In-vitro diagnostic reagent supply chain whole-chain monitoring information integrated management method

Abstract

This invention discloses a method for integrated management of full-chain monitoring information in the in vitro diagnostic reagent supply chain, specifically relating to the field of medical supply chain information management technology. It addresses the problems of low delivery efficiency and insufficient reagent quality assurance in existing in vitro diagnostic reagent in-hospital distribution processes due to fixed scheduling rules failing to adapt to dynamic environments. The method involves establishing an initial path set by acquiring reagent attribute information and hospital environmental structure information; identifying business logic dependencies through time-series similarity clustering analysis of reagent consumption fluctuation patterns; binding associated reagents as delivery units and generating aggregated attribute information; quantifying the demand intensity of delivery units by real-time assessment of logistics resource contention conflicts; solving for optimal delivery priorities using a multi-agent game model; and finally generating delivery instructions based on priority-based planning of specific paths and time schedules. The execution results are then fed back to the full-chain monitoring system to achieve refined and adaptive management of the in vitro diagnostic reagent supply chain.

Description

Technical Field

[0001] This invention relates to the field of medical supply chain information management technology, and in particular to a method for integrated management of full-chain monitoring information of in vitro diagnostic reagent supply chain. Background Technology

[0002] Supply chain management for in vitro diagnostic reagents involves multiple stages, including production, warehousing, transportation, and in-hospital distribution. Its core requirement is ensuring the stability and traceability of reagent quality throughout the entire process. Especially in the in-hospital logistics stage, reagents must be delivered to multiple end-stage departments such as the laboratory and operating room under strict temperature control conditions, while simultaneously meeting the urgency and diversity requirements of clinical applications. Currently, the industry generally uses information systems to record and track the status of reagent distribution, relying on fixed rules (such as order sequence, departmental priority, or simple route planning) for scheduling and arranging in-hospital distribution tasks. While this approach improves management efficiency to some extent, it still relies on static task allocation and experience-based decision-making.

[0003] However, existing technologies are insufficient to effectively coordinate the multi-objective optimization needs under various dynamic constraints in hospital logistics. Because in vitro diagnostic reagents are strictly sensitive to temperature control and timeliness, and the time windows during which each reagent can be removed from the temperature-controlled environment differ, coupled with the constantly changing hospital environment (e.g., departmental priorities, competition for logistics resources, and constantly changing route conditions), fixed-rule scheduling methods cannot respond to these coupled variable factors in real time. This results in significant conflicts between optimizing overall distribution efficiency, ensuring reagent safety, and meeting urgent clinical needs, thereby affecting the reliability of reagent distribution and the overall supply chain management level. Summary of the Invention

[0004] This invention addresses the technical problems existing in the prior art by providing an integrated management method for monitoring information across the entire supply chain of in vitro diagnostic reagents.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] A comprehensive management method for monitoring and managing the entire supply chain of in vitro diagnostic reagents, including:

[0007] S1. Obtain the attribute information, upstream batch and quality information, and hospital environment structure information of the reagents to be delivered;

[0008] S2. Based on the hospital's environmental structure information, establish an initial path set covering all accessible paths within the hospital;

[0009] S3. Based on time-series similarity clustering analysis, analyze the correlation of consumption fluctuation patterns in the historical consumption data of each reagent to determine whether there is a business logic dependency.

[0010] S4. When there is a business logic dependency, the associated reagents are bound to the same delivery unit, and the aggregated attribute information of the delivery unit is generated based on the attribute information of the reagents to be delivered.

[0011] S5. Real-time assessment of logistics resource contention conflicts, and quantification of the real-time demand intensity of each delivery unit for logistics resources based on aggregated attribute information, business logic dependencies, and upstream batch and quality information.

[0012] S6. Construct each delivery unit into a multi-agent game relationship, and solve the game equilibrium allocation scheme based on real-time demand intensity and logistics resource competition conflict to generate the delivery priority of each delivery unit.

[0013] S7. Based on delivery priority, plan specific travel routes and time arrangements for each delivery unit in the initial path set to generate delivery instructions, and after execution, feed them back to the supply chain full-chain monitoring system.

[0014] Furthermore, the attribute information of the reagents to be delivered is obtained, including the safe time window during which the reagents are allowed to be removed from the temperature-controlled environment and the temperature requirements for reagent transportation.

[0015] Obtain upstream batch and quality information, including reagent production batch number, expiration date, and temperature control records during upstream distribution.

[0016] Obtaining hospital environmental structure information includes hospital building distribution maps, department location maps, and information on the location of passageways and elevators.

[0017] Furthermore, establishing an initial path set based on hospital environmental structure information includes: constructing a hospital indoor and outdoor space connectivity graph based on the hospital building distribution map, department location map, and corridor and elevator location information; traversing all nodes corresponding to department location information in the graph, calculating the shortest path between nodes, and forming an initial path set.

[0018] Furthermore, the correlation of consumption fluctuation patterns of historical consumption data of each reagent based on time-series similarity clustering analysis includes: normalizing the historical consumption time-series data of each reagent, calculating the morphological similarity between different historical consumption time-series data of reagents, classifying reagents with morphological similarity exceeding a preset threshold into the same cluster group; and labeling the main consuming departments associated with each cluster group.

[0019] Determining whether a business logic dependency exists includes: if reagents are grouped into the same cluster, then it is determined that there is a business logic dependency between these reagents.

[0020] Furthermore, calculating the morphological similarity between time series data of historical consumption of different reagents includes: normalizing the time series data of historical consumption of each reagent to eliminate dimensional differences, and quantifying the morphological similarity by calculating the matching degree of shape and fluctuation trend between sequences.

[0021] Furthermore, binding related reagents into the same delivery unit means allocating reagents belonging to the same cluster group to the same logical container for unified delivery management;

[0022] The aggregated attribute information of the delivery unit is generated based on the attribute information of the reagents to be delivered. This includes the earliest expiration time within the safe time window during which all reagents in the group are allowed to leave the temperature-controlled environment, and the most stringent temperature requirement among the reagent transportation temperature requirements for all reagents in the group.

