An industrial collaboration-oriented digital economy resource optimal allocation system and method

Abstract

The application provides an industrial collaboration-oriented digital economic resource optimal allocation system and method, relates to the technical field of digital economic resource allocation, adopts three-dimensional tensor operation, fuses real-time supply and demand and pre-calculated industrial multi-dimensional complementary relationship, realizes fine and global quantification of the combined collaborative value of "supply side-demand side-resource type", solves through construction of a binary integer programming model, automatically outputs an optimal matching list maximizing the overall collaborative value, automatically decomposes the matching list into an execution step graph with logical dependence based on a pre-constructed process library, generates a scheduling plan according to historical time consumption fluctuation and sorting, enhances the executability and adaptability of the scheme, finally, verifies the output scheme through model detection, fundamentally ensures that the output scheme has no logical deadlock and meets all timing constraints, and thus delivers a high-quality and reliable resource allocation scheme.

Description

Technical Field

[0001] This invention relates to the field of digital economy resource allocation technology, and in particular to a digital economy resource optimization allocation system and method for industrial collaboration. Background Technology

[0002] In the context of the digital economy, industrial collaboration is becoming increasingly frequent. A key challenge is how to efficiently allocate dispersed digital economy resources, such as computing power, data, and software services, to improve the overall efficiency of the industrial chain. Current resource allocation methods often focus on static or simple pairwise supply and demand matching, lacking global dynamic quantification and optimization of complex collaborative relationships among multiple industries and resource types. Existing technologies often struggle to integrate and process massive, heterogeneous industrial supply and demand data in real time, failing to deeply integrate multi-dimensional factors such as historical collaboration reputation, resource technology attribute matching degree, and inter-industry correlation strength into a computable collaborative potential model. Furthermore, when generating configuration schemes, existing methods typically separate matching from execution scheduling, failing to fully consider the risks of temporal dependency logic conflicts during task execution and the uncertainties brought about by historical execution volatility. This results in poor executability and a lack of flexibility in the formulated plans, making it difficult to maximize cross-industry collaborative value while meeting the capacity and reputation constraints of multiple parties. Therefore, there is an urgent need for a resource allocation method that can achieve a closed-loop process from real-time supply and demand perception, global value quantification, intelligent matching optimization to elastic scheduling and formal verification to solve the above problems.

[0003] Therefore, it is necessary to provide a digital economy resource optimization allocation system and method oriented towards industrial collaboration to solve the above-mentioned technical problems. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a digital economy resource optimization and allocation system and method for industry collaboration, achieving the beneficial effect of intelligent and efficient digital economy resource allocation.

[0005] This invention provides a method for optimizing the allocation of digital economy resources for industrial collaboration, including:

[0006] S1: Based on real-time business data from various industries, generate standardized real-time supply and demand matrices.

[0007] S2: Perform tensor operations on the real-time supply matrix and real-time demand matrix with the pre-constructed inter-industry resource complementarity coefficient tensor to obtain the three-dimensional potential collaborative value tensor.

[0008] S3: Based on the real-time supply matrix, the three-dimensional potential collaborative value tensor, the capacity upper limit vector, and the credit score threshold, construct a binary integer programming model and solve it to obtain a list of collaborative matching pairs.

[0009] S4: Based on the pre-built business collaboration process library, the collaborative matching pair list is decomposed into multiple execution step nodes with temporal dependencies, and an execution step dependency graph is constructed.

[0010] S5: Based on historical execution step time records and preset risk buffer coefficients, calculate the buffer time of each execution step node in the execution step dependency graph, and generate an elastic scheduling plan through topology sorting;

[0011] S6: Compile the elastic scheduling plan into a verifiable scheduling timing model, verify it using a model testing tool, and output the final resource configuration scheme after successful verification.

[0012] Preferably, in step S1, generating the standardized real-time supply matrix and real-time demand matrix includes:

[0013] Based on a predefined global industry set and a standard resource type set, a supply matrix framework and a demand matrix framework with empty initial values ​​are constructed. In the matrix framework, the rows correspond to industries and the columns correspond to resource types.

[0014] Collect real-time business data for each industry, including industry identifier, resource type identifier, quantity, unit of measurement, and business type identifier;

[0015] Clean the real-time business data and map the resource type identifier and unit of measurement to a standard resource type identifier and standard unit of measurement to obtain standardized data records.

[0016] Based on the business type identifier in the standardized data records, the standardized data records are classified into supply datasets and demand datasets respectively;

[0017] The supply and demand datasets are populated into the supply matrix framework and demand matrix framework according to the two dimensions of industry identification and standard resource type identification, respectively, to generate the real-time supply matrix and real-time demand matrix.

[0018] Preferably, in step S2, the construction steps of the pre-constructed inter-industry resource complementarity coefficient tensor include:

[0019] Based on a predefined global industry set and standard resource type set, a three-dimensional inter-industry resource complementarity coefficient tensor framework with an initial value of empty is constructed. The three dimensions correspond to the supply-side industry, the demand-side industry, and the standard resource type, respectively.

[0020] Based on the historical industry collaborative transaction record database, the historical transaction success rate is calculated for each element position in the three-dimensional inter-industry resource complementarity coefficient tensor framework, corresponding to the supply-side industry, demand-side industry, and standard resource type, and is used as the historical collaborative coefficient.

[0021] Based on the resource attribute knowledge base, the technical parameter matching degree is calculated for each element position in the three-dimensional inter-industry resource complementarity coefficient tensor framework, corresponding to the supply-side industry, demand-side industry and standard resource type, and used as the resource adaptation coefficient.

