Blockchain-based business process instance migration method, medium and electronic device

Abstract

The application discloses a kind of based on blockchain's business process instance migration method, medium and electronic equipment.Predefined business process model on blockchain will change with time elapses and the change of multi-party business needs, however, the current business process management method cannot provide enough flexibility to support such changes.The application combines the candidate instance prefix trajectory matching algorithm based on dictionary tree and the optimal candidate model version selection algorithm based on accurate graph edit distance, which can support efficient migration of instances on the chain.Through the method proposed in the application, the business process in the blockchain-based business process management engine can be effectively optimized, so that the blockchain-based business process dynamic change method can obtain better practicality and timeliness.

Description

Technical Field

[0001] This invention relates to the field of business process management technology, and in particular to a blockchain-based business process instance migration method, medium, and electronic device. Background Technology

[0002] As business processes become increasingly complex, many are no longer confined to a single organization, resulting in numerous business processes requiring cross-organizational collaboration. In these complex cross-organizational business processes, predefined business process models often evolve over time and with changing business needs from multiple parties. However, current business process management methods lack sufficient flexibility to support these changes, leading to various problems such as low enterprise management efficiency and increased security risks. Furthermore, regarding the evolution of business process models, a common practice for process instances currently executing according to older versions of the model but not yet completed is to terminate and restart the current process, which often means significant work loss. Therefore, since predefined business process models may not be perfect, business process management requires continuous modification of these models and support for appropriate business process evolution methods to optimize processes. Simultaneously, for these currently executing user process instances, adopting appropriate instance migration strategies can reduce process rollback operations and lower labor costs, thereby improving enterprise management efficiency and achieving the strategic goals of enterprise business process management. Therefore, designing comprehensive process model evolution and dynamic process instance solutions to help relevant personnel address the dynamic changes in cross-organizational business processes is of great significance to business process management.

[0003] In recent years, a number of methods or frameworks have emerged that control complex business processes through smart contracts. These methods focus on cross-organizational business process modeling and blockchain business process framework implementation. However, they still lack effective strategies for process model evolution and process instance migration. Migrating process instances currently executing on the blockchain is not easy, requiring consideration of the evolution of the process instance's reference model and the compliance of the process instance's state. Most current research integrates solutions to support model evolution and instance migration within traditional single-organizational business process frameworks, but these solutions are not suitable for cross-organizational business processes. Furthermore, most existing blockchain prototype frameworks ignore the immutable nature of smart contracts once deployed; further research needs to consider design methods such as general-purpose smart contracts to address the issue of low framework flexibility. Summary of the Invention

[0004] To overcome the shortcomings of the existing technologies, a blockchain-based business process instance migration method, medium, and electronic device are proposed, which can effectively solve the above problems.

[0005] The specific technical solution adopted in this invention is as follows:

[0006] In a first aspect, the present invention provides a blockchain-based method for migrating business process instances, which includes the following steps:

[0007] S1. Assemble all evolvable business process models across the target organization into a process model set; use an orchestration tool to model the process for each business process model in the process model set, forming a business process orchestration diagram in BPMN format;

[0008] S2. Select a business process model from the process model set as the original process instance on-chain reference model, and deploy the corresponding BPMN format business process orchestration diagram on-chain. Execute the business process instance on the deployed on-chain reference model.

[0009] S3. When a new business process arises that requires modification of the on-chain reference model, candidate business process models that meet the requirements of the new business process are selected from the process model set to form an evolvable candidate model set, Collections. VT Then, the complete business process trajectory is extracted from the off-chain log file of each candidate business process model, and a trie is constructed for the complete business process trajectory of each candidate business process model. At the same time, the on-chain business process instances that are being executed but have not yet been completed in the current on-chain reference model are extracted, and the executed trajectory of each on-chain business process instance is prefix matched with the trie corresponding to the candidate business process model to obtain the candidate business process model with the highest matching degree. All matched candidate business process models form the intermediate candidate model set Collections. candi_ ;

[0010] S4. For the intermediate candidate model set Collections candi_ For each candidate business process model in the dataset, the exact graph edit distance between it and the on-chain reference model is calculated to obtain the similarity between the two models. This is then used to select the intermediate candidate model set, Collections. candi_ The candidate business process model that is most similar to the on-chain reference model is selected as the optimal candidate model for instance migration.

[0011] S5. The version initiator imports the BPMN format business process orchestration diagram data corresponding to the optimal candidate model into Ethereum and conducts a model version vote. If the vote passes, a new on-chain reference model version is generated, and the on-chain reference model is updated to the optimal candidate model.

[0012] S6. Iterate through the on-chain business process instances that are currently being executed but not yet completed in the old version of the on-chain reference model. For each on-chain business process instance, it is necessary to determine whether it can be migrated to the new version of the on-chain reference model for continued execution. If it can, migrate it to the new version of the on-chain reference model for continued execution. If it cannot, continue to execute based on the old version of the on-chain reference model, or directly terminate the on-chain business process and create a new business process instance on the new version of the on-chain reference model.

[0013] As a preferred embodiment of the first aspect mentioned above, when the on-chain reference model is first deployed on the chain, smart contracts need to be compiled and deployed. The smart contracts include a tool library contract, an access control contract, a voting contract, a scaffolding contract, a version control contract, and an instance control contract.

[0014] As a preferred embodiment of the first aspect above, in S3, the intermediate candidate model set Collections candi_ The construction method is as follows:

[0015] S31. Traversing the Collections of Evolvable Candidate Models VT Extract the complete business process trajectory of each candidate business process model from the off-chain log file. If there are multiple loops in the complete business process trajectory, retain the first loop and delete other redundant loops. This gives the final business process trajectory corresponding to each candidate business process model.

[0016] S32, Traversing the Collections of Evolvable Candidate Models VT The final business process trajectory of each candidate business process model is passed into the trie construction function createTree, thereby completing the construction of the trie corresponding to the model;

[0017] S33. Extract all currently executing and incomplete on-chain business process instances from the current on-chain reference model. Using the trie matching function `startWith`, perform prefix matching on the execution trajectory of each on-chain business process instance across all candidate business process models in the trie. Select the candidate business process model with the highest matching degree as the similar business process model of the on-chain business process instance. Combine all similar business process models into an intermediate candidate model set `Collections`. candi_ .

