Data-driven drug supply chain tracing management method

Abstract

The invention discloses a data-driven drug supply chain tracing management method, and relates to the technical field of data processing. The method comprises the following steps: calling a drug supply chain for data collection to obtain a multi-link data set, and constructing a block chain network based on the multi-link data set; constructing a decision tree model, traversing the block chain network to trigger the decision tree model to carry out prediction, and generating multiple pieces of drug prediction data; setting N tracing targets according to the plurality of drug prediction data, wherein N is an integer greater than 0; traversing the block chain network according to the N traceability targets to carry out rapid positioning, and determining a plurality of target links; and performing intelligent tracing on the medicine supply chain based on the plurality of target links. The technical problems of difficult data tracing, low transparency and low tracing efficiency in the drug supply chain in the prior art are solved, and the technical effect of improving the transparency, accuracy and efficiency of drug supply chain tracing is achieved through the block chain technology.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically to a data-driven method for pharmaceutical supply chain traceability management. Background Technology

[0002] In today's globalized pharmaceutical market, the management and traceability of the drug supply chain face unprecedented challenges. With consumers increasingly demanding higher drug safety and quality, and regulatory agencies strengthening their oversight of the entire drug lifecycle, traditional drug traceability methods are no longer sufficient to meet market needs. Traditional traceability systems often rely on centralized databases, which suffer from problems such as data tampering, low traceability efficiency, and insufficient transparency. These issues directly impact the healthy development of the pharmaceutical market and consumer trust. Summary of the Invention

[0003] This application provides a data-driven method for pharmaceutical supply chain traceability management, which solves the technical problems of difficult data traceability, low transparency, and low traceability efficiency in the pharmaceutical supply chain in the prior art.

[0004] This application provides a data-driven method for pharmaceutical supply chain traceability management, the method comprising: Data is collected from the pharmaceutical supply chain to obtain multi-stage datasets, and a blockchain network is constructed based on these datasets. A decision tree model is built, and the blockchain network is traversed to trigger predictions, generating multiple pharmaceutical prediction data. N traceability targets are set according to the multiple pharmaceutical prediction data, where N is an integer greater than 0. The blockchain network is traversed according to the N traceability targets to quickly locate and determine multiple target stages. Intelligent traceability of the pharmaceutical supply chain is then performed based on the multiple target stages.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, data is collected from the pharmaceutical supply chain to obtain multi-stage datasets, and a blockchain network is constructed based on these datasets. Next, a decision tree model is built, and traversing the blockchain network triggers the decision tree model to make predictions, generating multiple predicted drug data sets. Then, N traceability targets are set according to the multiple predicted drug data sets, where N is a positive integer. Next, the blockchain network is traversed based on the N traceability targets for rapid location, identifying multiple target stages. Finally, intelligent traceability of the pharmaceutical supply chain is performed based on these multiple target stages. This solves the technical problems of difficult data traceability, low transparency, and low traceability efficiency in the pharmaceutical supply chain in existing technologies. Through blockchain technology, it achieves the technical effect of improving the transparency, accuracy, and efficiency of pharmaceutical supply chain traceability. Attached Figure Description

[0006] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0007] Figure 1 A schematic diagram of the data-driven drug supply chain traceability management method provided in this application embodiment; Figure 2 This is a schematic diagram illustrating the process of generating multiple drug prediction data in the data-driven drug supply chain traceability management method provided in the embodiments of this application. Detailed Implementation

[0008] This application addresses the technical problems of difficulty in data traceability, low transparency, and low traceability efficiency in the pharmaceutical supply chain by providing a data-driven traceability management method.

[0009] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0010] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0011] Examples, such as Figure 1 As shown, this application provides a data-driven method for pharmaceutical supply chain traceability management, wherein the method includes: Data is collected from the pharmaceutical supply chain to obtain multi-stage datasets, and a blockchain network is built based on these datasets.

[0012] The pharmaceutical supply chain involves multiple stages, including but not limited to raw material procurement, production and processing, quality inspection, packaging and labeling, warehousing and logistics, sales and distribution, and final consumption. Data collection from the pharmaceutical supply chain creates a multi-stage dataset rich in information. A blockchain network is then built upon this multi-stage data. This involves using a blockchain framework and protocol to encrypt and store data from each stage of the dataset in blocks, and using technologies such as hash pointers to connect these blocks in a chain, thus constructing a complete blockchain network. In this blockchain network, each node can participate in data verification and storage, ensuring the authenticity and reliability of the data. Furthermore, due to the immutable nature of blockchain, once data is recorded on the chain, it cannot be easily modified or deleted, providing strong evidentiary support for subsequent traceability. In addition, the blockchain network offers high transparency, allowing all parties in the supply chain to view the latest data status in real time, enhancing the overall collaboration and efficiency of the supply chain.