[0022] Based on the industry classification map, the strength of the industrial chain linkage is calculated for the supply-side industry and the demand-side industry corresponding to each element position in the three-dimensional inter-industry resource complementarity coefficient tensor framework, and is used as the industry linkage coefficient.

[0023] For each element position in the three-dimensional inter-industry resource complementarity coefficient tensor framework, its corresponding historical synergy coefficient, resource adaptation coefficient, and industry correlation coefficient are fused and calculated according to preset weights and filled into that position to generate the inter-industry resource complementarity coefficient tensor.

[0024] Preferably, in step S2, the steps for obtaining the three-dimensional potential collaborative value tensor include:

[0025] By performing an outer product operation between each column vector of the real-time supply matrix and the corresponding column vector of the real-time demand matrix, a three-dimensional intermediate tensor is obtained.

[0026] The three-dimensional intermediate tensor is multiplied element-wise with the pre-constructed inter-industry resource complementarity coefficient tensor, and the tensor obtained after element-wise multiplication is used as the three-dimensional potential collaborative value tensor.

[0027] Preferably, in step S3, the step of obtaining the list of cooperative matching pairs includes:

[0028] The combination of the supply-side industry, demand-side industry, and standard resource type corresponding to each element in the three-dimensional potential collaborative value tensor is defined as a binary decision variable. A value of 1 for the binary decision variable indicates that the combination is selected to establish a collaborative relationship, and a value of 0 indicates that it is not selected.

[0029] The optimization objective is to maximize the sum of the products of all binary decision variables and the corresponding element values ​​in the three-dimensional potential collaborative value tensor.

[0030] Establish capacity constraints, namely, for each supplier industry and each standard resource type, the sum of the supply quantities of the resource in the real-time supply matrix associated with all decision variables that select that supplier industry to provide that standard resource type must not exceed the upper limit value specified for that industry and that standard resource type in the capacity upper limit vector;

[0031] Establish a credit constraint condition: only when the credit score of the demand-side industry is not lower than the credit score threshold can it be set to 1 as the decision variable of the demand-side.

[0032] Solve the binary integer programming model consisting of optimization objective, capacity constraints, and reputation constraints. Output the industry and resource combinations corresponding to the decision variables with a value of 1 in the solution results to generate a list of collaborative matching pairs.

[0033] Preferably, in step S4, the business collaboration process library stores standard task process templates for different industry collaboration scenarios. Each template defines the standard execution step sequence required to complete a collaboration and the logical dependencies between each step.

[0034] Preferably, in step S4, constructing the execution step dependency graph includes:

[0035] Based on the industry and resource types involved in the matching pairs in the collaborative matching pair list, query the standard task process templates in the business collaboration process library, instantiate the template, and use the instantiated steps as nodes. Connect them according to the dependencies in the template to construct an execution step dependency graph.

[0036] Preferably, in step S5, calculating the buffer time for each execution step node in the execution step dependency graph includes:

[0037] From the historical execution step time records, obtain the actual time of all historical steps of the same type as the node, calculate the standard deviation of the time, and multiply the standard deviation by the preset risk buffer coefficient to obtain the buffer time of the node.

[0038] Preferably, in step S6, the verification using a model checking tool includes:

[0039] Verification can verify whether the scheduling timing model satisfies two formal reduction properties: deadlock-free and time window constraints for all steps.

[0040] This invention provides a digital economy resource optimization and allocation system for industry collaboration, comprising:

[0041] The data standardization module is used to generate standardized real-time supply and demand matrices based on real-time business data from various industries.

[0042] The value quantification module is used to perform tensor operations on the real-time supply matrix and real-time demand matrix with the pre-constructed inter-industry resource complementarity coefficient tensor to obtain a three-dimensional potential collaborative value tensor.

[0043] The matching optimization module is used to construct a binary integer programming model based on the real-time supply matrix, the three-dimensional potential collaborative value tensor, the capacity upper limit vector, and the credit score threshold, and solve it to obtain a list of collaborative matching pairs.

[0044] The task decomposition module is used to decompose the list of collaborative matching pairs into multiple execution step nodes with temporal dependencies based on a pre-built business collaboration process library, and to construct an execution step dependency graph.

[0045] The elastic scheduling module is used to calculate the buffer time of each execution step node in the execution step dependency graph based on historical execution step time records and preset risk buffer coefficients, and generate an elastic scheduling plan through topology sorting.

[0046] The scheme verification module is used to compile the elastic scheduling plan into a verifiable scheduling timing model, verify it through a model testing tool, and output the final resource configuration scheme after successful verification.

[0047] Compared with related technologies, the digital economy resource optimization allocation system and method for industry collaboration provided by this invention has the following beneficial effects:

[0048] This invention solves the integration challenge of multi-source heterogeneous real-time data by predefining a global set and standardizing mappings, providing a structurally consistent supply and demand matrix for computation. Employing three-dimensional tensor operations, it integrates real-time supply and demand with pre-calculated multi-dimensional complementary relationships within the industry, achieving a refined and global quantification of the collaborative value of the "supply-demand-resource type" combination. By constructing a binary integer programming model with this three-dimensional value tensor as the benefit coefficient and strictly incorporating capacity and reputation constraints, it can automatically output the optimal matching list that maximizes the overall collaborative value while meeting business constraints. Based on a pre-built process library, the matching list is automatically decomposed into execution step diagrams with logical dependencies. Elastic buffer time is injected based on historical time volatility, and a scheduling plan is generated through sorting, enhancing the executability and adaptability to uncertainty. Finally, by compiling the scheduling plan into a formal temporal model and performing model testing and verification, it fundamentally ensures that the output plan is free of logical deadlocks and meets all temporal constraints, thus delivering a high-quality, reliable, deterministic resource allocation plan. Attached Figure Description

[0049] Figure 1 A flowchart of a digital economy resource optimization allocation method for industry collaboration according to the present invention;

[0050] Figure 2 This is a module structure diagram of a digital economy resource optimization and allocation system for industry collaboration according to the present invention. Detailed Implementation

[0051] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the drawings, not all structures. Moreover, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0052] It should also be noted that, for ease of description, the accompanying drawings show only the parts relevant to the invention and not all of them. Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the operations (or steps) as sequential processes, many of the operations can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but it may also have additional steps not included in the drawings. The process may correspond to a method, function, procedure, subroutine, subroutine, etc.