[0023] Furthermore, real-time assessment of logistics resource contention conflicts refers to monitoring the real-time occupancy status of each passage and elevator in hospital access and elevator location information through IoT sensors;

[0024] The real-time demand intensity of each delivery unit for logistics resources is quantified based on aggregated attribute information, business logic dependencies, and upstream batch and quality information. This includes: first, comparing the remaining time window of each delivery unit; the shorter the remaining time, the higher the demand intensity. When the remaining time is similar, the real-time demand intensity is determined based on the department priority information of the main consuming departments associated with the cluster group to which the delivery unit belongs. When there is a quality risk situation such as reagent expiration date approaching or temperature value in the upstream circulation process temperature control record continuously exceeding the reagent transportation temperature requirement threshold for a preset time, the real-time demand intensity level of the delivery unit is increased accordingly.

[0025] Furthermore, constructing each delivery unit as a multi-agent game relationship means treating each delivery unit as an independent agent;

[0026] The process of generating delivery priorities for each delivery unit based on the real-time demand intensity and logistics resource contention conflict to solve the game equilibrium allocation scheme includes: using the real-time demand intensity of each delivery unit for logistics resources as the basis of the utility function of each agent, using the logistics resource contention conflict as the resource allocation constraint, solving the Nash equilibrium point through a non-cooperative game process among agents, and determining the delivery priority of each delivery unit based on the strategy payoff of each agent at the equilibrium point.

[0027] Furthermore, solving for the Nash equilibrium point through a non-cooperative game process among agents involves: each agent proposing a resource usage strategy based on its own real-time demand intensity, and through an iterative process of multiple rounds of strategy adjustment and payoff calculation, ultimately converging to a Nash equilibrium state where all agent strategies are stable.

[0028] Furthermore, generating delivery instructions based on delivery priority and initial path set includes: determining the final delivery order of each delivery unit according to delivery priority, and planning specific travel routes and time arrangements for the final delivery order based on the initial path set;

[0029] Feedback to the supply chain monitoring system after execution refers to uploading real-time status information during the delivery task execution process and the final actual delivery route information to the supply chain monitoring system.

[0030] The beneficial effects of this invention are:

[0031] 1. By establishing a dynamic path planning system covering the entire hospital environment and combining it with intelligent clustering analysis of reagent consumption patterns, the system has achieved precise allocation of logistics resources within the hospital. Through temporal similarity clustering, the system identifies the business logic dependencies between reagents and binds reagents with collaborative usage characteristics into a unified delivery unit, effectively reducing the frequency of repeated deliveries. The introduction of a multi-agent game model enables the system to dynamically generate the optimal delivery priority based on real-time demand intensity and resource contention, thereby significantly improving its adaptability to the complex and ever-changing factors in the hospital environment.

[0032] 2. By generating aggregated attribute information and implementing a full-chain monitoring and feedback mechanism, the quality controllability of reagent delivery is ensured throughout the entire process. Key attributes such as reagent temperature control requirements and safety time windows are intelligently aggregated, and the optimal route is planned based on real-time logistics resource status. This not only ensures the timeliness and safety of reagents but also maximizes delivery efficiency. Data feedback throughout the process is fed back to the monitoring system to form a closed-loop optimization, providing data support for continuous improvement of supply chain management and effectively enhancing the overall control level of the in vitro diagnostic reagent supply chain. Attached Figure Description

[0033] Figure 1 This is a flowchart of the method for integrated management of monitoring information across the entire supply chain of in vitro diagnostic reagents according to the present invention;

[0034] Figure 2 This is a flowchart for generating delivery priorities for each delivery unit in this invention. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] Example 1: Figure 1 This invention provides a method for integrated management of the entire supply chain monitoring information of in vitro diagnostic reagents, including:

[0037] S1. Obtain the attribute information, upstream batch and quality information, and hospital environment structure information of the reagents to be delivered;

[0038] S2. Based on the hospital's environmental structure information, establish an initial path set covering all accessible paths within the hospital;

[0039] S3. Based on time-series similarity clustering analysis, analyze the correlation of consumption fluctuation patterns in the historical consumption data of each reagent to determine whether there is a business logic dependency.

[0040] S4. When there is a business logic dependency, the associated reagents are bound to the same delivery unit, and the aggregated attribute information of the delivery unit is generated based on the attribute information of the reagents to be delivered.

[0041] S5. Real-time assessment of logistics resource contention conflicts, and quantification of the real-time demand intensity of each delivery unit for logistics resources based on aggregated attribute information, business logic dependencies, and upstream batch and quality information.

[0042] S6. Construct each delivery unit into a multi-agent game relationship, and solve the game equilibrium allocation scheme based on real-time demand intensity and logistics resource competition conflict to generate the delivery priority of each delivery unit.

[0043] S7. Based on delivery priority, plan specific travel routes and time arrangements for each delivery unit in the initial path set to generate delivery instructions, and after execution, feed them back to the supply chain full-chain monitoring system.

[0044] When acquiring the attribute information, upstream batch and quality information, and hospital environmental structure information of the reagents to be delivered, the basic attribute data of the reagents to be delivered is first extracted from the reagent management database. This reagent management database is maintained and updated daily and in real time by the hospital's reagent management department through the central server of the supply chain system to ensure the accuracy and timeliness of the data. The attribute information of the safe time window for the reagent to be removed from the temperature-controlled environment is obtained by querying the product stability study report provided by the reagent manufacturer regarding the maximum permissible exposure time of the reagent under uncontrolled conditions. For example, the maximum permissible exposure time for a certain immunoassay reagent at 25 degrees Celsius is 30 minutes. This data is accurately recorded in minutes and stored in the corresponding field of the database. This value is set based on the results of accelerated stability testing verified by the manufacturer to ensure its scientific validity and reliability.

[0045] The temperature requirements for reagent transportation are obtained based on the temperature range clearly indicated in the reagent packaging instructions or the transportation specifications document provided by the manufacturer. For example, it is required to maintain the temperature between 2 and 8 degrees Celsius throughout the process. This information will be parsed and stored as two numerical fields: minimum temperature and maximum temperature, for subsequent verification and monitoring. The temperature range is usually set based on the heat sensitivity study of the active ingredient of the reagent. The requirements may vary for different reagents. For example, some freezing reagents require transportation below -20 degrees Celsius.