[0018] As a preferred embodiment of the first aspect above, the method for determining the optimal candidate model in S4 is as follows:

[0019] S41, Based on the old version of the reference model S i The order of elements in the reference model S i Establish model structure diagram g iAdditionally, the intermediate candidate model set Collections candi_ A model structure diagram (Tg) is also established for all similar business process models. t This yields the graph structure list Tg;

[0020] S42. Traverse the graph structure list Tg, and calculate the model structure graph g respectively. i Each model structure diagram in the graph structure list Tg t The minimum evolution distance and the maximum evolution distance, wherein the minimum evolution distance is the distance between the model structure diagram g and the maximum evolution distance. i Evolutionary model structure diagram Tg t The minimum required graphic editing distance and the maximum evolution distance are the model structure graph g. i Evolutionary model structure diagram Tg t The required maximum graphic editing distance; then, the old version of the reference model S is calculated by dividing the difference between the maximum and minimum evolution distances by the maximum evolution distance. i Similar business process model S t The similarity representation between the models is used to select the similar business process model with the smallest similarity representation as the optimal candidate model S. x .

[0021] As a preferred embodiment of the first aspect above, in S6, for each on-chain business process instance, it is necessary to determine whether it can be migrated to the new version of the on-chain reference model for continued execution through both state compliance checks and migration compliance checks. Only after passing both state compliance checks and migration compliance checks can the on-chain business process instance be migrated to the new version of the on-chain reference model.

[0022] The conditions for passing the state compliance check are: the next orchestration activity element to be executed in the current on-chain business process instance is in an active state, and the messages of all preceding orchestration activity elements of the current orchestration activity element are in a receiving state.

[0023] The condition for passing the migration compliance check is that the completed preceding trajectories in the current on-chain business process instance are a subset of any off-chain business process trajectory corresponding to the new version of the on-chain reference model.

[0024] As a preferred embodiment of the first aspect above, the specific method for migrating the on-chain business process instance to the new version of the on-chain reference model for continued execution in step S6 is as follows:

[0025] S61. For the on-chain business process instance with instance number i_id to be migrated, obtain the corresponding optimal candidate model S through the contract address. x Store references in the blockchain and determine the optimal candidate model S. x The start event;

[0026] S62, Based on the optimal candidate model S x Call the `generateVersion` method of the version control contract (VCC) to obtain the S... x The corresponding version number is bestversion;

[0027] S63. Obtain the storage reference of the on-chain business process instance corresponding to the instance number i_id on the blockchain, and then update the reference model version of the on-chain business process instance i.

[0028] S64. Reset the state of on-chain business process instance i, and in the new version of the reference model S x The system automatically activates elements that have already been executed in the original instance, and activates the status of messages corresponding to the orchestration activities according to the original instance. After processing is complete, the instance migration ends.

[0029] As a preferred embodiment of the first aspect above, in S5, if a new on-chain reference model version is generated through voting, the scaffolding contract is redeployed through the Truffle environment, and the optimal candidate model S is imported through the contract interface. x The corresponding data structure is thus initialized; the optimal candidate model S x The data structures include a list of roles, orchestration elements, messages, and model versions across organizational processes.

[0030] In a second aspect, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the blockchain-based business process instance migration method as described in any of the solutions in the first aspect above.

[0031] Thirdly, the present invention provides a computer electronic device, which includes a memory and a processor;

[0032] The memory is used to store computer programs;

[0033] The processor is configured to, when executing the computer program, implement the blockchain-based business process instance migration method as described in any of the solutions of the first aspect above.

[0034] Compared to existing cross-organizational business process management frameworks, this invention has the following advantages:

[0035] This invention proposes a candidate instance trajectory prefix matching algorithm based on a trie, which can efficiently match transferable candidate models. Furthermore, it combines this with an optimal candidate model version selection algorithm based on precise graph edit distance to select an optimal cost process model, reducing process rollback operations and lowering the time and manpower costs associated with process evolution. Additionally, to achieve a smooth transition between old and new reference models, this invention requires checking the state compliance and migration compliance of on-chain instances before migration, thereby improving the security of instance execution and migration. Attached Figure Description

[0036] Figure 1 This is a flowchart of the blockchain-based business process instance migration method of the present invention.

[0037] Figure 2 It describes the commonly used elements of a layout diagram.

[0038] Figure 3 This is an example of a cross-organizational business process orchestration diagram in a healthcare setting.

[0039] Figure 4 This is an example diagram illustrating the evolution and migration of a cross-organizational business process model.

[0040] Figure 5 This is a schematic diagram illustrating the dynamic migration of a process instance based on the optimal prefix according to the present invention;

[0041] Figure 6 Example diagram of the original process model for reference in process instance;

[0042] Figure 7 This is a flowchart illustrating the distribution of process operation categories in the experimental data of this invention.

[0043] Figure 8 This is a diagram showing the distribution of user execution trajectory lengths in the experimental data of this invention.

[0044] Figure 9 The figure shows the impact of the method of the present invention on the model evolution efficiency.

[0045] Figure 10 This is a diagram showing the similarity results between the original process model and the candidate model using the method of the present invention;

[0046] Figure 11 This figure illustrates the impact of the method of the present invention on instance migration efficiency. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings.

[0048] Conversely, this invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of the invention as defined in the claims. Furthermore, to provide a better understanding of the invention, certain specific details are described in detail below. However, those skilled in the art will fully understand the invention even without these detailed descriptions.

[0049] To facilitate understanding, some definitions in this invention will be explained before describing the specific implementation of the invention.

[0050] Definition 1, Orchestration Diagram: TM = {E, M, D, T} represents an orchestration diagram, where E, M, D, and T represent the elements (including events, gateways, and orchestration activities), messages, decision conditions, and time constraints of the orchestration diagram, respectively. The element E = E startEvent E endEvent G parallel G exclusive ,TOT},E startEvent It is the starting event, E endEvent It is the end event, G parallel It is a parallel gateway, G exclusive A is a mutual exclusion gateway, and TOT is a time-constrained orchestration activity. Therefore, the orchestration diagram in this definition is a time-constrained orchestration diagram that takes into account the time constraints between orchestration activities, but this invention is also applicable to ordinary orchestration diagrams that do not consider the time constraints between orchestration activities.

[0051] Definition 2, Access Control Organization: CO = {U, R, UR} is an organizational structure that includes all participants in the business processes. Here, U represents a participant in the access control organization, R represents a role in the access control organization, and UR represents the mapping relationship between participants and roles in the access control organization.

[0052] Definition 3, Process Model: ML = {TM, CO} is a reference model for instance creation and execution. Here, TM represents the timing orchestration diagram, and CO represents the access control organization.