[0053] Example 1

[0054] A digital economy resource optimization allocation method oriented towards industrial collaboration, in its specific implementation process, such as... Figure 1 As shown, a flowchart of a digital economy resource optimization allocation method for industry collaboration according to the present invention is illustrated, including:

[0055] Step S1: Generate a standardized real-time supply matrix and a real-time demand matrix based on real-time business data from various industries.

[0056] Specifically, in step S1, generating the standardized real-time supply matrix and real-time demand matrix includes:

[0057] Based on a predefined global industry set and a standard resource type set, a supply matrix framework and a demand matrix framework with empty initial values ​​are constructed. In the matrix framework, the rows correspond to industries and the columns correspond to resource types.

[0058] Collect real-time business data for each industry, including industry identifier, resource type identifier, quantity, unit of measurement, and business type identifier;

[0059] Clean the real-time business data and map the resource type identifier and unit of measurement to a standard resource type identifier and standard unit of measurement to obtain standardized data records.

[0060] Based on the business type identifier in the standardized data records, the standardized data records are classified into supply datasets and demand datasets respectively;

[0061] The supply and demand datasets are populated into the supply matrix framework and demand matrix framework according to the two dimensions of industry identification and standard resource type identification, respectively, to generate the real-time supply matrix and real-time demand matrix.

[0062] In the specific implementation process, firstly, based on a pre-defined and stored global industry set and a pre-defined and stored standard resource type set, a supply matrix framework is constructed where rows correspond to each industry in the global industry set and columns correspond to each standard resource type in the standard resource type set, with all cells having an initial value of zero. Simultaneously, a demand matrix framework with the same row and column structure and initial values ​​is also constructed. Next, real-time business data is collected from the various industry business systems connected to the system. Each real-time business data record must contain a clear industry identifier, resource type identifier, quantity expressed numerically, unit of measurement, and a business type identifier indicating whether the record provides or requires resources. Subsequently, each collected real-time business data record is cleaned. For example, according to a pre-defined mapping rule table, the original resource type identifier and unit of measurement in the record are uniquely mapped to the corresponding standard resource type identifier and standard unit of measurement in the standard resource type set, thereby obtaining a standardized... Data recording ensures the uniformity and comparability of all data in terms of resource type and unit of measurement. Then, based on the business type identifier carried in each standardized data record, it is precisely sorted into two independent datasets: records whose business type identifier indicates resource provision are assigned to the supply dataset, and records whose business type identifier indicates resource demand are assigned to the demand dataset. Finally, each record in the supply dataset is traversed, and the corresponding row in the supply matrix framework is located based on the industry identifier in the record, and the corresponding column is located based on its standard resource type identifier. The quantity in the record is then added to the cell value of that row and column. After the traversal is completed, the final real-time supply matrix is ​​generated. At the same time, the demand dataset is traversed with the exact same logic, and the quantity is added to the corresponding cell in the demand matrix framework to generate the final real-time demand matrix. This completes the entire transformation process from raw heterogeneous business data to standardized, structured matrix data, providing a dimensionally defined and numerically accurate input foundation for all subsequent calculations.

[0063] Step S2: Perform tensor operations on the real-time supply matrix and real-time demand matrix with the pre-constructed inter-industry resource complementarity coefficient tensor to obtain the three-dimensional potential collaborative value tensor.

[0064] Specifically, in step S2, the construction steps of the pre-constructed inter-industry resource complementarity coefficient tensor include:

[0065] Based on a predefined global industry set and standard resource type set, a three-dimensional inter-industry resource complementarity coefficient tensor framework with an initial value of empty is constructed. The three dimensions correspond to the supply-side industry, the demand-side industry, and the standard resource type, respectively.

[0066] Based on the historical industry collaborative transaction record database, the historical transaction success rate is calculated for each element position in the three-dimensional inter-industry resource complementarity coefficient tensor framework, corresponding to the supply-side industry, demand-side industry, and standard resource type, and is used as the historical collaborative coefficient.

[0067] Based on the resource attribute knowledge base, the technical parameter matching degree is calculated for each element position in the three-dimensional inter-industry resource complementarity coefficient tensor framework, corresponding to the supply-side industry, demand-side industry and standard resource type, and used as the resource adaptation coefficient.

[0068] Based on the industry classification map, the strength of the industrial chain linkage is calculated for the supply-side industry and the demand-side industry corresponding to each element position in the three-dimensional inter-industry resource complementarity coefficient tensor framework, and is used as the industry linkage coefficient.

[0069] For each element position in the three-dimensional inter-industry resource complementarity coefficient tensor framework, its corresponding historical synergy coefficient, resource adaptation coefficient, and industry correlation coefficient are fused and calculated according to preset weights and filled into that position to generate the inter-industry resource complementarity coefficient tensor.

[0070] Specifically, in step S2, the steps for obtaining the three-dimensional potential collaborative value tensor include:

[0071] By performing an outer product operation between each column vector of the real-time supply matrix and the corresponding column vector of the real-time demand matrix, a three-dimensional intermediate tensor is obtained.

