CN-122022784-A - Circulating package sharing dispatching recovery system and method based on big data

CN122022784ACN 122022784 ACN122022784 ACN 122022784ACN-122022784-A

Abstract

The invention discloses a circulating package sharing dispatching recovery system and method based on big data, and relates to the technical field of data analysis, wherein the method comprises the following steps of obtaining circulation state data and inventory change data of a circulating packing box, and checking the data through a block chain; the method comprises the steps of carrying out hash processing on key circulation events, carrying out hierarchical storage according to data access frequency and evidence storage requirements, constructing a full life cycle tracing chain based on circulation event hash association, extracting historical circulation data of each node from a blockchain, constructing a characteristic data set, identifying a requirement fluctuation mode to construct a characteristic early warning information set, monitoring the node circulation data in real time, matching an abnormal mode, calculating comprehensive scheduling urgent coefficients of corresponding nodes, screening high-priority scheduling nodes, inquiring blockchain idle empty boxes according to the high-priority scheduling nodes, generating a scheduling list uplink, and generating a push recycling list when the recycling node comparison information reaches the standard.

Inventors

DU SONGLIN
ZHANG LIYANG
LEI TAO
Sadamu Shadik
WANG BINGQUAN
WANG XIAOFENG
YU JIONG
DU XUSHENG

Assignees

杭州骋风而来数字科技有限公司
新疆丝路融创网络科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260414

Claims (9)

