination will have a higher weight. The weights of the two risks are added together to a fixed 100%. The two indices are then weighted and summed to obtain a single, integrated risk index that represents the overall threat level at that location.

[0038] The comprehensive assessment method is applied to the entire warehouse. The same calculation is performed on all other ordinary storage locations while calculating the fusion risk index for the conflict area. After this step, the fusion risk index value for each storage location at the same time is obtained, thus logically generating a global map reflecting the overall risk distribution across the entire warehouse space.

[0039] Dynamic spatial division is performed based on a real-time global risk map. Key thresholds representing different risk levels are pre-set, and the fusion risk index value calculated for each storage location is compared with the thresholds to classify them into different functional management areas.

[0040] For example, all storage locations whose fusion risk index reaches or exceeds the high-risk threshold are dynamically designated as core control areas. These areas have the highest risk and require the strictest and highest priority intervention. Storage locations whose fusion risk index is below the high-risk threshold but reaches or exceeds the low-risk threshold are designated as buffer monitoring areas. These areas require enhanced monitoring and preparation for risk escalation. Storage locations whose fusion risk index is below the low-risk threshold are designated as standard storage areas, where routine monitoring can be maintained. These area divisions are not permanent. Whenever the fusion risk index value of a storage location changes in subsequent updates and crosses the preset threshold boundary, its functional area assignment will be automatically and dynamically adjusted accordingly.

[0041] Finally, dynamic partitioning instructions are generated and published. The results of the above real-time partitioning, that is, which functional area each storage location belongs to, are clearly updated and displayed graphically on the digital twin map of the warehouse or the warehouse management interface.

[0042] S5 generates the optimal collaborative control strategy based on different warehouse functional areas.

[0043] The technical solutions described in the embodiments of this application have at least the following technical effects or advantages: By intelligently sensing and dynamically matching the storage environment with the characteristics of pharmaceuticals, the problem of declining pharmaceutical quality caused by unreasonable environmental settings is effectively solved. Its technological effects include significantly improving the quality of pharmaceutical storage and ensuring drug efficacy and storage safety. Advantages include achieving adaptive environmental control, reducing manual intervention costs, improving storage efficiency, and being particularly suitable for the refined, low-cost management of conventional warehouses in climate-changing environments.

[0044] Example 2: In Example 1, an integrated sensor network was deployed to build a fusion database, and a dual static digital profile of the drug was established for both microbial and photosensitive components, thereby enabling risk conflict identification and zoning management based on real-time environmental data. However, this technical solution treats both the photosensitive and microbial sensitivity parameters of the drug as static thresholds that remain fixed throughout the storage period. In reality, the stability of the active ingredients and the efficacy of preservatives in drugs undergo nonlinear decay over time due to the cumulative effects of environmental factors such as temperature, humidity, and light. Their tolerance to the external environment is a dynamic process of decline. The static threshold model cannot reflect the dynamic decay characteristics of drugs during actual storage, leading to a mismatch between the storage strategy and the actual drug state. This may cause two risks: first, when the drug has already experienced potential quality decay, the initial lenient threshold is still used, resulting in insufficient protection and quality risks; second, when the drug is still in good condition, unnecessary strict control cannot be lifted, leading to excessive protection and wasted resources. To achieve precise management of the stability of drugs throughout their entire life cycle and enable the storage strategy to adaptively match the real-time quality decay state of the drug, the static risk assessment model needs to be dynamically and predictively improved.

[0045] In some embodiments, a digital profile of each drug with both sensitive bacteria and photosensitive characteristics is formed. Step S2 further includes: S21. Input the initial physicochemical properties and accelerated stability test data of the drug into the digital profile to form the initial characteristic database of the drug.