[0013] Furthermore, the method for retrieving data from the pharmaceutical supply chain to obtain multi-stage datasets includes: Based on the drug supply chain, multiple data sources are identified, and data is collected from these multiple data sources to generate a drug supply dataset. The drug supply dataset is then standardized and matched according to multiple links in the drug supply chain to generate a multi-link dataset.

[0014] The pharmaceutical supply chain process involves multiple data sources, including but not limited to the ERP systems of manufacturers, the WMS systems of logistics companies, the CRM systems of sales channels, and the databases of government regulatory departments. Each data source carries key information from a specific link in the pharmaceutical supply chain, collectively forming a complete data ecosystem for the pharmaceutical supply chain. Data is collected from multiple data sources, including but not limited to drug batch numbers, production dates, expiration dates, quality inspection reports, logistics information, and sales records. The collected data constitutes a pharmaceutical supply dataset. The pharmaceutical supply dataset undergoes standardization to eliminate inconsistencies and redundancy between data, ensuring data accuracy and comparability. This is achieved by defining unified data standards, converting data formats, and standardizing data naming, transforming the pharmaceutical supply dataset into a clearly structured and easily understood standardized dataset. Data is then matched and categorized according to multiple links in the pharmaceutical supply chain, which typically includes raw material procurement, production and processing, quality inspection, packaging and warehousing, logistics and distribution, sales and distribution, and after-sales service. The standardized data is divided and matched according to these links, forming a multi-link dataset, which contains data information from multiple links.

[0015] A decision tree model is constructed, and the blockchain network is traversed to trigger the decision tree model to make predictions, generating multiple drug prediction data.

[0016] Decision tree models can be constructed by using collected multi-stage datasets as training data, selecting appropriate decision tree algorithms such as CART, ID3, and C4.5, and training the model with the training data. Once trained, the decision tree model is obtained. The trained model is then deployed to a blockchain network, enabling it to receive supply chain data and make predictions. Within the blockchain network, whenever new supply chain data is updated, the decision tree model is triggered to make predictions, generating multiple drug prediction datasets.

[0017] Furthermore, such as Figure 2 As shown, traversing the blockchain network triggers the decision tree model to make predictions, generating multiple drug prediction data. The method includes: The process involves traversing the blockchain network to extract drug supply trend information, drug supply type information, and drug supply environment information. Based on the drug supply environment information, the drug supply trend information is dynamically collected according to the drug supply type information to generate a critical value for drug supply change. Trigger constraints are constructed, and it is determined whether the critical value for drug supply change meets the trigger constraints. If the critical value for drug supply change meets the trigger constraints, a trigger command is generated to activate the decision tree model for prediction, generating multiple drug prediction data.

[0018] The system extracts drug supply trend information, drug supply type information, and drug supply environment information from the blockchain network. Drug supply trend information includes historical transaction data, inventory changes, and sales records. Drug supply type information includes drug types and specifications. Drug supply environment information includes supply chain participants (such as manufacturers, distributors, and retailers), logistics conditions, and policy environment. Based on the drug supply environment information, the system dynamically collects drug supply trend information according to drug supply type information to generate drug supply change thresholds. This involves calculating the supply change threshold for each drug type based on its supply change patterns under different environments; when the supply reaches or exceeds this threshold, supply chain adjustments may be triggered. Constraints are defined for the triggering decision tree model based on business needs and supply chain management strategies. The system assesses the match between the drug supply change threshold and the triggering constraints. If the threshold satisfies the triggering constraints, a trigger command is generated. Upon receiving a trigger command, the pre-trained decision tree model is activated, and the current drug supply environment information, supply type information, and the latest supply data are input to perform predictions, generating multiple drug prediction data sets.

[0019] Furthermore, the method for triggering the constraint conditions includes: The drug supply trend information, drug supply type information, and drug supply environment information are used to assign weights to obtain multiple weight coefficients; the blockchain network is updated according to the multiple weight coefficients to generate an associated connection list; the associated connection list is marked with edge data according to a preset last-digit ratio threshold to obtain the triggering constraint condition.