[0046] For obtaining upstream batch and quality information, the reagent production batch number and information are obtained by scanning the barcode or QR code on the outer packaging of the reagent and parsing the identification data contained therein that conforms to international coding standards (such as GS1 standard). The batch number is associated with the record in the production database to ensure its uniqueness and traceability. The scanning operation is completed using a handheld barcode scanner or mobile device camera, and the parsed data is verified for its validity through a verification algorithm.

[0047] The reagent expiration date is obtained from the barcode parsing results or by directly querying the batch expiration list provided by the manufacturer. This date information is stored in year-month-day format and compared with the current system date to verify validity. For example, the system calculates the difference in days between the current date and the expiration date; if it is less than or equal to 30 days, a warning is triggered. Temperature control records from the upstream process are obtained by accessing the logistics company's temperature monitoring platform application interface to extract the full-process temperature time-series data generated by the temperature recorder during the previous transportation stage. This record includes temperature values ​​collected at least once per minute and their corresponding timestamps. For example, one record may contain 1200 temperature data points from loading to unloading. This data is downloaded completely and temporarily stored in a local database for subsequent quality analysis. The temperature recorder's accuracy is typically required to be within ±0.5 degrees Celsius to ensure data reliability.

[0048] In the process of acquiring hospital environmental structure information, the hospital building distribution map is obtained by importing building information model files or computer-aided design drawings provided by the hospital's infrastructure department and converting them into a standardized digital map containing floor, room number and building structure information. This map will mark key topological information such as the name of each building, floor height, staircase and entrance location. For example, the floor height of the 1st to 5th floors of building A is 4 meters, and the geometric data of each structure is stored in vector graphics format.

[0049] The acquisition of the department location map is based on the above building distribution map, and further integrates the department deployment data in the hospital information system. The unique identifier of each department is bound to its actual building number, floor number and room number, thereby generating a visual map that can reflect the precise geographical location of each department. For example, the cardiology department is located in rooms 301 to 305 on the 3rd floor of building B. This binding relationship is realized through the association table of the department management database.

[0050] The acquisition of passage and elevator location information is achieved by extracting passage connection data from the building distribution map and equipment installation drawings provided by the elevator manufacturer. This process clearly marks the specific locations of all main passages, emergency passages, elevator shafts, and elevator floors within the building. This information is then stored as a set of node and edge data with a coordinate system. For example, the width attribute of each passage is recorded as a numerical value (unit: meters), and the load and speed parameters of the elevators are also stored for subsequent path planning algorithm calls and processing.

[0051] All acquired environmental structure information undergoes coordinate standardization and topology verification to ensure data consistency and usability. For example, coordinates from drawings from different sources are converted to a relative coordinate system with the hospital's main entrance as the origin, and the connectivity of passageways is logically verified to avoid isolated nodes or unconnected paths. The entire information acquisition process is ensured to have completeness through data verification mechanisms. For example, each reagent record is checked to ensure its attribute fields are not empty, and the environmental structure data is checked to ensure its topological logic is correct. When missing or incorrect information is found, a data re-acquisition or manual correction process is automatically triggered.

[0052] When implementing the step of establishing an initial path set covering all accessible paths in the hospital based on the hospital's environmental structure information, the first step is to construct an indoor and outdoor spatial connectivity map of the hospital based on the acquired hospital building distribution map, department location map, and corridor and elevator location information. The construction process is based on the building structure data contained in the hospital building distribution map. Each independent room, corridor intersection, stairwell, elevator lobby, and building entrance / exit is abstracted as a node in the graph structure, and each node is assigned a unique identifier and spatial coordinate information. The coordinate information comes from the actual dimensions marked on the drawings; for example, a scale where 1 centimeter represents 5 meters in actual distance is used to calculate the actual distance between nodes.

[0053] The connections between nodes are established through the location information of passageways and elevators. The connecting passageway between two parallel corridors is abstracted as an edge, and the weight of the edge is set according to the actual length of the passageway. For example, the weight of the edge corresponding to a corridor with a length of 20 meters is 20. The weight unit is uniformly used in meters to ensure consistency of units. For vertical transportation facilities such as elevators and stairs, they are abstracted as special edges connecting corresponding positions on different floors, and are assigned corresponding weight values ​​according to their transportation efficiency. For example, the weight of the elevator is set as the difference in floor height multiplied by a coefficient, which is usually between 1.5 and 2.0 to reflect the time consumption of vertical movement relative to horizontal movement.

[0054] Based on the construction of a complete connected graph, the system traverses all nodes corresponding to the department locations. These department nodes are determined by the correspondence between departments and rooms recorded in the department location map; for example, the cardiology department corresponds to node N305, and the laboratory department corresponds to node N102. The shortest path between nodes is calculated using a classic graph theory algorithm. This algorithm begins by initializing a distance matrix between all nodes. In the matrix, the distance from a node to itself is set to 0, the distance between directly connected nodes is set to the actual length of the passage, and the distance between non-directly connected nodes is initialized to a sufficiently large value, such as 10,000 meters, which should be greater than the maximum possible distance between any two nodes within the hospital. The core iterative process of the algorithm involves traversing all node pairs, comparing the path length via intermediate nodes with the currently recorded shortest path length. If a shorter path is found, the distance matrix and path information are updated. The iteration continues until the shortest path for all node pairs no longer changes or the maximum number of iterations is reached; for example, setting the upper limit of the number of iterations to three times the number of nodes ensures convergence. During the calculation process, some special cases also need to be considered. For example, when encountering temporarily closed passages or elevators under maintenance, these edges need to be temporarily removed from the graph and these unusable paths need to be excluded during path calculation.