[0046] Specifically, the initial physicochemical properties of the drug are extracted and entered from drug registration documents, quality standard documents, and stability study reports. These initial physicochemical properties include the drug's chemical structure and molecular formula, the detailed composition and ratio of the active pharmaceutical ingredient (API) and excipients, and its physical form (e.g., tablets, capsules, solutions). A complete description of the drug packaging is entered, including the inner packaging material that directly contacts the drug, the characteristics of the sealing components, and the outer packaging form. Accelerated stability test and long-term stability test data are integrated, including the test values ​​of key quality attributes (e.g., related substance content, main component content) over time under different temperature, humidity, and light conditions. For photostability, the photolysis test report must be entered, specifying the sensitive wavelength range; for microbial stability, data on preservative efficacy verification or packaging integrity testing must be entered. All information is stored in a structured manner to form a drug initial characteristic database.

[0047] Using the unique identifier of the drug (e.g., drug code) and the spatial coordinates of the storage location as key association fields, records in the initial drug feature database are matched with real-time environmental monitoring sequences of corresponding locations in the fusion database. For each drug storage location unit, its initial physicochemical properties and accelerated stability test data (as a static baseline) are bound to real-time environmental data collected at that point over time (as a dynamic variable). Based on a unified time axis, environmental monitoring data (e.g., temperature fluctuation sequences, cumulative light dose) are aligned with the corresponding critical threshold conditions in the drug stability test model.

[0048] S22. Extract real-time environmental monitoring sequences from each storage location from the fusion database, associate them with the digital feature files of the corresponding drugs in the storage location, and store them in time series to form a traceable dynamic database of drug-environment interaction.

[0049] S23. Based on the environmental historical data sequence and initial characteristics of drugs in the dynamic database, a drug active ingredient degradation rate prediction model and a drug preservative efficacy decay model are constructed by applying pharmacokinetic principles. The predicted degradation curve of the drug active ingredient and the real-time decay status of its preservative efficacy are output.

[0050] Specifically, for the target drug storage location, complete historical time-series data on temperature, humidity, and light intensity at various wavelengths are extracted from the drug-environment interaction dynamic database. Simultaneously, initial characteristic parameters of the drug are extracted, including the activation energy of photodegradation, pre-exponential factor, and initial preservative concentration. Using the extracted historical environmental data sequences and initial drug content change data, specific parameters in the overall degradation rate prediction model, composed of a thermal degradation rate term expressed as a temperature function, a photodegradation rate term expressed as a light intensity function at various wavelengths, and a humidity correction function, are determined through multivariate nonlinear regression fitting.

[0051] Based on historical environmental data and preservative activity detection data, a kinetic model for the decay of preservative activity was established through parameter fitting. This model describes the change in the residual active concentration of the preservative as a decay process driven by both temperature and humidity.

[0052] The prediction accuracy of the above model is verified using reserved historical data or recent sampled test data. When the prediction error exceeds the allowable range, a recursive parameter estimation algorithm is used to calibrate and update the key parameters in the model online.

[0053] By combining the calibrated model with real-time environmental monitoring data, a predicted concentration change curve of the drug's active ingredient over time is calculated and generated. Simultaneously, the residual activity percentage of the preservative at the current moment is calculated and output to characterize its real-time decay status.

[0054] S24. Based on the predicted degradation curve and the real-time decay status of the preservative efficacy, calculate the decay index reflecting the current stability status of the drug; dynamically adjust the maximum allowable cumulative light exposure of the drug according to the predicted degradation curve; and dynamically adjust its comprehensive microbial susceptibility index by combining the real-time microbial load and the predicted decay status of the preservative efficacy.

[0055] The process involves dynamically updating parameters based on the predicted degradation curve of the active pharmaceutical ingredient and the real-time decay status of preservative efficacy output from step S23. The current stability decay index of the drug is calculated, which is derived by weighting the predicted remaining percentage of active pharmaceutical ingredient and the real-time residual activity percentage of preservatives, with the weights pre-set according to the drug dosage form. Based on the stability decay trend reflected in the predicted degradation curve, the maximum permissible cumulative light exposure threshold of the drug is dynamically lowered. The new threshold is determined by calculating a function relating the initial threshold to the decay index or the remaining percentage of active pharmaceutical ingredient reflecting the current stability. Combining the microbial load data obtained from real-time monitoring and the predicted residual activity percentage of preservatives, the comprehensive microbial susceptibility index of the drug is dynamically adjusted upwards. This adjustment is achieved by multiplying the initial comprehensive microbial susceptibility index by a product factor composed of a preservative efficacy decay factor and a real-time microbial load adjustment function.