[0020] The importance of drug supply trend information, drug supply type information, and drug supply environment information is determined through expert evaluation, and a weight coefficient is assigned to each piece of information. These weight coefficients reflect the degree of influence of the feature on drug supply chain management decisions. In the blockchain network, each node may represent a link in the supply chain. Based on the calculated weight coefficients, the connection strength or trust level between nodes in the blockchain network is updated. Based on the weight updates, a list of associated connections is generated, which shows the relative importance of each node. A bottom-ranking proportion threshold is set to identify marginal data or low-weight connections in the associated connection list. Based on the bottom-ranking proportion threshold, the associated connection list is filtered to identify connections with weights below the threshold. These marginal connections are marked as potential risk points or areas requiring special attention. Based on the results of the marginal data identification, constraints that trigger the decision tree model are defined.

[0021] Based on the aforementioned multiple drug prediction data, N traceability targets are set, where N is an integer greater than 0.

[0022] Analyze multiple drug forecast data, and based on the analysis results, determine N key indicators as traceability targets, such as the demand for specific drugs, inventory levels, and supply chain disruption risks, where N is an integer greater than 0.

[0023] Based on the N traceability targets, the blockchain network is traversed to quickly locate and determine multiple target links.

[0024] Based on N traceability targets, by traversing the blockchain network, multiple target links directly related to the traceability targets are identified, including different stages in the supply chain such as production, processing, packaging, transportation, distribution, and sales.

[0025] Intelligent traceability of the pharmaceutical supply chain is performed based on the aforementioned multiple target links.

[0026] Based on these identified target links, intelligent traceability of the pharmaceutical supply chain can be achieved through mechanisms such as smart contracts and timestamps in the blockchain network, enabling transparent and traceable management of the entire life cycle of pharmaceuticals. The entire traceability process is efficient and accurate, providing strong protection for the quality and safety of pharmaceuticals.

[0027] Furthermore, the method for intelligent traceability of the pharmaceutical supply chain based on the aforementioned multiple target links includes: Using the multiple target links as index data, the drug supply chain is traversed for anomaly matching to generate multiple abnormal target links; multiple drug anomaly data are extracted based on the multiple abnormal target links; an anomaly level sequence is introduced, and the multiple drug anomaly data are labeled according to the anomaly level sequence to generate multiple drug anomaly levels, with a correspondence between the multiple drug anomaly data and the multiple drug anomaly levels; early warning instructions are generated according to the multiple drug anomaly levels, and the drug supply chain is traced for anomaly early warning response based on the early warning instructions.

[0028] Preferably, each target link in the pharmaceutical supply chain is used as the basis for index data, and each target link is associated with a unique identifier (such as ID, hash value, etc.) to facilitate subsequent data retrieval and matching. Anomaly matching is performed by traversing the pharmaceutical supply chain based on the index data, that is, matching detected abnormal data with the previously defined target link index to determine which target links have anomalies, generating multiple abnormal target links. Specific pharmaceutical anomaly data is extracted from the abnormal target links, including drug batch, production date, production location, anomaly type, and anomaly description. Based on the anomaly level sequence and combining multiple pharmaceutical anomaly data, each anomaly data is labeled and assigned a corresponding anomaly level, ensuring a clear correspondence between each pharmaceutical anomaly data and its corresponding anomaly level. Early warning rules are formulated based on the anomaly level and business needs, clarifying the early warning measures and response procedures to be taken under different anomaly levels. Early warning instructions are automatically generated based on the early warning rules and labeled anomaly levels, including early warning content, early warning level, recipients, and response time. Based on the early warning instructions, corresponding traceability anomaly early warning response actions are initiated, including problem investigation, risk assessment, emergency handling, and improvement measures.

[0029] Furthermore, using the multiple target links as index data, the drug supply chain is traversed for anomaly matching to generate multiple abnormal links. The method includes: An anomaly matching channel is established, and the index data is synchronized to the anomaly matching channel to identify anomalies in the drug supply chain. It is determined whether the index data meets a preset rule threshold. If the index data does not meet the preset rule threshold, an anomaly report is generated and encrypted to obtain an anomaly encrypted report. Multiple anomaly links corresponding to the index data are extracted, and the anomaly encrypted report is added to the multiple anomaly links.