[0072] The three-dimensional intermediate tensor is multiplied element-wise with the pre-constructed inter-industry resource complementarity coefficient tensor, and the tensor obtained after element-wise multiplication is used as the three-dimensional potential collaborative value tensor.

[0073] In the specific implementation process, firstly, an offline calculation process for a pre-constructed inter-industry resource complementarity coefficient tensor is performed. Based on a predefined global industry set and standard resource type set, a three-dimensional tensor framework is constructed, with the row dimension corresponding to all supply-side industries, the column dimension corresponding to all demand-side industries, and the depth dimension corresponding to all standard resource types, and all elements having an initial value of zero. Next, from the historical industry collaborative transaction record database, for each element position in this three-dimensional tensor framework determined by a specific supply-side industry, a specific demand-side industry, and a specific standard resource type, all historical transaction records involving that resource type are extracted from both parties, and the number of transactions marked as successful is counted. The historical collaboration coefficient for a given location is calculated by using the proportion of total transactions. For example, considering supply-side industry A, demand-side industry B, and standard resource type x, all records of A trading resource x with B are selected from the historical transaction record database. The number of records marked "successful" is counted, and this number is divided by the total number of transaction records. The resulting ratio is the historical transaction success rate for that location. Simultaneously, from the resource attribute knowledge base, detailed technical parameter attributes of the specific standard resource type are obtained for the same element location. The matching degree between the technical parameters of the resources provided by the supply-side industry and the demand-side industry is calculated to determine the resource success rate for that location. For example, the adaptation coefficient is obtained from the resource attribute knowledge base, which retrieves multiple technical parameters corresponding to the standard resource type, including but not limited to accuracy, speed, compatibility version, and their values. Typical values ​​of each technical parameter are statistically analyzed when supply-side industry A and demand-side industry B used the resource in historical collaborations. By comparing the similarity of these typical values, the technical parameter matching degree is quantified. Furthermore, based on a pre-set industry classification map, the system analyzes the degree to which the supply-side and demand-side industries involved in the same element position co-occur in the same product chain within the industry chain topology, thereby calculating the matching degree for that position. For example, based on an industry classification graph, industries are considered as nodes, and upstream and downstream supply and supporting synergy relationships are considered as edges. An industry relationship network graph is constructed, and the shortest path length between industry A and industry B in the industry relationship network graph is calculated. The reciprocal of the shortest path length is taken as a measure of the strength of the industry chain association. After the parallel calculation of the above three coefficients is completed, a preset and fixed fusion weight is assigned to the historical synergy coefficient, resource adaptation coefficient and industry association coefficient respectively. The three coefficients at each element position are weighted and summed, and the calculation results are filled into the corresponding position of the tensor frame, thereby completing the construction and storage of the inter-industry resource complementarity coefficient tensor.Subsequently, during online resource allocation calculations, a pre-constructed inter-industry resource complementarity coefficient tensor is loaded, and real-time supply and demand matrices are obtained. Each column in the real-time supply matrix, representing the supply volume data of all industries for a specific standard resource type, is treated as a column vector. Simultaneously, the column vectors corresponding to the same standard resource type in the real-time demand matrix are extracted. These two column vectors are then multiplied by an outer product to obtain a two-dimensional matrix. The value of each cell in this two-dimensional matrix equals the product of the supply volume of the supplying industry for that resource and the demand volume of the demanding industry for that resource. This operation is repeated for all standard resource types, and the resulting multiple... Two-dimensional matrices are stacked according to resource type dimension to form a three-dimensional intermediate tensor. Finally, this three-dimensional intermediate tensor is multiplied element-wise with the pre-constructed inter-industry resource complementarity coefficient tensor. That is, the values ​​corresponding to the same supply-side industry, the same demand-side industry, and the same standard resource type in the two tensors are multiplied. The result is directly used as the element value of the corresponding position in the three-dimensional potential collaborative value tensor. This value comprehensively represents the quantitative collaborative potential generated by a specific supplier providing specific resources to a specific demand-side under the current real-time supply and demand state, which combines historical relationships, technology matching, and industrial linkages, thereby obtaining the three-dimensional potential collaborative value tensor.

[0074] Step S3: Based on the real-time supply matrix, the three-dimensional potential collaborative value tensor, the capacity upper limit vector, and the credit score threshold, construct a binary integer programming model and solve it to obtain a list of collaborative matching pairs.

[0075] Specifically, in step S3, the steps for obtaining the list of cooperative matching pairs include:

[0076] The combination of the supply-side industry, demand-side industry, and standard resource type corresponding to each element in the three-dimensional potential collaborative value tensor is defined as a binary decision variable. A value of 1 for the binary decision variable indicates that the combination is selected to establish a collaborative relationship, and a value of 0 indicates that it is not selected.

[0077] The optimization objective is to maximize the sum of the products of all binary decision variables and the corresponding element values ​​in the three-dimensional potential collaborative value tensor.

[0078] Establish capacity constraints, namely, for each supplier industry and each standard resource type, the sum of the supply quantities of the resource in the real-time supply matrix associated with all decision variables that select that supplier industry to provide that standard resource type must not exceed the upper limit value specified for that industry and that standard resource type in the capacity upper limit vector;

[0079] Establish a credit constraint condition: only when the credit score of the demand-side industry is not lower than the credit score threshold can it be set to 1 as the decision variable of the demand-side.

[0080] Solve the binary integer programming model consisting of optimization objective, capacity constraints, and reputation constraints. Output the industry and resource combinations corresponding to the decision variables with a value of 1 in the solution results to generate a list of collaborative matching pairs.