[0056] S25, recalculate the photodegradation and microbial contamination risk index of drugs in each storage location, reclassify the risk level based on the new risk index, and automatically generate and execute a storage location dynamic migration instruction when the drug risk level is upgraded, transferring the drug to a functional area that better matches its current stability state.

[0057] The technical solutions described in the embodiments of this application have at least the following technical effects or advantages: By introducing a dynamic drug stability degradation prediction and adaptive control mechanism, the level of intelligence in warehouse management has been significantly improved. Its technical effectiveness lies in its ability to predict the quality degradation trend of drugs in the storage environment in real time and dynamically adjust environmental thresholds and storage strategies, thereby effectively avoiding drug deterioration caused by unsuitable environments. Advantages include significantly improving drug storage quality and safety, reducing the cost of manual intervention, enhancing adaptability to seasonal changes and drug condition fluctuations, improving warehousing efficiency through optimized resource allocation, and ensuring the stability of drugs throughout their entire lifecycle.

[0058] Example 3: In Example 2, by introducing a dynamic prediction model for drug stability and chemical interaction analysis, a forward-looking assessment of drug quality degradation trends and cross-contamination risks during warehousing was achieved. However, the risk assessment and control decisions of this scheme still exhibit an isolated and sequential response mode at the implementation level. Specifically, its decision-making process is only triggered when a single drug's storage location is planned, passively simulating and warning of the physical stress of that specific movement path or the chemical cross-contamination risk of the target storage location. This mode has two key limitations: first, the decision-making perspective is localized, failing to comprehensively consider the chain risks and path conflicts that may be caused by multiple concurrent or continuous drug movement operations within the overall warehouse; second, the control strategy is lagging, only triggering countermeasures after the risk prediction results exceed the threshold, lacking a proactive mechanism to systematically avoid risks by optimizing the overall warehouse layout and operation scheduling before risks materialize. Faced with the complex intertwining of risks and resource competition brought about by multiple drugs and multiple tasks operating in parallel in dynamic warehouse operations, this model based on single-point, post-judgment inevitably suffers from insufficient global optimization capabilities and low response efficiency.

[0059] In some embodiments, step S22, which involves storing a dynamic database of traceable drug-environment interactions in a time-series manner, further includes: S221, Before the first entry of a drug into the warehouse, expand its dynamic database for each drug and add records of chemical interaction characteristics.

[0060] Among them, the chemical interaction characteristics record includes the volatile component profile (VOCs fingerprint) of the drug, chemical compatibility data with other common drugs, and physical stress tolerance thresholds such as vibration and pressure.

[0061] S222, when planning to adjust the storage location of medicines or perform inbound / outbound operations, based on the digital twin model of the warehouse and combined with the physical stress tolerance parameters in the chemical interaction characteristics, the planned movement path is simulated to calculate the physical environmental stress that the medicines will accumulate during the entire movement process.

[0062] Specifically, after a drug storage location adjustment or transfer instruction is generated, the planned movement path is extracted from the warehouse management digital twin model. This path is discretized into a series of continuous path segments, each of which is associated with the physical attributes of its specific transport equipment or medium, such as a conveyor belt, elevator, or automated guided vehicle, and these attributes are mapped to the corresponding physical modules in the digital twin model.

[0063] For each defined path segment, the stress sources in its physical environment are parameterized. Vibration stress is characterized as a frequency-dependent power spectral density function. Steady-state pressure is defined as a constant value. For sporadic impact events that may occur during transportation, a half-sine wave pulse model is used, defined by two parameters: peak acceleration and pulse duration.