[0055] After calculation, the shortest path information between all department node pairs is stored as an initial path set. This set is organized in a list structure, with each path record containing the starting department node, the ending department node, the sequence of nodes traversed, and the total path length. For example, the shortest path from node N102 to node N305 might be recorded as passing through nodes N102, N110, N205, and N305, with a total length of 125 meters. To ensure path feasibility, the special characteristics of elevators are considered during the calculation process. Elevator edge weights include not only vertical distance but also the distance equivalent converted from the average waiting time; for example, a 30-second waiting time is converted into a 15-meter horizontal distance. The final generated initial path set contains the shortest reachable paths between nodes corresponding to the location information of all departments in the hospital, providing basic data support for subsequent delivery route planning. The entire construction process is implemented through automated scripts, with environmental structure information data files as input and structured path set data as output, which can be directly called by subsequent steps. Before outputting the results, the generated path set needs to be validated. For example, several node pairs can be randomly selected, and the shortest paths calculated can be manually verified to ensure their accuracy. Additionally, a regular update mechanism needs to be established to promptly reconstruct the initial path set when the hospital environment changes, ensuring data timeliness.

[0056] When implementing the step of cluster analysis based on time-series similarity to determine the correlation of consumption fluctuation patterns in the historical consumption time-series data of each reagent, the historical consumption time-series data of each reagent is first normalized. This process involves extracting the daily consumption sequence of each reagent within a preset time range, for example, selecting the daily consumption data of the most recent 365 days as input, and performing maximum and minimum value normalization on the consumption sequence of each reagent. This linearly transforms the original consumption data to the interval between 0 and 1, eliminating the dimensional influence caused by the difference in consumption levels of different reagents. The normalization process is calculated by subtracting the minimum value of the sequence from each data point and then dividing by the sequence range.

[0057] During the normalization process, it is necessary to handle possible missing values. For example, when the consumption data for a certain day is missing, the average of the consumption data for the three days before and after is used to fill in the missing data to ensure data continuity. At the same time, for cases where multiple consecutive days are missing, the upper limit of the number of missing days is set to 7 days. Sequences exceeding this number of days will be marked as invalid data.

[0058] After normalization, the morphological similarity between time series data of historical reagent consumption is calculated. The morphological similarity is calculated by comparing the matching degree of two normalized time series in terms of shape features and fluctuation trends. Specifically, the dynamic time warping method is used to calculate the similarity distance between the series. This method can effectively handle the nonlinear scaling problem of time series on the time axis. For example, the cumulative distance matrix of two reagent consumption series at each time point is calculated. The optimal warping path is found through dynamic programming algorithm. The total path distance is finally obtained as a quantitative index of morphological similarity. The smaller the distance value, the higher the morphological similarity. The distance value ranges from 0 to positive infinity. The smaller the value, the higher the similarity.

[0059] After obtaining the morphological similarity measure between all reagents, reagents with morphological similarity exceeding a preset threshold are grouped into the same cluster. The preset threshold is determined by analyzing the actual distribution of reagent consumption patterns in historical data. For example, by calculating the average and standard deviation of the morphological similarity distance between all reagent pairs, the threshold is set to the average minus one standard deviation, so that about 15% of reagent pairs can be classified into the same cluster. This threshold setting method is based on the normal distribution assumption and can be adjusted according to the specific data distribution characteristics in practical applications.

[0060] After each cluster is generated, it is necessary to label it with the information of the main consuming departments. This process is achieved by statistically analyzing the consumption frequency and consumption ratio of all reagents in each department within the cluster. For example, if a cluster contains 5 reagents, and the consumption of 4 of them accounts for more than 70% in the laboratory, then the main consuming department of the cluster is labeled as the laboratory. During the labeling process, a percentage threshold of 50% is set, that is, when the consumption ratio of a certain department exceeds 50%, it is identified as a main consuming department.

[0061] The process of determining whether a business logic dependency exists is based on the clustering results. If reagents are grouped into the same cluster, it is determined that there is a business logic dependency between these reagents. This dependency indicates that these reagents have synergy or substitutability in clinical use. For example, multiple reagents required for the same test item often show highly similar consumption fluctuation patterns and are therefore classified into the same cluster. The criteria for determining a business logic dependency are that the number of reagents in the cluster is at least 2 and the average morphological similarity distance between reagents in the group is lower than the overall average distance.

[0062] The entire analysis process is implemented using a hierarchical clustering algorithm. This algorithm starts with each reagent as a separate cluster and gradually merges the clusters with the highest morphological similarity until the similarity between all clusters is lower than a preset threshold. The final clustering results reflect the inherent correlation of reagent consumption patterns and provide a basis for binding subsequent delivery units.

[0063] Cluster analysis requires setting reasonable termination conditions, such as stopping merging when the maximum number of clusters reaches a preset limit or the minimum similarity between clusters falls below a certain threshold. This ensures the rationality and interpretability of the clustering results. The termination condition parameters are set based on historical data analysis results; for example, the maximum number of clusters could be set to 20% of the total reagent quantity. Finally, all formed clusters undergo manual review and verification, such as inviting clinical experts to evaluate the rationality of the clustering results and ensure the accuracy of business logic dependency determination. During verification, an expert approval threshold of 80% is set; that is, when more than 80% of the clustering is approved by experts, the clustering results are considered valid. The entire process also includes an anomaly handling mechanism. For example, when abnormal fluctuations occur in the consumption data of a reagent, an automatic data review process is triggered to ensure the reliability of the analysis results. The criterion for abnormal fluctuations is that the daily consumption exceeds three standard deviations of the average.

[0064] When implementing the step of binding related reagents to the same delivery unit and aggregating the attribute information of the delivery unit based on the attribute information of the reagents to be delivered when business logic dependencies exist, the first step is to allocate reagents belonging to the same cluster group to the same logical container for unified delivery management, based on the clustering results obtained in the previous step. The logical container exists in the system in the form of a specific data structure, which includes attribute information such as a unique container identifier, a list of reagents within the container, and container capacity limits. The unique container identifier uses a 16-bit alphanumeric code, and the container capacity limit is calculated based on the total volume and weight of the reagents within the group, typically set to 1.2 times the total volume of the reagents within the group plus 1.1 times the total weight of the reagents within the group, to ensure sufficient storage space and a safe load-bearing range.