[0053] Definition 4, Locus: σ = <e1,2,..., |σ| > is a non-empty finite sequence of activities, representing the sequence of all orchestration activities occurring in a given process instance, where e |t| (t∈[1,|σ|]) represents the t-th orchestration activity (or start event, end event, or gateway) in the trajectory σ.

[0054] Definition 5, Process Example: I = (VID, D, List) part(t, σ) represents a user process created or executed on the blockchain business process engine based on a process model with version number VID. Here, VID represents the model version number to which the instance belongs; D = {IE, IM, IG} represents the instance's structural information, where IE, I, and IG represent instance elements, instance message content, and instance global data, respectively; List part σ represents the list of participants for the instance; σ represents the set of states for the instance; σ represents the execution trajectory of the process instance.

[0055] Definition 6, State: St = {St Inst ,t Eles ,t Mess} represents the set of states for a process instance. Where St Inst ={Running,Finish} represents the two states of a process instance: running and finished.

[0056] St Eles ={Waiting, Enabled, Completed} represents the state of an element in a process instance. Waiting, Enabled, and Completed represent the waiting, activated, and completed states, respectively, corresponding to binary values ​​00, 01, and 10; St Mess =Wait,Send,Receive,Reject represent the message status in the process instance, where Wait, Send, Receive, and Reject represent the message waiting, sending, receiving, and rejecting statuses, respectively, corresponding to binary values ​​00, 01, 10, and 11; using binary storage and retrieval of element status (or message) can reduce the execution time of instance activities.

[0057] Definition 7, Model Version: V = (VID, Status, ML) represents the version information of the reference model, a generic element component based on version control. Where VID represents the version number within this component, and Status represents the version status (Init means proposed, Pass means approved).

[0058] Definition 8, Flow Operation: τ=<op1,...,op m > represents operations that evolve within a specific process model. These operations include Move, Delete, and Add operations for moving orchestration activities, events, and gateways in the timing orchestration diagram.

[0059] Definition 9, Process Operation Distance: This represents the total number of operations (OPs). For example, if a process model uses three Move operations and one Delete operation during its evolution, then the process operation distance |τ| = 4.

[0060] Definition 10, Set of Process Models: Δ(S) i )=ζ1(τ1),…,ζ n (τ n ) represents the process model S i The set of all evolvable process models, where ζ1(τ) j )=τ(S i ,S′ j ) represents the process model S i Evolved to S′ j The evolutionary set. These evolvable process models are typically formed by process managers through multiple orchestrations based on the actual execution of the original business processes. For example, a process model evolves to a process model ζ with version number k. i (τ k )={|τ|=5,τ∈(|Move|=1,|Delete|=2,|Add|=2)}, which means that the process model can be transformed into a process model with version number k through one Move operation, two Delete operations, and two Add operations.

[0061] Definition 11, Utility Library Contract: CC = {evar, Efun} is a smart contract that encapsulates the public enumeration variables and computation functions that can be called by the above smart contracts. Here, Evar represents the public enumeration variable, and Efun represents the public computation function.

[0062] Definition 12, Access Control Contract: ACC = {CO addressMap ,RList,UList,ACC link This is a smart contract used to authorize access to smart contract operators. CO is... addressMap It stores the linked addresses of the access control organization (CO), RList = {role1, ..., role} m} All roles in the process of translating the storage metamodel (ML) to the contract, UList = {user1, ..., user} m} is the set of blockchain user accounts for all execution instances. The Access Control Organization (CO) = {U, R, UR} is an organizational structure containing all business process participants, where U represents a participant, R represents a role within the Access Control Organization, and UR represents the mapping between participants and roles within the Access Control Organization. ACC link This represents a reference to the access control contract after it has been deployed.

[0063] Definition 13, Voting Contract: VC = {VP, ACC} link ,VC link} is a smart contract used to publish proposals and have participants vote on them, which is implemented through ACC. linkIt links to the Access Control Contract (ACC) described in Chapter 3. The proposal VP = {ID, SC} address V ID Vote require Vote now `State` includes the proposal ID, scaffolding address, current version number, number of people required for the proposal to pass, number of people who have already passed the proposal, and proposal status (including Init status and Pass status). VC link This indicates a contract reference after the voting contract is deployed.

[0064] Definition 14, Scaffolding Contract: SC = {TM, ACC} link ,VC link ,SC link ,Name} is a smart contract that stores a timed orchestration graph™ structure, which is accessed via ACC link VC link The access control contract ACC and the voting contract VC are linked respectively. SC link This indicates a contract reference after the scaffolding contract is deployed, where Name is the name of the deployed scaffolding contract.

[0065] Definition 15, Version Control Contract: VCC = {V group ,SC link ,VC link VCC link} is a smart contract used to manage model versions, which is implemented through SC link and VC link It links to the scaffolding contract SC and the voting contract VC, respectively. V group ={(<E1,V1>,<M1,V1>,<D1,V1>,<T1,V1>),…,(<E n V n >, <M n V n >, <D n V n >, <T n V n >)} is a set of model versions used to manage elements E, messages M, and decisions D. VCC link This indicates a reference to the contract after the version-controlled contract has been deployed.

[0066] Definition 15, Version Control Contract: VCC = {V group ,SC link ,VC link VCC link} is a smart contract used to manage model versions, which is implemented through SC link and VC linkIt links to the scaffolding contract SC and the voting contract VC, respectively. V group ={(<E1,V1>,<M1,V1>,<D1,V1>,<T1,V1>),…,( <E n V n >, <M n ,Vn>, <D n V n >, <T n V n >)} is a set of model versions used to manage elements E, messages M, and decisions D. VCC link This indicates a reference to the contract after the version-controlled contract has been deployed.

[0067] Definition 16, Instance Control Contract: ICC = {IS, ACC} link ,SC link VCC link ICC link} is a smart contract used to manage process instances, which is implemented through ACC link SC link VCC link These are linked to the Access Control Contract (ACC), the Scaffolding Contract (SC), and the Version Control Contract (VCC), respectively. Here, IS represents the collection storing all process instances I. ICC link This indicates a reference to the contract after the instance control contract is deployed.

[0068] Definition 17: Minimum evolutionary distance.d min (g i ,g j ) represents the process model S i Evolved to S j The minimum edit distance, d, when converted to a graph. min It can be expressed as the following formula:

[0069]

[0070] Where, τ t Δ(S) represents the process operations (including Delete, Add, and Move) given in Definition 4.9; c is a cost function used to measure the distance of the process operation. According to Definition 4.10, the cost of one Delete, Add, or Move operation (i.e., the distance of one edit) is 1; Δ(S) i ) is the set of process models given in definition 4.11.