[0064] Based on predefined physical stress tolerance parameters in the drug's digital archive, the stress levels experienced by the drug in each path segment are calculated. For vibration stress assessment, Miner's linear cumulative damage theory is used. The degree of vibration damage is determined by the ratio of the drug's exposure time in that path segment to the failure time calculated based on the stress-life curve and vibration power spectral density function. For steady-state pressure, the safety criterion is that the steady-state pressure value of the path segment is less than the drug's maximum tolerance pressure. For impact events, the safety criterion is that the peak ground acceleration of the impact is less than the drug's maximum tolerance acceleration.

[0065] Along the entire planned movement path, the various physical environmental stresses that the drug will experience are cumulatively analyzed. The total cumulative vibration damage is the sum of the vibration damage across all path segments. Early warning criteria are set as follows: the total cumulative vibration damage reaches or exceeds the critical damage threshold; or the steady-state pressure value of any path segment reaches or exceeds the drug's maximum tolerance pressure; or the peak impact acceleration of any path segment reaches or exceeds the drug's maximum tolerance acceleration. If any of the above early warning criteria are met, a physical stress over-limit risk warning is generated.

[0066] S223. Based on the calculated cumulative physical environmental stress and combined with the physical stability model of the drug, a stability decay prediction curve for the drug during this movement is constructed.

[0067] Specifically, this is based on the cumulative physical environmental stress data of the drug along its planned movement path. First, key parameters of its physical stability model are retrieved from the drug's digital profile. These key parameters include vibration sensitivity and pressure sensitivity, obtained through standardized accelerated stability testing, used to quantify the expected impact of unit physical stress on the drug's critical quality attributes.

[0068] Based on these parameters and the calculated cumulative physical environmental stress, a stability degradation prediction curve for the drug during the entire movement process is constructed. For vibration stress, the expected stability degradation is calculated by multiplying the vibration sensitivity coefficient by the total cumulative vibration damage. For steady-state pressure, the expected stability degradation is calculated by multiplying the pressure sensitivity coefficient by the pressure difference exceeding a preset stability impact threshold.

[0069] Combining the two attenuation factors mentioned above with a function reflecting the correlation between the attenuation process and the movement time or path progress, a predicted stability index for the drug at any time or location during its movement is calculated. This index value decreases from the initial stability baseline before the movement.

[0070] The generated predicted stability decay curve is continuously compared with predefined stability safety thresholds in the drug's digital file. When the stability index value at any point on the predicted curve falls below the preset safety threshold, it is determined that the movement operation will expose the drug to stability risks exceeding acceptable limits, triggering a formal stability risk warning. This warning information includes the specific predicted decay value and the identified risk exceeding the limit locations along the movement path.

[0071] S224, based on a computational fluid dynamics model, simulates in real time the airflow state in the warehouse and the diffusion path and concentration distribution of VOCs released from existing stocked drugs; when a drug is planned to be moved to the target storage location, it calls the VOCs fingerprint of the drug and its chemical compatibility data with existing drugs around the target storage location in real time, and calculates the potential cross-reaction risk index after its volatiles are mixed with VOCs in the surrounding environment.

[0072] Specifically, the system integrates environmental parameters acquired from a real-time monitoring network distributed throughout the warehouse, including temperature field data for each area and airflow velocity field boundary conditions collected by an array of wind speed sensors. It then iterates through the inventory of all medicines stored in the warehouse, calculating the volatile organic compound (VOC) flux and modeling it as a spatially distributed VOC release source based on the VOC fingerprint recorded in the digital profile of each medicine and its physical properties at the current local ambient temperature. By superimposing the release source terms of all medicines, a spatial source term field reflecting the VOC release situation of the entire warehouse is constructed for simulation.

[0073] Based on the integrated environmental conditions and the constructed VOCs emission source field, a computational fluid dynamics model was initiated for transient solutions. Within each computational time step, the model first solved the governing equations to update the air velocity and pressure distribution within the warehouse. For each volatile organic compound component, the mass transport equations were solved to calculate the real-time three-dimensional concentration distribution of that component throughout the entire warehouse space.