[0065] The creation of logical containers is completed automatically by the system. Each logical container corresponds to a cluster group. For example, a cluster group containing 5 reagents corresponds to a logical container. When creating a container, it is necessary to verify the compatibility of reagents within the cluster group, including chemical property compatibility and physical storage requirement compatibility, to avoid placing reagents that react with each other in the same logical container.

[0066] When generating aggregated attribute information for a delivery unit based on the attribute information of the reagents to be delivered, the first step is to process the earliest expiration time within the safe time window for all reagents in the group to be removed from the temperature-controlled environment. This aggregation process is achieved by comparing the safe time window parameters of each reagent in the logical container. Specifically, the safe time window values ​​recorded in the attribute information of each reagent are extracted. These values ​​are stored in minutes. For example, if the safe time window for one reagent is 30 minutes and the safe time window for another reagent is 45 minutes, then the minimum value of 30 minutes is taken as the aggregated safe time window value for the entire delivery unit through a comparison algorithm.

[0067] The comparison algorithm uses a traversal comparison method, starting from the safe time window value of the first reagent and comparing it with the safe time window values ​​of subsequent reagents in turn, always keeping the current minimum value, until the safe time window values ​​of all reagents have been traversed.

[0068] Next, the most stringent temperature requirement among all reagents in the group is processed. This polymerization process is achieved by comparing the transport temperature requirement range of each reagent within the logic container. The temperature requirement range includes two parameters: minimum temperature and maximum temperature. For example, if one reagent requires a transport temperature of 2 to 8 degrees Celsius, and another reagent requires a transport temperature of -20 to -15 degrees Celsius, then the most stringent temperature requirement range within the entire group is selected using a temperature range comparison algorithm. The temperature range comparison algorithm first compares the minimum temperature requirements of all reagents and takes the minimum value, then compares the maximum temperature requirements of all reagents and takes the maximum value. However, when the range formed by the minimum minimum temperature and the maximum maximum temperature cannot cover the temperature requirements of all reagents, the most stringent range needs to be selected and manual confirmation is required. For example, when there are two temperature requirements within the group: 2 to 8 degrees Celsius and 15 to 25 degrees Celsius, 2 to 8 degrees Celsius is selected as the most stringent range.

[0069] During the polymerization process, the system automatically detects the compatibility of temperature requirement ranges and sets temperature compatibility verification rules. When the temperature requirement ranges of reagents within a group overlap, the overlapping portion is taken as the polymerization temperature requirement. If there is no overlap, the strictest range is selected and manual confirmation is required. The entire polymerization attribute information generation process also includes a data verification mechanism. For example, when an abnormal temperature requirement for a reagent is detected, such as the upper limit of the temperature range being lower than the lower limit or the temperature value exceeding a reasonable range, the system will trigger an alert and require manual review to ensure the accuracy of the polymerization results.

[0070] The data verification mechanism includes range verification, logic verification, and consistency verification. Range verification ensures that the temperature value is within a reasonable range of -50 degrees Celsius to +50 degrees Celsius. Logic verification ensures that the upper temperature limit is not lower than the lower temperature limit. Consistency verification ensures that the temperature requirements of reagents in the same group are compatible.

[0071] The final generated aggregated attribute information of the delivery unit will be used in subsequent delivery priority calculation and route planning processes, providing a basis for decision-making throughout the entire delivery process. The aggregated attribute information of the delivery unit also includes other derived attributes, such as total volume, total weight, and urgency score. These attributes are derived through corresponding aggregation calculation rules. Total volume is calculated as the sum of the volumes of all reagents within the group, with the unit uniformly set to cubic centimeters. Total weight is calculated as the sum of the weights of all reagents within the group, with the unit uniformly set to grams. The urgency score is calculated based on factors such as the shortest expiration date and the highest priority of the reagents within the group. The calculation formula is a normalized product of the expiration date urgency coefficient and the priority coefficient. The expiration date urgency coefficient is the logarithmic function of the number of days remaining until the nearest expiration date. The priority coefficient is divided into five levels from 1 to 5 based on the clinical importance of the reagent.

[0072] The entire binding and aggregation process is automated, while also providing a manual intervention interface. This allows operators to adjust binding results or modify aggregation attribute information in special circumstances. The manual intervention interface offers a visual interface displaying the current binding results and aggregation attribute information, along with modification suggestions and conflict warnings, ensuring the system's flexibility and usability. The system also includes an operation log function, recording the content, time, and operator information of all manual intervention operations for subsequent auditing and traceability. The entire process also incorporates an anomaly handling mechanism. When data anomalies or system errors occur, contingency plans are automatically activated, including data rollback, operation prompts, and administrator notifications, ensuring the system's stability and reliability.

[0073] In implementing the step of real-time assessment of logistics resource contention and quantifying the real-time demand intensity of each delivery unit for logistics resources based on aggregated attribute information, business logic dependencies, and upstream batch and quality information, the first step involves monitoring the real-time occupancy status of each passageway and elevator using IoT sensors deployed at key locations throughout the hospital. These IoT sensors include various types such as infrared sensors, pressure sensors, and video sensors. For example, a pair of infrared sensors is deployed every 20 meters in main passageways to detect personnel passage, pressure sensors are installed inside elevator cars to monitor load status, and video sensors are installed at passageway intersections to identify congestion. Sensor data is transmitted in real-time to the central processing system via the hospital's internal wireless network, with a sampling frequency set to once per second to ensure timely capture of changes in the status of logistics resources. The central processing system analyzes and processes the sensor data, assessing the degree of contention by calculating the real-time occupancy rate of each passageway and elevator. The occupancy rate is calculated as the ratio of the current number of users or load capacity to the maximum capacity. For example, if the maximum capacity of a passageway is 10 people passing through simultaneously, and 5 people are currently detected passing through, the occupancy rate is 50%. The system classifies logistics resources into three levels based on occupancy rate: smooth, busy, and congested. The classification criteria are: occupancy rate below 30% is smooth, 30% to 70% is busy, and above 70% is congested. Different status levels correspond to different contention and conflict coefficients, which are used for subsequent demand intensity calculations.