[0071] Definition 18: Maximum evolutionary distance.d max (g i ,g j ) represents the process model S i Evolved to Sj Transformed into the maximum edit distance of the graph. Specifically, d max (g i ,g j The editing process can be represented as follows: First, edit Figure g... i Delete all nodes and edges, and then insert graph g. j All corresponding vertices and edges. Let graph g be... i There are a total of p i vertices and q i Edge, g j There are a total of p j vertices and q j Edge, d max (g i ,g j It can be expressed as the following formula:

[0072] d max (g i ,g j ) = p i +q i +p j +q j

[0073] Definition 19: Model Similarity. For two process models S... i and S j The model similarity is defined as the ratio of the evolutionary distance between models, as shown in the following formula:

[0074]

[0075] Based on the above definition, the specific implementation process of the present invention will be described in detail below.

[0076] like Figure 1 As shown, the blockchain-based business process instance migration method of the present invention may specifically include the following steps:

[0077] S1. Assemble all evolvable business process models across the target organization into a process model set; use an orchestration tool to model each business process model in the process model set to form a business process orchestration diagram in BPMN format.

[0078] In this step, because a large amount of business process trajectory data is accumulated in the off-chain business process log files of cross-organizational enterprises, when building the process model set, process managers generally need to sort out cross-organizational business processes from the log files to form a series of evolvable business process models, providing a reference for process modeling. Simultaneously, it is also necessary to predict the possible evolution direction of business process models based on future business development, model them, and add them to the set of process models that need to be uploaded to the blockchain. After determining the corresponding business process models that need to be uploaded to the blockchain based on the analysis results, process managers can orchestrate the business process orchestration diagram™ in the chor-js modeler. Figure 2 It describes the common elements of an orchestration diagram, including events, gateways, orchestration activities, and sequence streams. Orchestration activities are divided into one-way activities and two-way activities based on the number of messages they carry. Figure 3 This is an example of a cross-organizational business process orchestration diagram in a healthcare scenario. The diagram involves six organizations: medical supply manufacturers, a supply distribution center, a transporter, a hospital (or clinic), a waste collection point, and a waste treatment plant. It includes six orchestration activities and messages. For example, corresponding to the first orchestration activity, "Medical Supply Distribution," after medical supplies are manufactured by the medical supply manufacturer, the activity is driven by the message "Medical Supply Distribution" as they are about to be distributed to the supply distribution center. Upon completion of the activity, the medical supplies are stored at the supply distribution center. Subsequent messages, such as "Supply Distribution Request," will then send the supplies to the hospital or clinic via a transporter. In this process, the orchestration diagram does not focus on the details within a single organization, such as ignoring the medical supply manufacturer's internal production process. Similarly, after the hospital uses the medical supplies, these discarded medical supplies need to be recycled. Therefore, the hospital and another organization, the waste collection point, drive the orchestration activity through the message "Waste Recycling Notification." Finally, the messages "Waste Transportation Notification" and "Waste Disposal Notification" drive inter-organizational collaboration, and the medical waste is then recycled by the waste treatment plant.

[0079] S2. Select a business process model from the process model set as the original process instance on-chain reference model, and deploy the corresponding BPMN format business process orchestration diagram on-chain. Execute the business process instance on the deployed on-chain reference model.

[0080] In an embodiment of the invention, the process manager parses the orchestration diagram TM in BPMN format into model components in JSON format using a translator, for use in the next step of model deployment. During model deployment, the process manager can perform the deployment through a Truffle environment. When the on-chain reference model is deployed on-chain for the first time, smart contracts need to be compiled and deployed. These smart contracts include the aforementioned utility library contract, access control contract, voting contract, scaffolding contract, version control contract, and instance control contract.

[0081] Therefore, the compilation and deployment process of the above smart contract specifically includes the following steps: first, deploy the tool library contract CC; then, link the tool library contract CC through the Truffle environment; finally, deploy the access control contract ACC, and obtain the contract reference ACC after the access control contract is deployed. link Then, the Truffle environment links to the library contract CC, and then passes in ACC. link Deploy the voting contract (VC) and obtain the contract reference (VC) after deployment. link Then, the Truffle environment links to the library contract CC, and then passes in ACC. link VC link Deploy the scaffolding contract SC using the scaffolding name (Name), and obtain the contract reference SC after the scaffolding contract is deployed. link Then, the tool library contract CC is linked through the Truffle environment, and then passed to VC. link SC link To deploy the version control contract (VCC), obtain the contract reference VCC after deployment. link Finally, the Truffle environment links to the library contract CC, and then passes in the ACC. link SC link VCC link Deploy the instance control contract (ICC) to obtain the contract reference ICC after deployment. link .

[0082] In an embodiment of this invention, the process manager needs to compile the Access Control Contract (ACC) and the Scaffolding Contract (SC) sequentially based on the Remix platform and the Truffle environment to obtain the ABI files corresponding to the two contracts. An ABI file is an Application Binary Interface file used to describe the interfaces in a smart contract. Furthermore, the process manager needs to deploy the Access Control Contract (ACC) and the Scaffolding Contract (SC) sequentially on the Ethereum platform to obtain the contract addresses after deployment. Then, by calling the Web3.js interface and passing in the ABI files and contract addresses of the two contracts, an instance of the contract is obtained. Finally, the control flow information of the metamodel and the Access Control Organization (CO) are obtained through the contract instance. The control flow information of the metamodel includes elements E of the timing orchestration graph, messages M, decision information D, and time constraint information T.

[0083] After the initial reference model deployment is completed, the business process model is put on-chain and begins to create corresponding process instances I according to user instructions. Although the initially selected business process model takes universality into consideration as much as possible, in complex cross-organizational business processes, the predefined business process model often changes over time and with changes in the business needs of various parties, or the predefined business process model may not be perfect. Therefore, in the process of business process management, it is necessary to continuously modify the business process model and support appropriate business process evolution methods to optimize the process. At this time, the cross-organizational process model evolution method of this invention is needed to continuously evolve the on-chain business process model and realize the dynamic update of the on-chain reference model. However, at this time, there are a large number of process instances still running on the original model. How to migrate these process instances is the most important key technology in the on-chain version update.