[0081] In the specific implementation process, firstly, for each element in the three-dimensional potential collaborative value tensor—that is, each unique combination consisting of a specific supply-side industry, a specific demand-side industry, and a specific standard resource type—a binary decision variable is created. This decision variable can only be assigned a value of one or zero, where an assignment of one represents the decision to establish a collaborative matching relationship between the supply-side industry and the demand-side industry providing the standard resource type, and an assignment of zero represents not establishing this matching relationship. Next, the objective function of the optimization model is constructed. Its mathematical form is to find the maximum value of the sum of the products of all binary decision variables and their corresponding element values ​​in the three-dimensional potential collaborative value tensor, i.e., the collaborative value of the combination. This objective function drives the model to prioritize industry and resource combinations with higher collaborative value. Then, the system constructs the constraints of the model. The first type is capacity constraints. For example, for the capacity upper limit value recorded in the capacity upper limit vector for each supply-side industry and each standard resource type, all decision variables assigned a value of one and involving the specific supply quantity of the resource recorded in the real-time supply matrix corresponding to the combination of the supply-side industry and the standard resource type are summed. This summation... The first type of constraint is a credit constraint. The credit score must be less than or equal to the upper limit of the production capacity to ensure that the total resources output by any supplier do not exceed its actual production capacity. The second type is a credit constraint. For example, only when the real-time credit score of a certain demander's industry is not lower than the platform's preset credit score threshold are all binary decision variables in the model allowed to be assigned a value of one during the solution process. If its credit score is lower than the threshold, these decision variables are forcibly assigned a value of zero, thus excluding demanders with poor credit from the matching range. After defining the objective function and the above two types of constraints, a complete binary integer programming model is constructed. Then, a professional mathematical programming solver is called to solve the model. The solver automatically calculates and outputs a set of binary decision variable assignment schemes that maximize the objective function and satisfy all constraints. Finally, the solution result is analyzed, and the supplier industry identifier, demander industry identifier, and standard resource type identifier corresponding to all decision variables assigned a value of one are extracted to form a clear matching record. The set of all matching records constitutes the final collaborative matching pair list, thus completing the automated generation process from quantified value to optimal matching decision.

[0082] Step S4: Based on the pre-built business collaboration process library, decompose the collaboration matching pair list into multiple execution step nodes with temporal dependencies, and construct the execution step dependency graph.

[0083] Specifically, in step S4, the business collaboration process library stores standard task process templates for different industry collaboration scenarios. Each template defines the standard execution step sequence required to complete a collaboration and the logical dependencies between each step.

[0084] Specifically, in step S4, constructing the execution step dependency graph includes:

[0085] Based on the industry and resource types involved in the matching pairs in the collaborative matching pair list, query the standard task process templates in the business collaboration process library, instantiate the template, and use the instantiated steps as nodes. Connect them according to the dependencies in the template to construct an execution step dependency graph.

[0086] In the specific implementation process, firstly, each matching record in the collaborative matching pair list is read sequentially. Each matching record in the collaborative matching pair list clearly defines the supply-side industry, demand-side industry, and standard resource type. Based on the industry type and resource type involved in the record, a combined query condition is used to retrieve and match the most suitable standard task process template in the business collaboration process library. The pre-built business collaboration process library is a knowledge base storing standard task process templates for various industry collaboration scenarios. Each standard task process template precisely defines the sequence of standard execution steps that must be performed to complete that type of industry collaboration, and clearly defines the logical dependencies between each standard execution step and the preceding and following steps. For example, a certain step can only begin after another specified step has been completed. Then, the template is instantiated, that is, a corresponding template is created for each standard execution step defined in the template. Each execution step node has a unique identifier. This instantiation process binds specific industry and resource information from the matching records to the node, making it an executable task. After instantiating all matching records into templates and obtaining all execution step nodes, directed edges are established between all relevant execution step nodes based on the logical dependencies defined within each instantiated standard task flow template. For example, if the template stipulates that step A must be completed before step B, then a directed edge is established between the instantiated step A node and step B node, indicating that node A is a prerequisite of node B. By traversing and connecting all nodes, a complete and visually representative execution step dependency graph is finally formed, representing the execution order and constraints of all tasks. This completes the transformation from abstract collaborative matching decisions to a concrete, time-constrained, operable task network.

[0087] Specifically, S5: Based on the historical execution step time records and the preset risk buffer coefficient, calculate the buffer time of each execution step node in the execution step dependency graph, and generate an elastic scheduling plan through topology sorting.

[0088] Specifically, in step S5, calculating the buffer time for each execution step node in the execution step dependency graph includes:

[0089] From the historical execution step time records, obtain the actual time of all historical steps of the same type as the node, calculate the standard deviation of the time, and multiply the standard deviation by the preset risk buffer coefficient to obtain the buffer time of the node.

[0090] In the specific implementation process, to calculate the dedicated buffer time for each execution step node in the execution step dependency graph, firstly, the historical execution step time record database is accessed. Based on the standard step type identifier of the current node to be calculated, the actual time consumption data recorded for all historical execution step instances with the same node type are retrieved from the database, forming a historical time consumption dataset for this type of step. Then, statistical analysis is performed on this historical time consumption dataset to calculate the standard deviation of all actual times in the dataset, thereby quantifying the time volatility and uncertainty exhibited by this type of step in historical execution. Subsequently, a fixed value pre-configured and stored by the system administrator, namely the preset risk buffer coefficient, is called. This coefficient is multiplied by the calculated standard deviation, and the product is determined as the buffer time for this execution step node. Then, the process continues in sequence. After calculating the buffer time for each node in the execution step dependency graph, these calculated buffer time values ​​are appended as time attributes to the corresponding execution step nodes. Then, a topological sorting algorithm is run on the complete execution step dependency graph with the buffer time attributes appended. This algorithm automatically analyzes the dependency edges between all nodes in the graph and outputs a linear sequence of execution steps in which all predecessor nodes of any node are placed before that node. At the same time, when generating the sequence, the algorithm comprehensively considers the basic time consumption of the steps represented by each node and its additional buffer time, thereby compiling an ordered execution plan with elastic time windows reserved on both critical and non-critical paths, which is a time-resistant scheduling plan. This plan realizes the transformation from a task network with dependencies to an operable scheduling scheme with time resilience.