[0074] When a drug is planned to be moved to a specific target storage location, all volatile component identifiers and their volatilization kinetic parameters for that drug are retrieved from the drug feature tag database. Based on the real-time ambient temperature of the target storage location, the volatilization rate of each component is calculated, and it is set as a temporary release source located at the target coordinates in the computational fluid dynamics model to quickly simulate the diffusion process of its volatiles in the existing flow field and background VOCs concentration field. Simultaneously, background concentration data of various VOCs at all spatial points within a specified radius around the target storage location are extracted from the calculated steady-state background concentration field.

[0075] A chemical compatibility database of the target drug and all existing drugs in the surrounding area was retrieved to obtain a quantitative incompatibility index between each volatile component of the target drug and each volatile component in the background environment. By weighting and synthesizing the predicted concentration distribution of the target drug after simulated diffusion, the extracted background concentration distribution, and the corresponding incompatibility index, the incremental risk of local cross-reactions at various points in the surrounding space after the introduction of the target drug was obtained (the calculation formula is: Where i represents all volatile organic compounds (VOCs) released by the target drug; j represents all VOCs released by other drugs in the warehouse background environment. This indicates that all possible combinations of interacting volatile components are traversed and accumulated; It is a chemical incompatibility index used to quantify the tendency or intensity of adverse chemical reactions between volatile component i of the target drug and volatile component j in the background environment. It is obtained through experimental data or theoretical calculations (prediction of molecular reactivity) and assigns a value to each pair of potentially interacting components. In a three-dimensional spatial location label, a spatial point (x, y, z) is defined in the length, width, and vertical height directions. The predicted steady-state concentration of the i-th volatile component released by the target drug at its planned storage location was obtained through computational fluid dynamics (CFD) simulation. The simulation used the VOCs release rate of the target drug at its planned storage location as the source term, and the spatial concentration distribution was obtained after diffusion calculations in the existing warehouse flow field and background concentration field. It is at a point in space (x, y, z) The concentration of the j-th volatile component already present in the warehouse's background environment is used as an example. This is also derived from a CFD simulation, which uses the VOCs release from all existing pharmaceuticals in the warehouse as the source term to calculate the background VOCs concentration distribution before the introduction of the target pharmaceutical product. Spatial aggregation (integration or taking the maximum value) is performed on all risk increments in this area (e.g., adjacent storage locations) to calculate the comprehensive cross-reactivity risk index caused by this pharmaceutical adjustment plan. The comprehensive cross-reactivity risk index is compared with a preset risk safety threshold for this area, which may be dynamically adjusted based on the value and sensitivity of the pharmaceutical products. If the risk index reaches or exceeds the safety threshold, it is determined that this storage location adjustment poses an unacceptable risk of cross-contamination, and an early warning message is automatically generated containing specific risk values, the main VOCs component pairs involved, and the coordinates of the location with the highest risk. Simultaneously, this risk event is associated with the relevant pharmaceutical batch and storage location.

[0076] The technical solutions described in the embodiments of this application have at least the following technical effects or advantages: By introducing chemical interaction analysis and physical stress movement simulation of pharmaceuticals, this technology overcomes the limitations of traditional warehouse management that only focuses on static environmental parameters. Its technological advantage lies in its ability to proactively predict and mitigate the risks of chemical contamination caused by cross-contamination of volatile substances between pharmaceuticals, as well as the risks of physical damage during handling and transportation. Its advantages are reflected in achieving a leap from environmental monitoring to the coordinated prevention and control of components and stress. Through digital twins and fluid dynamics simulation, potential risks are accurately quantified in three-dimensional space, allowing for early warnings and optimized decisions before warehouse location adjustments are implemented. This significantly improves the quality and safety assurance level of pharmaceuticals throughout the entire warehousing and circulation process, achieving truly proactive, preventative, and refined management.