[0030] Preferably, an anomaly matching channel is constructed to enable real-time monitoring of index data. This channel includes a data receiving module, a processing module, a rule matching module, and a result output module. Based on business needs and supply chain characteristics, a series of preset rule thresholds are defined to determine whether the index data is abnormal. The rule matching module compares the index data with these preset rule thresholds to determine if the data conforms to the rules. If the index data does not conform to any preset rule threshold, it is considered abnormal. For identified anomalies, a detailed anomaly report is generated, including the anomaly time, anomaly type, anomaly data, and scope of impact. Simultaneously, the anomaly report is encrypted using encryption algorithms (such as AES and RSA) to protect sensitive information from unauthorized access. The encrypted anomaly report is stored in a secure location, such as an encrypted database or a private area on the blockchain. Based on the index data, multiple associated anomaly links are extracted, and the encrypted anomaly report is associated with each extracted anomaly link to ensure that each anomaly link has a corresponding anomaly report as support.

[0031] Furthermore, the method for the anomaly level sequence includes: Based on the drug supply trend information, combined with the drug supply type information and the drug supply environment information, a risk assessment is conducted to generate drug anomaly standards; based on the drug anomaly standards, anomaly analysis is performed to define multiple anomaly levels; the multiple anomaly levels are arranged in descending order to generate the anomaly level sequence.

[0032] Preferably, risk assessment is conducted based on drug supply trend information, combined with drug supply type information and drug supply environment information, to identify risk factors that may affect the stability and safety of drug supply; based on the risk assessment results, the conditions under which drug supply is considered abnormal are defined, and the criteria for drug abnormality may include a sudden drop in supply, failure to meet quality standards, and supply interruption time exceeding a specific threshold; anomaly analysis is performed on the criteria for drug abnormality, that is, according to the severity of the abnormality, the abnormality is divided into multiple levels, such as mild, moderate, severe, etc., and the abnormality levels are arranged from high to low severity to generate an abnormality level sequence.

[0033] Furthermore, the method of generating early warning instructions according to the multiple drug anomaly levels, and responding to the drug supply chain anomaly based on the early warning instructions, includes: An impact analysis is performed based on the multiple drug anomaly levels to generate anomaly impact factors. When the anomaly impact factors reach a preset impact threshold, a traceability path is generated and added to the early warning command. The entire drug supply chain is traced according to the traceability path to generate a traceability trajectory. The anomaly early warning response module is activated based on the traceability trajectory for verification and processing.

[0034] Preferably, based on the drug anomaly level, statistical analysis and machine learning methods are used to assess the potential impact of the anomaly on the supply chain, identify and quantify the impact of the anomaly on each link of the supply chain (such as production, transportation, and sales), and generate anomaly impact factors. Based on corporate policies, industry standards, and historical experience, preset thresholds are set for different types of anomaly impact factors. The actual calculated anomaly impact factors are compared with the preset thresholds to determine whether they have reached or exceeded the thresholds. When an anomaly impact factor reaches the preset threshold, a detailed traceability path is planned based on the link where the anomaly occurred, the drug batches involved, and supplier information. The traceability path information is added to the early warning command to ensure that traceability can be carried out along the correct path during the response process. Along the traceability path, relevant data is collected from each link of the supply chain, including production records, inspection reports, transportation information, and inventory status, forming a complete traceability trajectory that shows the entire process of the drug from production to use. Based on the traceability trajectory, the anomaly early warning response module is activated, initiating corresponding verification and processing procedures. Key nodes in the traceability trajectory are verified to confirm the specific cause, scope of impact, and possible consequences of the anomaly. Based on the verification results, corresponding handling measures are formulated and implemented, such as recalling problematic drugs, repairing supply chain vulnerabilities, and strengthening supervision.

[0035] In summary, the embodiments of this application have at least the following technical effects: First, data is collected from the pharmaceutical supply chain to obtain multi-stage datasets, and a blockchain network is constructed based on these datasets. Next, a decision tree model is built, and traversing the blockchain network triggers the decision tree model to make predictions, generating multiple predicted drug data sets. Then, N traceability targets are set according to the multiple predicted drug data sets, where N is a positive integer. Next, the blockchain network is traversed based on the N traceability targets for rapid location, identifying multiple target stages. Finally, intelligent traceability of the pharmaceutical supply chain is performed based on these multiple target stages. This solves the technical problems of difficult data traceability, low transparency, and low traceability efficiency in the pharmaceutical supply chain in existing technologies. Through blockchain technology, it achieves the technical effect of improving the transparency, accuracy, and efficiency of pharmaceutical supply chain traceability.