[0074] When quantifying the real-time demand intensity of logistics resources for each delivery unit based on aggregated attribute information, business logic dependencies, and upstream batch and quality information, the system first compares the remaining time of the delivery unit's safe time window. The remaining time is calculated as the time difference between the current time and the earliest expiration time recorded in the delivery unit's aggregated attribute information, with minutes as the uniform time unit. Delivery units with shorter remaining time have higher demand intensity. The system categorizes delivery units into multiple urgency levels based on the remaining time; for example, less than 30 minutes remaining is classified as "Urgent," 30 to 60 minutes as "Emergency," 60 to 120 minutes as "Normal," and more than 120 minutes as "Low Priority." Each urgency level corresponds to a basic demand intensity score; for example, the basic score for "Urgent" is 100 points, for "Emergency" it is 80 points, for "Normal" it is 60 points, and for "Low Priority" it is 40 points. When the remaining time is similar, i.e., the difference between the remaining time of two delivery units is within a preset range (e.g., less than 10 minutes), the real-time demand intensity is determined based on the departmental priority information of the main consuming departments associated with the cluster group to which the delivery unit belongs. Departmental priority information is derived from the departmental importance rating system within the hospital management system. Typically, critical departments such as the emergency department and operating room are assigned the highest priority, general wards the medium priority, and administrative departments the low priority. Departmental priorities are divided into five levels, from 1 to 5, with higher levels corresponding to higher priority coefficients. For example, level 5 has a priority coefficient of 1.5, level 4 1.2, level 3 1.0, level 2 0.8, and level 1 0.5. The demand intensity score is obtained by multiplying the base score by the priority coefficient.

[0075] When there is a quality risk situation, such as reagents nearing their expiration date or temperature values ​​in the upstream handling process continuously exceeding the reagent transportation temperature requirement threshold for a preset duration, the real-time demand intensity level of the delivery unit will be increased accordingly. "Nearing expiration date" means the reagent's expiration date is less than 7 days away. "Quality risk situation" means that the temperature control record shows three consecutive sampling temperatures exceeding the reagent transportation temperature requirement threshold; for example, if the temperature requirement is 2 to 8 degrees Celsius, and three consecutive sampling temperatures reach 9 degrees Celsius or 1 degree Celsius. When encountering a quality risk situation, the demand intensity score is multiplied by a risk coefficient. The risk coefficient is set between 1.2 and 2.0 according to the severity of the risk; for example, a slight risk coefficient is 1.2, a moderate risk is 1.5, and a severe risk is 2.0. The risk level is determined based on the magnitude and duration of temperature deviation from the standard. For example, a temperature deviation of 1 to 2 degrees Celsius for less than 30 minutes is considered a slight risk; a deviation of 2 to 5 degrees Celsius or a duration of 30 to 60 minutes is considered a moderate risk; and a deviation exceeding 5 degrees Celsius or a duration exceeding 60 minutes is considered a severe risk. The final demand intensity score is a comprehensive score after all adjustments, used for subsequent delivery prioritization and resource allocation decisions. The entire quantification process is recalculated every 5 minutes to ensure timely reflection of the latest status of logistics resources and delivery demand. The system also has an anomaly handling mechanism; when sensor data anomalies or calculation errors occur, the most recent valid data is automatically used as a replacement, and a maintenance alarm is issued to ensure stable system operation. All calculation processes and results are recorded in the system log, including timestamps, calculation parameters, intermediate results, and final scores, for subsequent auditing and analysis. This multi-factor comprehensive evaluation method can accurately quantify the real-time demand intensity of logistics resources for each delivery unit, providing a scientific and reasonable decision-making basis for hospital reagent delivery.

[0076] Figure 2A flowchart illustrating the generation of delivery priorities for each delivery unit according to this invention is provided. In the step of constructing a multi-agent game relationship between each delivery unit and generating delivery priorities by solving a game equilibrium allocation scheme based on real-time demand intensity and logistics resource contention conflicts, each delivery unit is first treated as an independent agent, each with independent decision-making capabilities and objective functions. The agent initialization process includes assigning a unique identifier, current state information, and decision parameters to each delivery unit agent. The unique identifier is the delivery unit number; the current state information includes the real-time demand intensity score, current location, and target location; and the decision parameters include the strategy space and payoff function parameters. Each agent's strategy space is defined as selectable logistics resource usage schemes, such as choosing different travel routes or elevator usage time periods. The size of the strategy space is determined based on the amount of available logistics resources and typically contains 3 to 5 selectable strategies. The game relationship between agents is established through shared logistics resource constraints. All agents compete for limited channel and elevator resources, and each agent's strategy choice affects the available resources and payoffs of other agents.

[0077] When generating delivery priorities for each delivery unit based on the game equilibrium allocation scheme derived from real-time demand intensity and logistics resource contention conflicts, the real-time demand intensity of each delivery unit for logistics resources serves as the basis for the utility function of each agent. The utility function is designed using a linear weighted form, with the real-time demand intensity score as the primary weighting factor, while also considering factors such as logistics resource utilization efficiency and waiting time. Logistics resource contention conflicts are used as resource allocation constraints, including maximum capacity limits for each passageway and elevator, minimum passage interval requirements, etc. These constraints constitute the feasible region of the game problem. The Nash equilibrium point is solved through a non-cooperative game process among agents. A Nash equilibrium point is a strategy combination where no agent can improve their own payoff by individually changing their strategy. The delivery priority of each delivery unit is determined based on the strategy payoff of each agent at the equilibrium point; agents with higher strategy payoffs have higher priority delivery units. Priorities are divided into multiple levels; for example, the top 20% are assigned the highest priority, the middle 60% the medium priority, and the bottom 20% the lowest priority, ranked from highest to lowest.

[0078] The utility function Ui can be expressed as: Ui=Si×ηi-α×Twait; where Ui is the utility value of agent i; Si is the real-time demand intensity score; ηi is the resource utilization efficiency coefficient, which is obtained from historical data statistics and has a value range of ηi∈[0.8,1.2]; Twait is the expected waiting time (unit: minutes); α is the waiting time penalty coefficient, with a value of α=0.1.