[0084] like Figure 4 As shown in the example, for a given cross-organizational business scenario within an enterprise, the original process model S serves as a reference model executed by some user instances. Due to changes in the actual business process, the orchestration activities of Activity(′Participant P1, Activity Q, Participant P2′) in the original process model S need to be reduced; that is, the original process model S needs to evolve. Considering the evolution process of the original process model S, the goal of this process is to reduce the process operation distance |τ|. For some given evolvable candidate models, the original process model S can usually be transformed into a new process model through multiple process operations, thereby achieving model evolution. For example... Figure 4 As shown, an efficient evolution operation τ(S,S') is as follows: First, the orchestration activity (′participant P2, activity R, participant P3′) is moved using the Move operation to the orchestration activity (′participant P1, activity H, participant P2′) and the parallel gateway and g1 In between, the parallel gateway is deleted via the Delete operation. g1and parallel gateway and g2 All elements in between eventually evolve into the process model S'. During the evolution of the business process model, dynamic migration of process instances generally needs to be considered, especially for process instances that are based on the old process reference model but have not yet finished executing. Considering the process instance migration process, for a given user process execution trajectory σ: A→B→C→…→H, when the next orchestrated activity (′Participant P2, Activity R, Participant P3′) is activated, the original process model S evolves. During this process, it is necessary to evaluate the relationship between the current user execution trajectory σ and the reference model (including the original process model S and candidate process models S'), i.e., to consider process compliance checks. If a user process instance does not meet process compliance requirements, it will continue to execute based on the original reference model S or the process will be restarted; conversely, user process instances that meet process compliance requirements need to be migrated. Specifically, before instance migration, the optimal process model S' needs to be selected from the candidate model set, and then the current user process instance is migrated to the new process model S'.

[0085] The following section will detail how this invention updates the reference model version of the business process and migrates user process instances. For example... Figure 5 As shown, this process is based on the optimal prefix to achieve matching and dynamic migration of process instances. It includes candidate model trajectory matching, optimal process model version selection and dynamic migration of process instances, which will be described in detail below.

[0086] S3. When a new business process arises that requires modification of the on-chain reference model, candidate business process models that meet the requirements of the new business process are selected from the process model set to form an evolvable candidate model set, Collections. VT Then, the complete business process trajectory is extracted from the off-chain log file of each candidate business process model, and a trie is constructed for the complete business process trajectory of each candidate business process model. At the same time, the on-chain business process instances that are being executed but have not yet been completed in the current on-chain reference model are extracted, and the executed trajectory of each on-chain business process instance is prefix matched with the trie corresponding to the candidate business process model to obtain the candidate business process model with the highest matching degree. All matched candidate business process models form the intermediate candidate model set Collections. candi_ .

[0087] In this invention, a trie is constructed for each candidate model, and an evolving set of intermediate candidate models is obtained by matching the trie with the user's executed trajectory. Most existing technologies often use arrays, hash tables, and tries to solve string flow trajectory matching problems. A trie is a data structure used for fast retrieval (such as longest prefix matching of strings). It utilizes the idea of ​​trading space for time, which can significantly reduce the time overhead of queries, thereby improving query efficiency. Let the number of candidate strings be n, and the length of the currently matched string be m. If an array is used for traversal matching, its time complexity is O(nm). Since traversal matching does not require additional space, the space complexity is O(1). If a hash table is used for storage, its space complexity will increase to... ( The time complexity will be reduced to O(1) (the sum of the lengths of the inserted strings), but hash tables need to solve the hash collision problem. Furthermore, hash tables do not support dynamic lookups; they must wait for the user to finish inputting the string to be matched before a lookup can be performed. Trie trees (tire) support dynamic string lookups, with a time complexity of O(m) and a space complexity of O(m). ( The sum of the lengths of the inserted strings is ω, where ω is the size of the character set (e.g., for lowercase English letters, ω = 26). In terms of time efficiency, its time complexity is comparable to a hash table and much faster than B-trees or other comparison-based indexing methods. In terms of space efficiency, building a trie incurs additional space overhead, but it can be improved by supporting node compression; research has already optimized its space complexity to O(cm) (where c is a constant coefficient, c ∈ (0.17~1.13)). Furthermore, the trie can be persistently stored on disk at any time and retrieved and loaded into memory when needed; it can also be created in memory and released after use without disk I / O. Based on the advantages and disadvantages of arrays, hash tables, and tries, this invention uses the trie to solve the problem of time-consuming searches for transferable process models. Moreover, the dynamic query characteristics of the trie are applicable to matching user trajectories of different lengths during process trajectory matching.

[0088] In embodiments of the present invention, the aforementioned intermediate candidate model set Collections candi_VT The construction method is as follows:

[0089] S31. Traversing the Collections of Evolvable Candidate Models VTExtract the complete business process trajectory of each candidate business process model from the off-chain log file. If there are multiple loops in the complete business process trajectory, retain the first loop and delete other redundant loops, thereby obtaining the final business process trajectory corresponding to each candidate business process model.

[0090] S32, Traversing the Collections of Evolvable Candidate Models VT The final business process trajectory of each candidate business process model is passed to the createTree function to construct the trie corresponding to the model.

[0091] S33. Extract all currently executing and incomplete on-chain business process instances from the current on-chain reference model. Using the trie matching function `startWith`, perform prefix matching on the execution trajectory of each on-chain business process instance across all candidate business process models in the trie. Select the candidate business process model with the highest matching degree as the similar business process model of the on-chain business process instance. Combine all similar business process models into an intermediate candidate model set `Collections`. candi_VT .

[0092] Prefix trajectory is defined as PT k (σ)=<e1,e2,…,e k >(1≤k≤|σ|) is a sequence consisting of k orchestration elements preceding the current element in the trajectory σ, where the orchestration elements include events, gateways, and orchestration activities in the timed orchestration graph. The candidate instance trajectory prefix matching algorithm based on the trie is represented in pseudocode as follows:

[0093]

[0094] The flow in the pseudocode of the above algorithm is as follows:

[0095] Step 1 (lines 13-15): Initialize the current candidate model set, Collections VT If the `tire` set `PrefixTree` is empty, proceed to step 2 to extract the complete trajectory and add it to the `Collections` set. VT .

[0096] Step 2 (lines 1-11): Perform loop reduction on the complete trajectory. Specifically, if there is a loop in the extracted complete trajectory, retain the first loop and delete the redundant loops.

[0097] Step 3 (line 16): Construct the trie of candidate models. The complete trajectory of each candidate model needs to be passed to the createTree method.