1. A big data-based cyclic packaging sharing scheduling recovery method is characterized by comprising the following steps: Acquiring circulation state data and inventory change data of a circulation packing box, and checking the data through a block chain; performing key circulation event hash processing on the collected circulation state data and inventory change data through a block chain, performing hierarchical storage according to data access frequency and evidence storage requirements, and constructing a full life cycle traceability chain based on circulation event hash association; extracting historical circulation data of each node from the block chain, and constructing a characteristic data set; identifying a demand fluctuation mode based on the characteristic data set, and constructing a characteristic early-warning information set; Monitoring node circulation data in real time, identifying an abnormal mode matched with the characteristic early warning information set, calculating a comprehensive scheduling urgent coefficient of a node triggering the early warning mode, and screening out a high-priority scheduling node based on the comprehensive scheduling urgent coefficient; And reading the box information at the recovery node and comparing with the record on the chain, and generating and pushing the recovery task list when the accumulated recovery box amount reaches a threshold value.
2. The method for recycling the circulating package sharing schedule based on big data according to claim 1, wherein the method is characterized by obtaining circulation state data and inventory change data of the circulating package, and verifying the data through a blockchain, and comprises the following specific steps: The method comprises the steps of obtaining circulation state data and inventory change data of circulation packing boxes at all nodes, wherein the circulation state data comprise box body identification marks, warehouse-in and warehouse-out time and operation node positions which are acquired through RFID read-write equipment, and handover types and manager information which are recorded through handover terminals, and the inventory change data comprise real-time empty box inventory and on-road box inventory of all nodes, and warehouse-out box quantity, warehouse-in box quantity and box retention time in each node history period which are extracted from a history database; And carrying out consistency check and integrity check on the acquired data through a block chain.
3. The cyclic packaging sharing scheduling recovery method based on big data according to claim 2, wherein the method is characterized in that key circulation event hash processing is carried out on the collected circulation state data and inventory change data through a blockchain, hierarchical storage is carried out according to data access frequency and evidence storage requirements, and a full life cycle tracing chain is built based on circulation event hash association, and the method specifically comprises the following steps: Carrying out hash processing on key circulation events of each circulation packing box to generate a unique certificate storage identifier of the circulation event, wherein the key circulation events comprise a warehouse-in event, a cross-node transfer event and a recovery handover event of a box body; According to the access frequency and the certification requirement of the box body data, a block chain is utilized to execute a hierarchical storage strategy, namely, the real-time position update and current state change data of the box body are written into a main chain, historical transfer record and past transfer certificate data of the box body are transferred to a side chain or a distributed storage network according to a preset storage rule, meanwhile, a unique time mark generated based on a block chain timestamp server is allocated to each piece of uplink data, the generation time or the uplink processing time of the data is recorded, and a full life cycle tracing chain is constructed in a mode that each new block or new record contains the hash value of the previous record based on the hash association among all transfer events.
4. The cyclic packing sharing scheduling recovery method based on big data according to claim 3, wherein the method is characterized by extracting the historical circulation data of each node from the blockchain and constructing a characteristic data set, and the specific steps comprise: Extracting historical circulation data of the circulating packing box of each node in a past preset time period from a constructed full life cycle traceability chain, wherein the historical circulation data comprise daily warehouse-out box quantity, daily warehouse-in box quantity, average retention time of a box body and node service type identification; Performing data cleaning and normalization processing on the extracted historical circulation data, removing outliers caused by abnormal data acquisition, and supplementing missing data through an interpolation method to generate a standardized time sequence data set; Extracting features of the standardized time sequence data set to obtain feature information sets of all nodes, wherein the feature information comprises periodic features, trend features and fluctuation features; summarizing the feature information sets of all the nodes to construct a global feature data set.
5. The cyclic packaging sharing scheduling recovery method based on big data, which is characterized by identifying a demand fluctuation mode and constructing a characteristic early warning information set, comprises the following specific steps: performing cluster analysis on the global feature data set, and dividing nodes with similar box demand modes into a plurality of node class clusters, wherein the box demand modes are used for representing box feature vectors for the nodes, which are formed by the extracted periodic features, trend features and fluctuation features; Extracting feature information of the type of node which triggers emergency allocation or causes a box missing event in history aiming at each node class cluster to form a feature early warning information set of the type of node, wherein the feature early warning information set comprises a plurality of feature items, each feature item corresponds to one feature extracted from the history circulation data and is provided with a corresponding early warning threshold range; And writing the characteristic early warning information set of each node class cluster into the blockchain.
6. The method for recycling cyclic packaging shared scheduling based on big data according to claim 5, wherein the method is characterized by monitoring node circulation data in real time and identifying an abnormal mode matched with a characteristic early-warning information set, and comprises the following specific steps: calculating real-time feature vectors in the current time window of each node by using real-time uplink node circulation data based on a feature extraction method of the history circulation data; According to the category cluster to which each node belongs, a corresponding feature early-warning information set is called, and each feature item in the real-time feature vector is matched with a corresponding feature item in the feature early-warning information set; If the feature values of two continuous time windows exist in the real-time feature vector of a certain node and are beyond the pre-warning threshold range of the corresponding feature item in the feature pre-warning information set, judging that the node triggers the pre-warning mode, and extracting two real-time records corresponding to the node in the two time windows.
7. The cyclic packaging shared scheduling recovery method based on big data according to claim 6, wherein the method is characterized by calculating comprehensive scheduling urgent coefficients of nodes triggering an early warning mode, and comprises the following specific steps: Acquiring two corresponding real-time records of a node triggering the early warning mode in two continuous time windows, namely R a and R b , wherein the time of R a is earlier than that of R b ; Calculating a set of distinguishing features F (ab) =f a -F a ∩F b for F a relative to F b , and a set of distinguishing features F' (ab) =f b -F a ∩F b for F b relative to F a ; Calculating the deviation degree of each characteristic item in F (ab) and the corresponding characteristic item in the characteristic early warning information set of the class cluster to which the node belongs, taking the product of all the deviation degrees as a first comprehensive scheduling urgent coefficient beta a , and defining as shown in beta a =∏ f∈Fab θ f , wherein theta f represents the deviation degree of the characteristic item F and is defined as the ratio of a real-time characteristic value to an early warning threshold value; For each characteristic item in F' (ab), calculating the deviation degree of the characteristic item corresponding to the characteristic item in the characteristic early warning information set of the class cluster to which the node belongs, taking the product of all the deviation degrees as a second comprehensive scheduling urgent coefficient beta b , and defining as beta b =∏ g∈F'ab θ g , wherein theta g represents the deviation degree of the characteristic item g and is defined as the ratio of a real-time characteristic value to an early warning threshold value; If beta b >β a and beta b -β a > eta, determining the node as a high-priority scheduling node, and adding the identification of the node to a node list to be scheduled, wherein eta represents a preset comprehensive urgency coefficient threshold.
8. The cyclic packaging sharing scheduling recovery method based on big data according to claim 1 is characterized in that a scheduling task list is generated by inquiring a block chain idle empty box source node according to a high-priority scheduling node and is stored through a block chain upper chain, when the recovery node reads box information in batches through RFID equipment and compares the box information with the on-chain record, the recovery task list is generated and pushed when a recovery threshold is reached, and the specific steps include: According to the high-priority scheduling node serving as a target node, inquiring real-time empty box stock quantity of each node from a blockchain, screening available source nodes with idle empty boxes, determining the scheduling box quantity according to the box shortage quantity of the target node, generating a scheduling task list comprising the scheduling box quantity, a source node identifier and a target node identifier, and uploading the scheduling task list to a chain through the blockchain for evidence; The recycling nodes read the box body identifications of the circulating packing boxes in batches through RFID read-write equipment, and compare the box body identifications with the records to be recycled stored on the blockchain, so as to realize checking and update the box body state to be recycled; And counting the accumulated recycling bin quantity of each recycling node in real time, generating a recycling task list comprising recycling node identifiers, recycling bin quantity and bin identifier lists when the accumulated recycling bin quantity reaches a preset recycling batch threshold, and pushing the recycling task list to a recycling executive party.
9. The circulating package sharing dispatching recovery system based on big data is applied to the circulating package sharing dispatching recovery method based on big data, which is characterized by comprising a data acquisition and verification module, a blockchain storage evidence tracing module, a characteristic analysis early warning module and an intelligent dispatching recovery module, wherein the output end of the data acquisition and verification module is connected with the input end of the blockchain storage evidence tracing module; The RFID acquisition unit acquires circulation data comprising box body identification and warehouse-in and warehouse-out time, the handover recording unit records handover type and manager information, synchronously acquires node inventory data, and the blockchain verification unit performs consistency and integrity verification on the data and eliminates abnormal data; The block chain storage certificate tracing module comprises a hash processing unit, a layered storage unit and a tracing construction unit, wherein the hash processing unit generates a unique certificate identifier for a box body key stream event, the layered storage unit divides data into a main chain and a side chain according to access frequency, and the tracing construction unit constructs a box body full life cycle tracing chain based on hash association; the characteristic analysis early warning module comprises a data set construction unit, a cluster analysis unit and a real-time monitoring unit, wherein the data set construction unit extracts historical data on a chain and processes the historical data to generate a characteristic data set, the cluster analysis unit divides node class clusters and constructs a characteristic early warning information set, and the real-time monitoring unit matches node real-time data with early warning information to identify an abnormal mode; The intelligent dispatching recovery module comprises a priority computing unit, a dispatching task unit and a recovery executing unit, wherein the priority computing unit is used for computing the comprehensive dispatching urgent coefficient of abnormal nodes, screening high-priority nodes, the dispatching task unit is used for inquiring idle empty boxes to generate dispatching orders and uploading the dispatching orders, the recovery executing unit is used for finishing box inventory through RFID, and the recovery task orders are generated and pushed when the accumulated recovery amount reaches the standard.