[0079] The Nash equilibrium is solved through a non-cooperative game process among agents. Each agent proposes a resource utilization strategy based on its real-time demand intensity. Each agent selects an initial strategy based on its current real-time demand intensity score and the effects of its historical strategies, using a greedy algorithm to choose the currently optimal strategy. Through multiple rounds of iterative strategy adjustment and payoff calculation, a dynamic optimal response method is employed. In each iteration, each agent selects the strategy that maximizes its own payoff based on the current strategies of other agents. Payoff calculation includes direct and indirect payoffs. Direct payoffs come from resource utilization efficiency, while indirect payoffs come from the synergistic effects of strategies; for example, two agents choosing to use the same resource at off-peak times can achieve synergistic payoffs. The convergence condition for the iterative process is that the strategies of all agents remain unchanged for three consecutive rounds, or the maximum number of iterations (100 rounds) is reached. The process ultimately converges to a stable Nash equilibrium state where each agent's strategy is the best response to the strategies of other agents. The entire game-solving process uses a distributed computing architecture. Each agent independently selects its strategy and calculates its payoff, exchanging strategy information through a message passing mechanism. A central coordinator monitors the convergence state and summarizes the final results. The system also incorporates an anomaly handling mechanism. When convergence fails or the equilibrium point is not unique, an arbitration procedure is initiated. Priority is determined based on real-time demand intensity scores, ensuring a definite delivery priority sequence is generated under all circumstances. The final delivery priority will guide subsequent route planning and resource allocation decisions, providing the optimal scheduling scheme for hospital reagent delivery. All game processes and results are recorded in the system log, including strategy selection, payoff calculations, and convergence status for subsequent analysis and optimization. This multi-agent game theory approach fully considers the demand characteristics and resource competition relationships of each delivery unit, generating a fair and efficient delivery priority scheme.

[0080] When implementing the step of planning specific travel routes and time schedules for each delivery unit in the initial path set based on delivery priority to generate delivery instructions, the final delivery order of each delivery unit is first determined according to the delivery priority calculated in the previous step. The delivery priority value is usually set between 0 and 100, with higher values ​​indicating higher priority. The system arranges delivery units in descending order of priority value to form the final delivery order queue. When multiple delivery units have the same priority value, a secondary sort is performed based on the remaining time of their safety time window, with delivery units with shorter remaining time being delivered first. For example, if one delivery unit has a priority of 95 points and another has a priority of 90 points, the delivery unit with the 95-point priority is delivered first; if two delivery units both have a priority of 90 points, but one has 25 minutes of remaining time and the other has 30 minutes, the delivery unit with 25 minutes of remaining time is delivered first. When determining the delivery order, the system also considers the business logic dependencies between delivery units. For delivery units belonging to the same cluster group, delivery times are arranged as close as possible to reduce multiple deliveries to the same department.

[0081] When planning specific routes and time schedules for the final delivery order based on the initial route set, the system selects the optimal route for each delivery unit from the initial route set. The selection of the optimal route comprehensively considers factors such as route length, estimated travel time, and current congestion level, using a weighted scoring method for evaluation. The weights for route length, estimated travel time, and current congestion level are all set at 0.4 and 0.3 respectively. For example, if a route is 100 meters long, has an estimated travel time of 2 minutes, and experiences light congestion (corresponding to a score of 2 points), its comprehensive score is 100 × 0.4 + 2 × 0.3 + 2 × 0.3 = 40 + 0.6 + 0.6 = 41.2 points. The system selects the route with the highest comprehensive score as the delivery unit's route. Time scheduling is based on delivery priority; higher-priority delivery units receive earlier delivery time windows, and sufficient safety intervals are reserved between adjacent delivery units to avoid resource conflicts. The time interval should be set based on the route length and estimated travel time, typically 1.2 times the estimated travel time. For example, if the estimated travel time is 10 minutes, the time interval should be set to 12 minutes. For delivery units requiring special temperature control conditions, the temperature maintenance requirements should be considered additionally when scheduling. For example, for delivery units requiring a temperature of 2–8 degrees Celsius, the delivery time should be appropriately advanced during high-temperature periods.

[0082] Post-execution feedback to the supply chain monitoring system refers to uploading real-time status information and final actual delivery path information during the delivery task execution process to the supply chain monitoring system. Real-time status information includes data such as the current location of the delivery unit, its speed, ambient temperature, and remaining time. This data is collected through IoT devices carried by delivery vehicles or personnel, including GPS locators, temperature sensors, and speedometers, with a sampling frequency of once every 30 seconds. The final actual delivery path information records the sequence of path nodes actually traversed by the delivery unit, the arrival time of each node, and the duration of each stop. Data uploads use encrypted transmission protocols to ensure data transmission security. The upload frequency is dynamically adjusted according to network conditions, typically every 5 minutes when the network is good and every 15 minutes when the network is poor. After receiving the data, the supply chain monitoring system performs integrity checks and logical verifications, such as checking the continuity of timestamps and the reachability of path nodes. When abnormal data is detected, an early warning mechanism is triggered, requiring on-site personnel to confirm and correct the data. All uploaded data is used to establish a complete delivery traceability file in the monitoring system, including comparative analysis of planned and actual routes, time deviation statistics, and abnormal event records, providing data support for subsequent delivery optimization. The system also features a data backup mechanism, regularly backing up delivery data to an off-site storage center once a day to ensure data security and recoverability. This comprehensive feedback mechanism enables visualized monitoring and traceability of the entire delivery process, improving the reliability and transparency of delivery services. Feedback data is also used to optimize subsequent route planning algorithms, such as adjusting the calculation parameters for estimated travel time based on actual travel time and updating the congestion assessment model based on actual congestion conditions, continuously improving the accuracy and efficiency of delivery planning. The entire feedback process forms a closed-loop optimization mechanism, enabling the delivery system to adapt to environmental changes and demand fluctuations, continuously improving the quality of delivery services.