[0098] Step 4 (lines 17-21): By calling the `startWith` method of the trie `tire`, prefix matching is performed between the current user trajectory PT3(σ) and the candidate models, ultimately obtaining the intermediate candidate model set `Collections`. candi_VT Intermediate candidate models are used for subsequent selection of the optimal process model version and dynamic instance migration.

[0099] S4. For the intermediate candidate model set Collections candi_VT For each candidate business process model in the dataset, the exact graph edit distance between it and the on-chain reference model is calculated to obtain the similarity between the two models. This is then used to select the intermediate candidate model set, Collections. candi_VT The candidate business process model that is most similar to the on-chain reference model is selected as the optimal candidate model for instance migration.

[0100] Graph edit distance (GED) is a graph similarity metric, similar to Levenshtein distance for strings. It is defined as the cost of the minimum number of edit operations (node ​​and edge edits) required to transform a source graph into a target graph. The cost of each operation is derived by defining a corresponding cost function, which is the distance value. Existing graph edit distance algorithms can be mainly divided into two categories: exact graph edit distance algorithms and approximate graph edit distance algorithms. Solving for exact graph edit distance can have a higher time complexity, but improving search efficiency can reduce the time complexity.

[0101] In an embodiment of the present invention, the above-mentioned optimal candidate model is achieved by solving the precise map edit distance, and the specific solution method is as follows:

[0102] S41, Based on the old version of the reference model S i The order of elements in the reference model S i Establish model structure diagram g i Additionally, the intermediate candidate model set Collections candi_VT A model structure diagram (Tg) is also established for all similar business process models. t This yields the graph structure list Tg;

[0103] S42. Traverse the graph structure list Tg, and calculate the model structure graph g respectively. i Each model structure diagram in the graph structure list Tg t Minimum evolution distance d min (g i ,Tg t ) and maximum evolutionary distance d max (g i ,Tgt The minimum evolution distance d min (g i ,Tg t To transform the model structure diagram g i Evolutionary model structure diagram Tg t Minimum required graphic editing distance, maximum evolution distance d max (g i ,Tg t To transform the model structure diagram g i Evolutionary model structure diagram Tg t The maximum required graphic editing distance; then based on the minimum evolution distance d min (g i ,Tg t ) and maximum evolutionary distance d max (g i ,Tg t ) Calculate the old version of the reference model S i Similar business process model S t Similarity representation between Characterizing TS(S) by similarity i ,S j The smallest similar business process model is selected as the optimal candidate model S. x .

[0104] S5. The version initiator imports the BPMN format business process orchestration diagram data corresponding to the optimal candidate model into Ethereum and conducts a model version vote. If the vote passes, a new on-chain reference model version is generated, and the on-chain reference model is updated to the optimal candidate model.

[0105] The model version control strategy based on a voting mechanism mainly includes three processes: releasing a new version, model and version binding, and instance version control.

[0106] (1) Releasing a New Version: The process manager, as the version initiator, first releases a new version based on the original actual business process execution. The process of releasing a new version mainly includes version submission, voting decision, and version release. The implementation logic of the new version is as follows: The input of the algorithm is the version initiator's address and the list of voting participants, and the output of the algorithm is the version number VID. First, the account address of the version initiator needs to be verified. Initiators whose permissions are verified can release a new version. Each time the model is deployed, a new version needs to be released. The new version will be a proposal for the participants to vote on. After the voting decision process is completed for the new version created by the version initiator, if the number of votes for the new version is insufficient, the new version status will change to the Refuse status and return false (vote not passed). The process model version in the Refuse status can only be re-proposed or abandoned; similarly, if the number of votes for the new version meets the required number, the new version status will change to the Pass status and return true (vote passed).

[0107] (2) Model and version binding: If a new version is successfully released, the process manager can then bind the released new version with the process model to obtain model version V. i Specifically, by calling the scaffolding contract and the version control contract, version number i is bound to elements, messages, decisions, and time constraints in the process model;

[0108] (3) Instance version control: Create new instances based on the reference model and execute the instances. During the instance version control process, a series of process instances I (e.g., process instance I1 and process instance I2) are bound to the model version V containing version number i. i In this case, subsequent instance execution requires model version V. i This achieves the effect of version control instances.

[0109] Therefore, in the embodiments of the present invention, for the voting to generate a new on-chain reference model version, the scaffolding contract is redeployed through the Truffle environment, and the optimal candidate model S is imported through the contract interface. x The corresponding data structure is then initialized. Optimal candidate model S x The data structures include a list of roles, orchestration elements, messages, and model versions across organizational processes.

[0110] S6. Iterate through the on-chain business process instances that are currently being executed but not yet completed in the old version of the on-chain reference model. For each on-chain business process instance, it is necessary to determine whether it can be migrated to the new version of the on-chain reference model for continued execution. If it can, migrate it to the new version of the on-chain reference model for continued execution. If it cannot, continue to execute based on the old version of the on-chain reference model, or directly terminate the on-chain business process and create a new business process instance on the new version of the on-chain reference model.

[0111] In the embodiments of the present invention, for each on-chain business process instance, it is necessary to determine whether it can be migrated to the new version of the on-chain reference model for continued execution through both state compliance checks and migration compliance checks. Only by passing both state compliance checks and migration compliance checks can the on-chain business process instance be migrated to the new version of the on-chain reference model.

[0112] The conditions for passing the state compliance check are: the next orchestration activity element to be executed in the current on-chain business process instance is in an active state, and all messages of the preceding orchestration activity elements of the current orchestration activity element are in a receiving state.

[0113] The condition for passing the migration compliance check is that the completed preceding trajectories in the current on-chain business process instance are a subset of any off-chain business process trajectory corresponding to the new version of the on-chain reference model.

[0114] In an embodiment of the present invention, the specific method for migrating an on-chain business process instance to a new version of the on-chain reference model for continued execution is as follows:

[0115] S61. For the on-chain business process instance with instance number i_id to be migrated, obtain the corresponding optimal candidate model S through the contract address. x Store references in the blockchain and determine the optimal candidate model S. x The start event;

[0116] S62, Based on the optimal candidate model S X Call the `generateVersion` method of the version control contract (VCC) to obtain the S... x The corresponding version number is bestversion;

[0117] S63. Obtain the storage reference of the on-chain business process instance corresponding to the instance number i_id on the blockchain, and then update the reference model version of the on-chain business process instance i.

[0118] S64. Reset the state of on-chain business process instance i, and in the new version of the reference model S xThe system automatically activates elements that have already been executed in the original instance, and activates the status of messages corresponding to the orchestration activities according to the original instance. After processing is complete, the instance migration ends.