[0091] Step S6: Compile the elastic scheduling plan into a verifiable scheduling timing model, verify it using a model testing tool, and output the final resource configuration scheme after successful verification.

[0092] Specifically, in step S6, the verification using the model detection tool includes:

[0093] Verification can verify whether the scheduling timing model satisfies two formal reduction properties: deadlock-free and time window constraints for all steps.

[0094] In the specific implementation process, the first step is to perform a compilation operation, converting the elastic scheduling plan into a verifiable scheduling time series model that can be directly read and analyzed by formal verification tools. A verifiable scheduling time series model, also known as a formal time series model, is a general concept in the field of formal methods. The elastic scheduling plan is compiled into a mathematical model for automatic verification by model checking tools, including but not limited to time automata network models. In this model, each execution step is mapped to an automaton process, the step time and buffer time are mapped to in-process clock variables, and the dependencies between steps are mapped to inter-process channel synchronization or shared variable constraints. This method is a common technique in the field of formal verification, and model checking tools, including but not limited to UPPAAL, [are used for this purpose]. TAPAAL can perform exhaustive and symbolic analysis on such models to verify whether they meet specified temporal logic specifications. For example, it maps each execution step node in the scheduling plan to an independent state machine in the temporal model, maps the temporal dependencies between nodes to channels for synchronization and communication between state machines, and maps the earliest start time, latest end time, fixed duration, and additional buffer time of each node to one or more clock variables within the corresponding state machine, along with their guard conditions and reset rules. This ensures that all temporal logic and constraints in the original scheduling plan are completely equivalent and accurately expressed in the temporal model. Subsequently, it can be verified that the scheduling temporal model formally defines two mandatory property properties that must be satisfied: the first is the deadlock-free property, which requires that no set of permanently blocked loops caused by mutual waiting for resources or signals appear on any possible state evolution path of the model; the second is the satisfiability of time window constraints for all steps, which requires that each mapped step in the model... The state machine process, once activated, always completes state transitions and triggers subsequent processes within the time window specified by its clock variable. There are no timeout or default paths that are inevitably caused by tight time windows or conflicts. After defining the specification attributes, a preset model checking tool is called to load the compiled verifiable scheduling timing model and the aforementioned formal specification attributes, and to start the automated verification process. The model checking tool uses a state space search algorithm to systematically traverse or calculate all possible behavioral trajectories of the model to rigorously verify whether the two defined specification attributes are true on all trajectories. After the verification process is completed, the model checking tool will output a clear verification conclusion. If the conclusion is passed, it indicates that the elastic scheduling plan is feasible in both logic and time and has no inherent contradictions. Then, it is packaged together with the cooperative matching pair list, the execution step dependency graph, and the elastic scheduling plan to generate a comprehensive and deterministic final resource allocation scheme document containing "matching relationships, task decomposition, timing arrangement, and formal verification certificate" and output it.If the verification result is unsuccessful, the model checking tool will simultaneously generate a specific counterexample path. This path precisely describes the state transition sequence that leads to the attribute violation. The system uses this counterexample path as feedback information. If the verification result is unsuccessful, the model checking tool will simultaneously generate a specific counterexample path describing the state transition sequence that leads to the attribute violation. This counterexample path will be used as key feedback information, automatically backtracking to step S5. In step S5, based on the time conflict or logical contradiction revealed by the counterexample path, the system will make targeted adjustments to, but not limited to, the preset risk buffer coefficient and scheduling logic, and recalculate the buffer time of the execution step nodes and regenerate the elastic scheduling plan. Subsequently, step S6 is iteratively executed to recompile and verify the new scheduling plan. This closed-loop process of "backtracking-adjustment-verification" will be repeated until a final solution that can pass verification is generated. To ensure that the process will inevitably terminate, a clear termination condition will be preset. For example, a maximum number of iterations will be set. If a feasible solution is not generated after the termination condition is reached, a diagnostic report containing the final counterexample will be output instead of the final solution. Through the above mechanism, this method ensures that any resource allocation scheme output has logical correctness and time feasibility verified by formal methods.

[0095] The working principle of the digital economy resource optimization allocation method for industry collaboration provided by this invention is as follows:

[0096] This invention first generates a unified, dimensionally defined real-time supply and demand matrix by predefining a global set of industry and resource types and cleaning and mapping real-time business data, thus establishing a reliable data foundation for subsequent calculations. Second, it innovatively introduces a three-dimensional tensor as the core computational model, performing calculations on the real-time supply and demand matrix with a pre-constructed multi-dimensional complementary coefficient tensor that integrates historical collaboration records, resource technical attributes, and industry-related knowledge. This generates a refined, quantitative collaborative value assessment for each "supply-demand-resource type" combination. Next, using this three-dimensional value tensor as benefit coefficients, and combining it with the upper limit of production capacity representing supply capacity and the reputation threshold representing cooperation risk as hard constraints, a binary integer programming model aimed at maximizing global collaborative value is constructed and solved, automatically outputting the optimal collaborative matching pair list. Then, based on a predefined standard business process knowledge base, the abstract matching list is automatically instantiated into a network diagram of specific execution steps with clear logical dependencies. Furthermore, to address the uncertainty of the execution process, elastic buffer time is injected into each step based on the statistical analysis results of historical time consumption data, and a scheduling plan with anti-fluctuation capabilities is generated through topological sorting. Finally, to ensure the absolute reliability of the output scheme, the scheduling plan is compiled into a formal time series model and rigorous logical deadlock and time satisfiability verification is performed using model checking tools, forming a closed loop of "generation-verification-iterative optimization", thereby ensuring that the final delivered resource allocation scheme has global optima, executability and theoretical correctness.