Description

Circulating package sharing dispatching recovery system and method based on big data Technical Field The invention relates to the technical field of data analysis, in particular to a cyclic packaging sharing dispatching recovery system and method based on big data. Background The traditional circulation packaging management mode is suitable for the whole-flow management of new material circulation boxes and is used for realizing the process of sharing scheduling, and is prone to various problems that firstly, the sharing scheduling lacks big data analysis support, the supply and demand matching efficiency is low, the traditional mode relies on manual statistics of the stock, idle and in-transit states of the new material circulation boxes of all nodes, the fluctuation rule of the box demands for the big data mining nodes cannot be used for leading to unbalance of supply and demand among multiple nodes, the problems of the box idling and the box shortage coexist, the real sharing scheduling is difficult to realize, secondly, the box checking relies on manual operation, a trusted data tracing system does not exist, the circulation data of the new material circulation boxes are distributed, such as the warehouse-in and the transfer of the new material circulation boxes is lack of a block chain technology for unified storage and verification, the data is easy to tamper and lose, the manual checking mode of the box-by-box is time consuming and has large checking errors, automatic checking cannot be realized, and furthermore, the traditional recovery link triggers an untimely recovery or no-load condition easily occurs. Disclosure of Invention The invention aims to provide a circulating package sharing dispatching recovery system and method based on big data, which are used for solving the problems in the prior art. In order to achieve the purpose, the invention provides the technical scheme that the cyclic packaging sharing scheduling recovery method based on big data comprises the following steps: Acquiring circulation state data and inventory change data of a circulation packing box, and checking the data through a block chain; performing key circulation event hash processing on the collected circulation state data and inventory change data through a block chain, performing hierarchical storage according to data access frequency and evidence storage requirements, and constructing a full life cycle traceability chain based on circulation event hash association; extracting historical circulation data of each node from the block chain, and constructing a characteristic data set; identifying a demand fluctuation mode based on the characteristic data set, and constructing a characteristic early-warning information set; Monitoring node circulation data in real time, identifying an abnormal mode matched with the characteristic early warning information set, calculating a comprehensive scheduling urgent coefficient of a node triggering the early warning mode, and screening out a high-priority scheduling node based on the comprehensive scheduling urgent coefficient; And reading the box information at the recovery node and comparing with the record on the chain, and generating and pushing the recovery task list when the accumulated recovery box amount reaches a threshold value. The method comprises the specific steps of obtaining circulation state data and inventory change data of the circulation packing box, and checking the data through a block chain, wherein the specific steps comprise: The method comprises the steps of obtaining circulation state data and inventory change data of circulation packing boxes at all nodes, wherein the circulation state data comprise box body identification marks, warehouse-in and warehouse-out time and operation node positions which are acquired through RFID read-write equipment, and handover types and manager information which are recorded through handover terminals, and the inventory change data comprise real-time empty box inventory and on-road box inventory of all nodes, and warehouse-out box quantity, warehouse-in box quantity and box retention time in each node history period which are extracted from a history database; The method comprises the steps of carrying out consistency check and integrity check on collected data through a blockchain, specifically, checking whether key fields of each piece of data are complete or not based on a preset data format rule, checking whether the data are tampered in the transmission process based on a hash algorithm, eliminating repeated uploading or data with abnormal time stamps, classifying the checked data based on the current circulation state and node attribute of a box body, marking each piece of data with a collection time stamp and a geographic position label, writing the classified data into a blockchain distributed account book according to a preset storage strategy, wherein the data related to the real-time position of the box body a