[0119] Furthermore, the above-described steps can be implemented as software functional units in the form of logical instructions in memory. When this software is 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 the present invention, or the part that contributes to the prior art, or a part 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, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0120] Similarly, based on the same inventive concept, another preferred embodiment of the present invention also provides a computer-readable storage medium corresponding to the blockchain cross-organizational process model evolution method provided in the above embodiments. The storage medium stores a computer program, which, when executed by a processor, implements the blockchain-based business process instance migration method as described in the foregoing scheme.

[0121] Similarly, based on the same inventive concept, another preferred embodiment of the present invention also provides a computer electronic device corresponding to the blockchain cross-organization process model evolution method provided in the above embodiments, which includes a memory and a processor;

[0122] The memory is used to store computer programs;

[0123] The processor is configured to implement the blockchain-based business process instance migration method as described in the foregoing scheme when executing the computer program.

[0124] It is understood that the aforementioned storage medium and memory can be random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Furthermore, the storage medium can also be any medium capable of storing program code, such as a USB flash drive, external hard drive, magnetic disk, or optical disk.

[0125] It is understood that the processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0126] It should also be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here. In the embodiments provided in this application, the division of steps or modules in the device and method is merely a logical functional division, and there may be other division methods in actual implementation. For example, multiple modules or steps may be combined or integrated together, and a module or step may also be split.

[0127] The following section applies the blockchain-based business process instance migration method shown in S1 to S6 to a specific example to demonstrate its technical effectiveness.

[0128] Example

[0129] The steps in this embodiment are the same as S1 to S6 in the specific implementation, and will not be repeated here. The following section demonstrates some of the implementation process and results:

[0130] Cross-organizational business process model acquisition: This embodiment compiles 42 cross-organizational business processes of a hearing aid company. These processes involve multiple participants, including individual customers, retail customers, raw material suppliers, manufacturers, logistics providers, and shipping carriers. This embodiment selects one as the original reference model example, and the rest as candidate models. The original reference model can be found here. Figure 6 During the similarity calculation between the original reference model and its candidate models, the candidate models undergo modification operations including moving, deleting, and adding. Only after these modifications can the original reference model evolve into a new candidate model. See... Figure 7In this embodiment, the modification operations for the remaining 41 candidate models of the experimental data are divided into the following four categories: (1) op_Add indicates that the new candidate model can be evolved only by adding operations compared to the original reference model. (2) op_Delete indicates that the new candidate model can be evolved only by deleting operations compared to the original reference model. (3) op_Move indicates that the new candidate model can be evolved only by moving operations compared to the original reference model. (4) op_All indicates that the original reference model can be evolved into a new candidate model only by combining the three operations of moving, deleting, and adding.

[0131] User process instance trajectory data acquisition: This embodiment simulates the creation and execution of 62 user process instances, and the relevant process instances are based on the aforementioned 42 cross-organizational business process models for reference and execution. Figure 8 The distribution of user execution trajectory lengths is given: (1) Approximately 12.9% of user execution trajectories are between 2 and 4. (2) Approximately 29.0% of user execution trajectories are between 4 and 6. (3) The largest proportion of user execution trajectory lengths are between 6 and 8, accounting for approximately 32.3%. (4) The remaining lengths are between 8 and 10, accounting for approximately 9.7%. (5) The remaining user execution trajectory lengths are greater than 10, accounting for approximately 16.1%.

[0132] To verify the technical effectiveness of the present invention, simulation experiments were conducted on the Ethereum private chain Ganache and Truffle framework. The present invention uses Gas as a cost metric for the framework. Gas is a unit of measurement for the computational resources required to execute an operation or deploy a smart contract on the blockchain, or to call a smart contract method. Table 1 shows the Gas consumption results for smart contract deployment involved in this invention, including Gas cost, Ether (ETH) size, and USD ($) size. One Ether is equal to 10. 9 Gwei, and one Ethereum is approximately equal to 2,586.37 US dollars.

[0133] Table 1 Contract Deployment Information

[0134]

[0135] This invention performs model evolution efficiency analysis: Figure 9 The time efficiency of the optimal candidate model version matching algorithm based on the exact graph edit distance is given, and in... Figure 10 The results of similarity calculation between the original reference model and the candidate model are presented. Figure 9 , 10As can be seen, the average operation time for op_Add is the shortest, and the average operation time for this edit is 1.49s. Specifically, if the graph corresponding to the original reference model can be obtained from the graph of the candidate model simply by adding the operation, then this algorithm is the most efficient, meaning the model evolution speed is the fastest. Furthermore, the graph shows that the average model similarity corresponding to the op_Add operation is relatively high. Figure 10 Clearly, the average operation time for `op_All` is the longest and significantly longer than other categories. This is because the operations in `op_All` are more complex. For example, if there are significant adjustments due to new actual business processes, the similarity between the new and original processes is low, resulting in lower efficiency in model evolution. Furthermore, it can be observed that the average model similarity of `op_Delete` and `op_Move` is not significantly different, and therefore their corresponding average operation times are also similar. Thus, as shown in the graph, model similarity and average operation time are generally negatively correlated. New and old process models with lower similarity often require more combinations of process operations to achieve editing and transformation, resulting in a higher average operation time consumed by the precise graph editing algorithm.

[0136] This invention includes an instance migration efficiency analysis: Figure 11 The efficiency of candidate process trajectory matching during instance migration varies with the length of the user process trajectory. As shown in the figure, both the ordinary loop traversal matching algorithm and the candidate instance trajectory prefix matching algorithm based on the trie (tire tree) of this invention show an increasing trend with the increase of the user process trajectory length. Overall, the increasing trend is more pronounced for the ordinary loop traversal matching algorithm, because the time complexity of the trie is O(m), while the time complexity of the ordinary loop traversal is O(nm). Furthermore, when the user process trajectory length is short, the ordinary loop traversal matching algorithm is faster because the trie is built and stored on disk, requiring time to retrieve. However, as the user process trajectory length increases, especially when the trajectory length is greater than 10, the average execution time of the ordinary loop traversal matching algorithm begins to exceed that of the candidate instance trajectory prefix matching algorithm based on the trie (tire tree) of this invention. In other words, the method of this invention is more efficient when the user trajectory length is large.

[0137] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the invention. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, all technical solutions obtained through equivalent substitution or transformation fall within the protection scope of the present invention.