[0097] Example 2

[0098] A digital economy resource optimization and allocation system oriented towards industrial collaboration, in its specific implementation process, such as... Figure 2 As shown, it illustrates a modular structure diagram of a digital economy resource optimization and allocation system for industry collaboration according to the present invention, including:

[0099] The data standardization module 100 is used to generate standardized real-time supply and demand matrices based on real-time business data from various industries.

[0100] The value quantification module 200 is used to perform tensor operations on the real-time supply matrix and real-time demand matrix with the pre-constructed inter-industry resource complementarity coefficient tensor to obtain a three-dimensional potential collaborative value tensor.

[0101] The matching optimization module 300 is used to construct a binary integer programming model based on the real-time supply matrix, the three-dimensional potential collaborative value tensor, the capacity upper limit vector, and the credit score threshold, and solve it to obtain a list of collaborative matching pairs.

[0102] The task decomposition module 400 is used to decompose the list of collaborative matching pairs into multiple execution step nodes with temporal dependencies based on a pre-built business collaboration process library, and to construct an execution step dependency graph.

[0103] The elastic scheduling module 500 is used to calculate the buffer time of each execution step node in the execution step dependency graph based on historical execution step time records and preset risk buffer coefficients, and generate an elastic scheduling plan through topology sorting.

[0104] The scheme verification module 600 is used to compile the elastic scheduling plan into a verifiable scheduling timing model, verify it through the model testing tool, and output the final resource configuration scheme after the verification is successful.

[0105] The working principle of the digital economy resource optimization and allocation system for industry collaboration provided by this invention is as follows:

[0106] This invention begins with a data standardization module 100, responsible for collecting and cleaning real-time business data and generating a structured supply and demand matrix. A value quantification module 200 receives these matrices, performs tensor operations with pre-stored complementary coefficient tensors, and outputs a quantified three-dimensional collaborative value tensor. A matching optimization module 300, using the three-dimensional collaborative value tensor as its core, combines capacity and reputation constraints to construct and solve a binary integer programming model, automatically determining the optimal list of collaborative matching pairs. A task decomposition module 400, based on a pre-stored standard process knowledge base, automatically instantiates this list into an execution step network diagram with clear logical dependencies. A flexible scheduling module 500 injects buffer time into each step in this network diagram based on historical time volatility and generates a flexible scheduling plan that can cope with uncertainty through sorting. Finally, a solution verification module 600 transforms the scheduling plan into a formal model and performs rigorous automatic verification to ensure the output solution is logically sound and time-sequentially feasible. These modules are sequentially connected and data-driven, together forming a fully automated and verifiable resource allocation system from raw data processing, intelligent decision-making, task planning to solution reliability assurance.

[0107] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0108] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compactdisc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0109] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

Claims

1. A method for optimizing the allocation of digital economy resources for industrial collaboration, characterized in that, The configuration method includes: S1: Based on real-time business data from various industries, generate standardized real-time supply and demand matrices. S2: Perform tensor operations on the real-time supply matrix and real-time demand matrix with the pre-constructed inter-industry resource complementarity coefficient tensor to obtain the three-dimensional potential collaborative value tensor. S3: Based on the real-time supply matrix, the three-dimensional potential collaborative value tensor, the capacity upper limit vector, and the credit score threshold, construct a binary integer programming model and solve it to obtain a list of collaborative matching pairs; S4: Based on the pre-built business collaboration process library, the collaborative matching pair list is decomposed into multiple execution step nodes with temporal dependencies, and an execution step dependency graph is constructed. S5: Based on historical execution step time records and preset risk buffer coefficients, calculate the buffer time of each execution step node in the execution step dependency graph, and generate an elastic scheduling plan through topology sorting; S6: Compile the elastic scheduling plan into a verifiable scheduling time sequence model, verify it using a model testing tool, and output the final resource configuration scheme after successful verification; In step S2, the construction steps of the pre-constructed inter-industry resource complementarity coefficient tensor include: Based on a predefined global industry set and standard resource type set, a three-dimensional inter-industry resource complementarity coefficient tensor framework with an initial value of empty is constructed. The three dimensions correspond to the supply-side industry, the demand-side industry, and the standard resource type, respectively. Based on the historical industry collaborative transaction record database, the historical transaction success rate is calculated for each element position in the three-dimensional inter-industry resource complementarity coefficient tensor framework, corresponding to the supply-side industry, demand-side industry, and standard resource type, and is used as the historical collaborative coefficient. Based on the resource attribute knowledge base, the technical parameter matching degree is calculated for each element position in the three-dimensional inter-industry resource complementarity coefficient tensor framework, corresponding to the supply-side industry, demand-side industry and standard resource type, and used as the resource adaptation coefficient. Based on the industry classification map, the strength of the industrial chain linkage is calculated for the supply-side industry and the demand-side industry corresponding to each element position in the three-dimensional inter-industry resource complementarity coefficient tensor framework, and is used as the industry linkage coefficient. For each element position in the three-dimensional inter-industry resource complementarity coefficient tensor framework, its corresponding historical synergy coefficient, resource adaptation coefficient and industry correlation coefficient are fused and calculated according to preset weights and filled into the position to generate the inter-industry resource complementarity coefficient tensor. In step S2, the steps for obtaining the three-dimensional potential collaborative value tensor include: By performing an outer product operation between each column vector of the real-time supply matrix and the corresponding column vector of the real-time demand matrix, a three-dimensional intermediate tensor is obtained. The three-dimensional intermediate tensor is multiplied element-wise with the pre-constructed inter-industry resource complementarity coefficient tensor, and the tensor obtained after element-wise multiplication is used as the three-dimensional potential collaborative value tensor.