[0036] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0037] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0038] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A data-driven method for pharmaceutical supply chain traceability management, characterized in that, The method includes: Data is collected from the pharmaceutical supply chain to obtain multi-stage datasets, and a blockchain network is constructed based on these multi-stage datasets. A decision tree model is constructed, and the blockchain network is traversed to trigger the decision tree model to make predictions, generating multiple drug prediction data. N traceability targets are set based on the multiple drug prediction data, where N is an integer greater than 0; Based on the N traceability targets, the blockchain network is traversed to quickly locate and determine multiple target links; Intelligent traceability of the pharmaceutical supply chain is performed based on the aforementioned multiple target links.

2. The data-driven pharmaceutical supply chain traceability management method as described in claim 1, characterized in that, The method for retrieving data from the pharmaceutical supply chain to obtain multi-stage datasets includes: Based on the drug supply chain, multiple data sources are identified, and data is collected from these multiple data sources to generate a drug supply dataset. The drug supply dataset is standardized and matched according to multiple links in the drug supply chain to generate the multi-link dataset.

3. The data-driven pharmaceutical supply chain traceability management method as described in claim 1, characterized in that, Traversing the blockchain network to trigger the decision tree model for prediction, generating multiple drug prediction data, the method includes: The blockchain network is traversed to extract drug supply trend information, drug supply type information, and drug supply environment information. Based on the drug supply environment information, the drug supply trend information is dynamically collected according to the drug supply type information to generate drug supply change thresholds; Construct trigger constraints, and determine whether the critical value of the change in drug supply meets the trigger constraints based on the trigger constraints. If the critical value of the change in drug supply meets the triggering constraint, a triggering instruction is generated, and the decision tree model is activated by the triggering instruction to make predictions and generate the multiple drug prediction data.

4. The data-driven pharmaceutical supply chain traceability management method as described in claim 3, characterized in that, The method for triggering the constraint conditions includes: Multiple weight coefficients are obtained by weighting the drug supply trend information, the drug supply type information, and the drug supply environment information. The blockchain network is updated according to the multiple weighting coefficients to generate a list of associated connections. The associated connection list is marked with edge data according to a preset last-digit ratio threshold to obtain the triggering constraint condition.

5. The data-driven pharmaceutical supply chain traceability management method as described in claim 3, characterized in that, The method for intelligent traceability of the pharmaceutical supply chain based on the aforementioned multiple target links includes: Using the multiple target links as index data, the drug supply chain is traversed to perform anomaly matching, generating multiple abnormal target links; Extract multiple abnormal drug data based on the aforementioned multiple abnormal target links; An anomaly level sequence is introduced, and multiple drug anomaly levels are generated by labeling the multiple drug anomaly data in combination with the anomaly level sequence. There is a corresponding relationship between the multiple drug anomaly data and the multiple drug anomaly levels. Early warning instructions are generated according to the multiple drug anomaly levels, and the drug supply chain is traced and anomaly warning response is performed based on the early warning instructions.

6. The data-driven pharmaceutical supply chain traceability management method as described in claim 5, characterized in that, Using the multiple target stages as index data, the drug supply chain is traversed for anomaly matching to generate multiple abnormal stages. The method includes: An anomaly matching channel is constructed, and the index data is synchronized to the anomaly matching channel to identify anomalies in the drug supply chain and determine whether the index data meets the preset rule threshold. If the index data does not meet the preset rule threshold, an anomaly report is generated and encrypted to obtain an anomaly encrypted report; Extract multiple abnormal links corresponding to the index data, and add the abnormal encrypted report to the multiple abnormal links.

7. The data-driven pharmaceutical supply chain traceability management method as described in claim 5, characterized in that, The abnormality level sequence, the method includes: Based on the drug supply trend information, combined with the drug supply type information and the drug supply environment information, a risk assessment is conducted to generate drug anomaly criteria. Anomaly analysis is performed based on the aforementioned drug anomaly criteria to classify multiple anomaly levels; The multiple anomaly levels are arranged in descending order to generate the anomaly level sequence.

8. The data-driven pharmaceutical supply chain traceability management method as described in claim 5, characterized in that, The method includes generating early warning instructions based on the multiple drug anomaly levels, and performing a traceability anomaly early warning response on the drug supply chain based on the early warning instructions. An impact analysis was performed on the multiple drug anomaly levels to generate anomaly impact factors. When the abnormal impact factor reaches the preset impact threshold, a tracing path is generated and the tracing path is added to the early warning instruction; The entire drug supply chain is traced according to the traceability path to generate a traceability trajectory. Based on the traceability trajectory, the abnormal early warning response module is activated for verification and processing.

Images