[0083] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0084] It should be noted that this invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0085] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wireless or wired route; wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; wireless transmission includes infrared, microwave, etc. Computer-readable storage media can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0086] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0087] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0088] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0089] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0090] If a function is implemented as a software module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0091] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0092] In conclusion, the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for integrated management of monitoring information across the entire in vitro diagnostic reagent supply chain, characterized in that: include: S1. Obtain the attribute information, upstream batch and quality information, and hospital environment structure information of the reagents to be delivered; S2. Based on the hospital's environmental structure information, establish an initial path set covering all accessible paths within the hospital; S3. Based on time-series similarity clustering analysis, analyze the correlation of consumption fluctuation patterns in the historical consumption data of each reagent to determine whether there is a business logic dependency. S4. When there is a business logic dependency, the associated reagents are bound to the same delivery unit, and the aggregated attribute information of the delivery unit is generated based on the attribute information of the reagents to be delivered. S5. Real-time assessment of logistics resource contention conflicts, and quantification of the real-time demand intensity of each delivery unit for logistics resources based on aggregated attribute information, business logic dependencies, and upstream batch and quality information. S6. Construct each delivery unit into a multi-agent game relationship, and solve the game equilibrium allocation scheme based on real-time demand intensity and logistics resource competition conflict to generate the delivery priority of each delivery unit. S7. Based on delivery priority, plan specific travel routes and time arrangements for each delivery unit in the initial path set to generate delivery instructions, and after execution, feed them back to the supply chain full-chain monitoring system.

2. The method for integrated management of full-chain monitoring information of in vitro diagnostic reagent supply chain according to claim 1, characterized in that, Obtain the attribute information of the reagents to be delivered, including the safe time window during which the reagents can be removed from the temperature-controlled environment and the temperature requirements for reagent transportation; Obtain upstream batch and quality information, including reagent production batch number, expiration date, and temperature control records during upstream distribution. Obtaining hospital environmental structure information includes hospital building distribution maps, department location maps, and information on the location of passageways and elevators.

3. The method for integrated management of full-chain monitoring information of in vitro diagnostic reagent supply chain according to claim 1, characterized in that, The initial path set is established based on the hospital's environmental structure information, including: constructing a hospital indoor and outdoor space connectivity map based on the hospital building distribution map, department location map, and corridor and elevator location information; traversing all nodes corresponding to the department location information in the map, calculating the shortest path between nodes, and forming the initial path set.

4. The method for integrated management of the entire supply chain monitoring information of in vitro diagnostic reagents according to claim 1, characterized in that, The correlation of consumption fluctuation patterns of historical consumption data of each reagent based on time-series similarity clustering analysis includes: normalizing the historical consumption time-series data of each reagent, calculating the morphological similarity between different historical consumption time-series data of reagents, classifying reagents with morphological similarity exceeding a preset threshold into the same cluster group, and labeling the main consuming departments associated with each cluster group. Determining whether a business logic dependency exists includes: if reagents are grouped into the same cluster, then it is determined that there is a business logic dependency between these reagents.

5. The method for integrated management of full-chain monitoring information of in vitro diagnostic reagent supply chain according to claim 4, characterized in that, Calculating the morphological similarity between time series data of historical consumption of different reagents includes: normalizing the time series data of historical consumption of each reagent to eliminate dimensional differences, and quantifying the morphological similarity by calculating the matching degree of shape and fluctuation trend between sequences.

6. The method for integrated management of full-chain monitoring information of in vitro diagnostic reagent supply chain according to claim 1, characterized in that, Binding related reagents to the same delivery unit means allocating reagents belonging to the same cluster group to the same logical container for unified delivery management; The aggregated attribute information of the delivery unit is generated based on the attribute information of the reagents to be delivered. This includes the earliest expiration time within the safe time window during which all reagents in the group are allowed to leave the temperature-controlled environment, and the most stringent temperature requirement among the reagent transportation temperature requirements for all reagents in the group.

7. The method for integrated management of full-chain monitoring information of in vitro diagnostic reagent supply chain according to claim 1, characterized in that, Real-time assessment of logistics resource contention conflicts refers to monitoring the real-time occupancy status of each passage and elevator in hospital access channels and elevator locations using IoT sensors. The real-time demand intensity of each delivery unit for logistics resources is quantified based on aggregated attribute information, business logic dependencies, and upstream batch and quality information. This includes: firstly, comparing the remaining duration of the delivery unit's safe time window; the shorter the remaining duration, the higher the demand intensity of the delivery unit. When the remaining time is similar, the real-time demand intensity is determined based on the departmental priority information of the main consuming departments associated with the cluster group to which the delivery unit belongs; when there is a quality risk situation where the reagent expiration date is approaching or the temperature value in the temperature control record of the upstream circulation process continuously exceeds the reagent transportation temperature requirement threshold for a preset time, the real-time demand intensity level of the delivery unit is increased accordingly.

8. The method for integrated management of full-chain monitoring information of in vitro diagnostic reagent supply chain according to claim 1, characterized in that, Constructing each delivery unit as a multi-agent game relationship means treating each delivery unit as an independent agent; The process of generating delivery priorities for each delivery unit based on the real-time demand intensity and logistics resource contention conflict to solve the game equilibrium allocation scheme includes: using the real-time demand intensity of each delivery unit for logistics resources as the basis of the utility function of each agent, using the logistics resource contention conflict as the resource allocation constraint, solving the Nash equilibrium point through a non-cooperative game process among agents, and determining the delivery priority of each delivery unit based on the strategy payoff of each agent at the equilibrium point.

9. The method for integrated management of full-chain monitoring information of in vitro diagnostic reagent supply chain according to claim 8, characterized in that, Solving for the Nash equilibrium through a non-cooperative game process among agents involves each agent proposing a resource usage strategy based on its own real-time demand intensity. Through multiple rounds of strategy adjustment and payoff calculation, the process eventually converges to a Nash equilibrium state where all agents' strategies are stable.

10. The method for integrated management of full-chain monitoring information of in vitro diagnostic reagent supply chain according to claim 1, characterized in that, Generating delivery instructions based on delivery priority and initial path set includes: determining the final delivery order of each delivery unit according to delivery priority, and planning specific travel routes and time arrangements for the final delivery order based on the initial path set; Feedback to the supply chain monitoring system after execution refers to uploading real-time status information during the delivery task execution process and the final actual delivery route information to the supply chain monitoring system.

Images