[0077] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for managing drug warehouse storage, characterized in that, include: S1. Deploy integrated sensor network nodes within the warehouse. Each node synchronously collects multi-source data from its location and adds timestamps and three-dimensional spatial location labels to all collected data to build a spatiotemporally synchronized fusion database. S2 establishes sensitivity labels for each drug stored in the warehouse in two dimensions: photosensitivity and microbial sensitivity, forming a digital profile of each drug with dual characteristics of sensitive bacteria and photosensitivity. S3, based on the fusion database and digital images, calculates the microbial contamination risk index and photodegradation risk index of medicines in each storage location under the current environment, and identifies risk conflict areas where both high microbial contamination risk and high photodegradation risk exist simultaneously. S4: For the identified risk conflict areas, calculate their fusion risk index, and dynamically divide different warehouse functional areas based on the real-time fusion risk index of each area in the entire warehouse. S5 generates the optimal collaborative control strategy based on different warehouse functional areas.

2. The drug warehouse storage management method as described in claim 1, characterized in that, The process of creating a digital profile of each drug with dual biosensitivity and photosensitivity specifically includes: analyzing the drug's photosensitivity based on its chemical composition, physical form, and photolysis characteristics; determining its sensitive wavelength range; establishing a photodegradation kinetic model; and determining its maximum allowable cumulative light exposure. Based on the drug's formulation, packaging materials, and preservative information, the process also includes assessing its microbial contamination sensitivity, determining its microbial sensitivity level, and identifying susceptible microbial types. Finally, the photosensitivity parameters and the quantification results of microbial sensitivity are structurally integrated to form a digital profile of the drug.

3. The drug warehouse storage management method as described in claim 1, characterized in that, The risk conflict area includes: extracting real-time environmental monitoring data for each storage location from a spatiotemporally synchronized fusion database, and retrieving digital feature files of the corresponding drugs stored in the storage location from the drug feature label library; calculating the photodegradation risk index of drugs in each storage location based on the light data in the extracted environmental monitoring data and the light sensitivity parameters in the digital feature files; calculating the microbial contamination risk index of drugs in each storage location based on the microbial load, temperature, and humidity data in the extracted environmental monitoring data, and the microbial sensitivity parameters in the digital feature files; setting high thresholds for photodegradation risk and microbial contamination risk, and identifying storage locations that simultaneously exceed both high thresholds as risk conflict areas.

4. The drug warehouse storage management method as described in claim 3, characterized in that, Step S2, which involves forming a digital profile of each drug with dual biosensitive and photosensitive features, further includes: S21. Input the initial physicochemical properties and accelerated stability test data of the drug into the digital profile to form the initial characteristic library of the drug. S22, extract the real-time environmental monitoring sequence of each storage location from the fusion database, associate it with the digital feature file of the corresponding drug in the storage location, and store it in time series to form a traceable dynamic database of drug-environment interaction; S23. Based on the environmental historical data sequence and the initial characteristics of the drug in the dynamic database, the drug active ingredient degradation rate prediction model and the drug preservative efficacy decay model are constructed by applying the principles of pharmacokinetics, and the predicted degradation curve of the drug active ingredient and the real-time decay status of its preservative efficacy are output. S24. Based on the predicted degradation curve and the real-time decay status of the preservative efficacy, calculate the decay index reflecting the current stability status of the drug; dynamically adjust the maximum allowable cumulative light exposure of the drug based on the predicted degradation curve; and dynamically adjust its comprehensive microbial susceptibility index by combining the real-time microbial load and the predicted decay status of the preservative efficacy. S25, recalculate the photodegradation and microbial contamination risk index of drugs in each storage location, reclassify the risk level based on the new risk index, and automatically generate and execute a storage location dynamic migration instruction when the drug risk level is upgraded, transferring the drug to a functional area that better matches its current stability state.

5. The pharmaceutical warehouse storage management method as described in claim 4, characterized in that, The process of outputting the predicted degradation curve of the active pharmaceutical ingredient and its real-time decay status of preservative efficacy specifically includes: establishing a degradation rate prediction model for the active pharmaceutical ingredient and a decay model for the preservative efficacy of the drug based on historical environmental data sequences and initial drug characteristic parameters in a dynamic database through parameter fitting; verifying the accuracy of the prediction model and the decay model using reserved monitoring data or sampling test data, and dynamically updating the model parameters according to the verification results; combining the parameter-updated model with real-time environmental monitoring data to generate and output the predicted degradation curve of the active pharmaceutical ingredient and its real-time decay status of preservative efficacy.