Claims

1. A blockchain-based business process instance migration method, characterized in that, The method comprises the following steps: S1, all the evolvable business process models of the target cross-organizational business process are combined into a process model set; S2, a business process model in the process model set is selected as an original on-chain reference model, and the corresponding BPMN format business process orchestration diagram is deployed on the chain, and a business process instance is executed on the deployed on-chain reference model; S3, when a new business process needs to modify the on-chain reference model, candidate business process models that meet the new business process are selected from the process model set to form an evolvable candidate model set; Then the complete business process traces are extracted from the off-chain log files of each candidate business process model, and the complete business process traces of each candidate business process model are constructed into a dictionary tree; meanwhile, the on-chain business process instances that are being executed and have not been completed in the current on-chain reference model are extracted, and the executed traces of each on-chain business process instance are prefix-matched with the dictionary tree corresponding to the candidate business process model to obtain the candidate business process model with the highest matching degree, and all the matched candidate business process models form an intermediate candidate model set; S4, for each candidate business process model in the intermediate candidate model set, the precise graph edit distance between the candidate business process model and the on-chain reference model is calculated to obtain the similarity between the two models, and the candidate business process model that is most similar to the on-chain reference model is selected from the intermediate candidate model set as the optimal candidate model for instance migration; S5, the version initiator imports the BPMN format business process orchestration diagram data corresponding to the optimal candidate model into Ethereum, and performs model version voting, if the voting is passed, a new on-chain reference model version is generated, and the on-chain reference model is updated to the optimal candidate model; S6, the on-chain business process instances that are being executed and have not been completed in the old version of the on-chain reference model are traversed, and each on-chain business process instance needs to be judged whether it can be migrated to the new version of the on-chain reference model for continuous execution, if yes, it is migrated to the new version of the on-chain reference model for continuous execution, if not, it continues to be executed based on the old version of the on-chain reference model, or directly terminates the on-chain business process and creates a new business process instance on the new version of the on-chain reference model. When the on-chain reference model is first deployed on the chain, a smart contract needs to be compiled and deployed, and the smart contract includes a tool library contract, an access control contract, a voting contract, a scaffold contract, a version control contract, and an instance control contract. 2.The blockchain-based business process instance migration method of claim 1, wherein, In S4, the optimal candidate model is determined as follows: 3.The blockchain-based business process instance migration method of claim 1, wherein, In the S3, the intermediate candidate model collection Collections candi_VT is constructed as follows: S31, traversing the set of evolvable candidate models Collections VT extracting the complete business process trace of each candidate business process model in the off-chain log file, and if there are multiple loops in the complete business process trace, retaining the first loop and deleting other redundant loops, thereby obtaining the final business process trace corresponding to each candidate business process model; S32, traverse the set of evolvable candidate models Collections VT The final business process track of each candidate business process model is transmitted into the dictionary tree construction function createTree, so as to complete the construction of the model corresponding dictionary tree. S33, extract all on-chain business process instances being executed and not yet completed in the current on-chain reference model, prefix match the executed trajectory of each on-chain business process instance in the dictionary tree of all candidate business process models through the matching function startWith of the dictionary tree, take the candidate business process model with the highest matching degree as the similar business process model of the on-chain business process instance, and group all similar business process models to form an intermediate candidate model set Collections candi_VT . 4.The blockchain-based business process instance migration method of claim 1, wherein, In S6, for each on-chain business process instance, state compliance check and migration compliance check are needed to judge whether it can be migrated to the new version of the on-chain reference model for continuous execution, and the on-chain business process instance can be migrated to the new version of the on-chain reference model only after passing the state compliance check and the migration compliance check. S41, according to the old version of the reference model S i The order of the elements, the reference model S i Establish model structure diagram g i ; In addition to the intermediate candidate model set Collections candi_VT All similar business process model in the middle of the model structure diagram Tg t , get the graph structure list Tg; S42, traverse the graph structure list Tg, respectively calculate the model structure graph g i The minimum evolutionary distance and the maximum evolutionary distance of each model structure graph Tg in the graph structure list Tg, the minimum evolutionary distance is the minimum graph edit distance required to evolve the model structure graph g t Tg into the model structure graph Tg i The maximum evolutionary distance is the maximum graph edit distance required to evolve the model structure graph g t Tg into the model structure graph Tg i The minimum evolutionary distance is the minimum graph edit distance required to evolve the model structure graph g t Tg into the model structure graph Tg i The maximum evolutionary distance is the maximum graph edit distance required to evolve the model structure graph g t The similarity between the old version of the reference model S x The similarity between the old version of the reference model S 5.The blockchain-based business process instance migration method of claim 1, wherein, ​ The passing condition of the state compliance check is that the next current orchestration activity element to be executed in the current on-chain business process instance is in an activated state, and the messages of all the preceding orchestration activity elements of the current orchestration activity element satisfy a receiving state; The passing condition of the migration compliance check is that the executed preceding track in the current on-chain business process instance is a subset of any off-chain business process track corresponding to the new version of the on-chain reference model. 6.The blockchain-based business process instance migration method of claim 1, wherein, In S6, the specific method of migrating the on-chain business process instance to the new version of the on-chain reference model for continuous execution is: S61, for the instance to be migrated, the instance number of the chain business process is i_id, the corresponding optimal candidate model S is obtained through the contract address x The storage reference in the block chain is determined, and the start event of the optimal candidate model S x ​ S62、According to the optimal candidate model S x , call the method generateVersion of the version control contract VCC to obtain S x The corresponding version number bestversion; S63, obtaining the storage reference of the on-chain business process instance corresponding to the instance serial number i_id on the blockchain, and then updating the reference model version of the on-chain business process instance i; S64, reset the state of the chain business process instance i, and migrate the instance to the new version of the reference model S x The elements that have been executed in the original instance are automatically activated, and the state of the messages corresponding to the orchestration activities is activated according to the original instance. After the processing is completed, the instance migration is ended.

7. The blockchain-based business process instance migration method of claim 1, wherein, The S5, for voting through the generation of a new chain reference model version, through the Truffle environment to deploy the scaffold contract again, and through the contract interface to import the optimal candidate model S x The corresponding data structure is completed; the optimal candidate model S x The data structure in the S includes the role list of the cross-organizational process, the arrangement element, the message and the model version.

8. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, and when the computer program is executed by the processor, the migration method of the business process instance based on the blockchain is realized.

9. A computer electronic device, comprising: comprising a memory and a processor; The memory is used for storing a computer program; The processor is used for realizing the migration method of the business process instance based on the blockchain when the computer program is executed.

Images