2. The method for optimizing the allocation of digital economy resources for industrial collaboration according to claim 1, characterized in that, In step S1, generating the standardized real-time supply matrix and real-time demand matrix includes: Based on a predefined global industry set and a standard resource type set, a supply matrix framework and a demand matrix framework with empty initial values ​​are constructed. In the matrix framework, the rows correspond to industries and the columns correspond to resource types. Collect real-time business data for each industry, including industry identifier, resource type identifier, quantity, unit of measurement, and business type identifier; Clean the real-time business data and map the resource type identifier and unit of measurement to a standard resource type identifier and standard unit of measurement to obtain standardized data records. Based on the business type identifier in the standardized data records, the standardized data records are classified into supply datasets and demand datasets respectively; The supply and demand datasets are populated into the supply matrix framework and demand matrix framework according to the two dimensions of industry identification and standard resource type identification, respectively, to generate the real-time supply matrix and real-time demand matrix.

3. The method for optimizing the allocation of digital economy resources for industrial collaboration according to claim 2, characterized in that, In step S3, the steps for obtaining the list of cooperative matching pairs include: The combination of the supply-side industry, demand-side industry, and standard resource type corresponding to each element in the three-dimensional potential collaborative value tensor is defined as a binary decision variable. A value of 1 for the binary decision variable indicates that the combination is selected to establish a collaborative relationship, and a value of 0 indicates that it is not selected. The optimization objective is to maximize the sum of the products of all binary decision variables and the corresponding element values ​​in the three-dimensional potential collaborative value tensor. Establish capacity constraints, namely, for each supplier industry and each standard resource type, the sum of the supply quantities of the resource in the real-time supply matrix associated with all decision variables that select that supplier industry to provide that standard resource type must not exceed the upper limit value specified for that industry and that standard resource type in the capacity upper limit vector; Establish a credit constraint condition: only when the credit score of the demand-side industry is not lower than the credit score threshold, it is allowed to be set to 1 as the decision variable of the demand-side. Solve the binary integer programming model consisting of optimization objective, capacity constraints, and reputation constraints. Output the industry and resource combinations corresponding to the decision variables with a value of 1 in the solution results to generate a list of collaborative matching pairs.

4. The method for optimizing the allocation of digital economy resources for industrial collaboration according to claim 3, characterized in that, In step S4, the business collaboration process library stores standard task process templates for different industry collaboration scenarios. Each template defines the standard execution step sequence required to complete a collaboration and the logical dependencies between each step.

5. A method for optimizing the allocation of digital economy resources for industrial collaboration as described in claim 4, characterized in that, In step S4, constructing the execution step dependency graph includes: Based on the industry and resource types involved in the matching pairs in the collaborative matching pair list, query the standard task process templates in the business collaboration process library, instantiate the template, and use the instantiated steps as nodes. Connect them according to the dependencies in the template to construct an execution step dependency graph.

6. The method for optimizing the allocation of digital economy resources for industrial collaboration according to claim 5, characterized in that, In step S5, calculating the buffer time for each execution step node in the execution step dependency graph includes: From the historical execution step time records, obtain the actual time of all historical steps of the same type as the node, calculate the standard deviation of the time, and multiply the standard deviation by the preset risk buffer coefficient to obtain the buffer time of the node.

7. A method for optimizing the allocation of digital economy resources for industrial collaboration as described in claim 6, characterized in that, In step S6, the validation using a model checking tool includes: Verification can verify whether the scheduling timing model satisfies two formal reduction properties: deadlock-free and time window constraints for all steps.

8. A digital economy resource optimization allocation system for industrial collaboration, characterized in that, The allocation system, applied to a digital economy resource optimization allocation method for industrial collaboration as described in any one of claims 1 to 7, comprises: The data standardization module is used to generate standardized real-time supply and demand matrices based on real-time business data from various industries. The value quantification module is used to perform tensor operations on the real-time supply matrix and real-time demand matrix with the pre-constructed inter-industry resource complementarity coefficient tensor to obtain a three-dimensional potential collaborative value tensor. The matching optimization module is used to construct a binary integer programming model based on the real-time supply matrix, the three-dimensional potential collaborative value tensor, the capacity upper limit vector, and the credit score threshold, and solve it to obtain a list of collaborative matching pairs. The task decomposition module is used to decompose the list of collaborative matching pairs into multiple execution step nodes with temporal dependencies based on a pre-built business collaboration process library, and to construct an execution step dependency graph. The elastic scheduling module is used to calculate the buffer time of each execution step node in the execution step dependency graph based on historical execution step time records and preset risk buffer coefficients, and generate an elastic scheduling plan through topology sorting. The scheme verification module is used to compile the elastic scheduling plan into a verifiable scheduling timing model, verify it through a model testing tool, and output the final resource configuration scheme after successful verification.

Images