6. The pharmaceutical warehouse storage management method as described in claim 5, characterized in that, Step S24 specifically includes: determining the current stability state of the drug based on the predicted degradation curve and real-time decay status; dynamically reducing the maximum permissible cumulative light exposure threshold of the drug based on the current stability state; and dynamically increasing the comprehensive microbial susceptibility index of the drug by combining real-time environmental monitoring data and real-time decay status.

7. The pharmaceutical warehouse storage management method as described in claim 3, characterized in that, In step S22, storing a dynamic database of traceable drug-environment interactions in a time-series manner includes: S221. Before the first entry of a drug into the warehouse, expand its dynamic database for each drug and add chemical interaction characteristic records. Among them, the chemical interaction characteristic records include the volatile component spectrum (VOCs fingerprint) of the drug, chemical compatibility data with other common drugs, and physical stress tolerance threshold. S222, when planning to adjust the storage location of medicines or to perform inbound / outbound operations, based on the digital twin model of the warehouse and combined with the physical stress tolerance parameters in the chemical interaction characteristics, the planned movement path is simulated to calculate the physical environmental stress that the medicines will accumulate during the entire movement process. S223. Based on the calculated cumulative physical environmental stress and combined with the physical stability model of the drug, a stability decay prediction curve of the drug during this movement is constructed. S224, based on a computational fluid dynamics model, simulates in real time the airflow state in the warehouse and the diffusion path and concentration distribution of VOCs released from existing stocked drugs; when a drug is planned to be moved to the target storage location, it calls the VOCs fingerprint of the drug and its chemical compatibility data with existing drugs around the target storage location in real time, and calculates the potential cross-reaction risk index after its volatiles are mixed with VOCs in the surrounding environment.

8. The pharmaceutical warehouse storage management method as described in claim 7, characterized in that, The calculation of the cumulative physical environmental stress that the drug will experience during the entire movement process is as follows: Based on the digital twin model of the warehouse, the planned drug movement path is extracted and discretized into multiple continuous path segments; the physical environmental stress sources of each path segment are parameterized and defined; based on the physical stress tolerance parameters of the drug, the physical environmental stress level borne by the drug in each path segment is calculated; the various physical environmental stresses that the drug will bear along the entire movement path are cumulatively analyzed, and a physical stress over-limit risk warning is generated based on the comparison results between the cumulative physical environmental stress and the physical stress tolerance threshold of the drug.

9. The pharmaceutical warehouse storage management method as described in claim 7, characterized in that, The construction of the drug's stability degradation prediction curve during this migration includes: Based on the calculated cumulative physical environmental stress and the physical stability model parameters of the drug, a stability decay prediction curve for the drug during this movement is constructed. According to the stability decay prediction curve, the predicted stability index of the drug at each moment during the movement is calculated. The predicted stability index is compared with the preset stability safety threshold, and a drug movement stability risk warning is triggered when the predicted stability index is lower than the safety threshold.

10. The pharmaceutical warehouse storage management method as described in claim 7, characterized in that, The potential cross-reaction risk index specifically includes: integrating real-time environmental parameters of the warehouse with the VOCs release characteristics of all stored drugs to construct a spatial source term field reflecting the VOCs release situation of the entire warehouse; based on the spatial source term field and environmental boundary conditions, obtaining the real-time three-dimensional concentration distribution of each VOCs component in the warehouse through computational fluid dynamics model; when a drug is planned to be moved to the target storage location, it is set as a temporary release source in the model, simulating its volatile diffusion and obtaining its predicted concentration distribution in the target area; extracting the background concentration distribution of each VOCs component around the target area from the background concentration field; obtaining the chemical compatibility data between the target drug and existing drugs in the surrounding area, and calculating the comprehensive cross-reaction risk index caused by the introduction of the target drug based on the predicted concentration distribution, background concentration distribution, and chemical compatibility data; comparing the comprehensive cross-reaction risk index with a preset dynamic safety threshold, and generating a cross-contamination risk warning if the threshold is exceeded.

Images