Process parameter iterative optimization method and system based on multi-source data

Abstract

The present application relates to the technical field of business data, and discloses a process parameter iterative optimization method and system based on multi-source data, which comprises the following steps: collecting multi-source business data, a resource association table and a task timing set of a target business process, analyzing the steady state and abnormal characteristics in the data to determine a reasonable adjustment interval of resource configuration and a task to be improved, and forming a preliminary optimization scheme; taking resources as the main body, fusing task timing and dependency relationship, and constructing a resource cooperation graph; mapping the preliminary scheme to the graph to simulate and deduce the full node load change; comparing the load increase value of the full node load change concentrated node with the available resource capacity of the node, and identifying a resource overload risk set of the business process; and adjusting the preliminary optimization scheme set according to the resource overload risk set to obtain a collaborative optimization execution sequence of the target business process; and the present application can improve the accuracy of process parameter optimization of the target business process.

Description

Technical Field

[0001] This invention relates to the field of commercial data technology, and in particular to a method and system for iterative optimization of process parameters based on multi-source data. Background Technology

[0002] In the field of optimizing process parameters for target business processes, existing technologies often struggle to fully integrate core information such as materials, equipment, and personnel from multi-source business data. They rely solely on single-dimensional data or localized process fragments to formulate optimization solutions, resulting in a lack of consideration for the overall operational logic of the business process. These solutions fail to effectively analyze the temporal relationships and resource dependencies of task execution, and cannot accurately identify the steady-state characteristics and root causes of abnormal fluctuations in the process. This causes the optimization direction to deviate from actual needs, making it difficult to fundamentally solve process bottlenecks and ultimately leading to resource imbalances and low task execution efficiency.

[0003] Meanwhile, existing technologies lack a systematic assessment of load changes across all nodes before implementing optimization solutions. They focus only on the direct load impact of individual nodes, neglecting the indirect load transmission effects caused by inter-node collaboration. This leads to situations where some nodes are overloaded while others are idle after optimization measures are implemented. Furthermore, adjustments for overload risks lack clear priority guidance and precise resource replacement matching mechanisms. The dynamic adjustment capability of optimization solutions is insufficient, making it difficult to adapt to complex changes in business processes. Ultimately, this affects the stability and sustainability of optimization results, and prevents the efficient and collaborative operation of business processes. Summary of the Invention

[0004] This invention provides a method and system for iterative optimization of process parameters based on multi-source data, the main purpose of which is to solve the problem of low accuracy in iterative optimization of process parameters based on multi-source data.

[0005] To achieve the above objectives, the present invention provides a method for iterative optimization of process parameters based on multi-source data, comprising:

[0006] Collect multi-source business data, resource association tables, and task time series sets for the target business process;

[0007] Based on the steady-state performance characteristics and abnormal fluctuation characteristics of the multi-source business data, the resource allocation adjustment range and the task items to be improved for the target business process are determined respectively, and a preliminary set of optimization schemes for the target business process is obtained.

[0008] Using the resource entities in the multi-source business data as objects, and based on the execution order of the task time sequence set and the dependency relationship in the resource association table, a resource collaboration graph of the target business process is constructed.

[0009] The preliminary optimization scheme set is mapped to the resource collaboration graph to obtain the full node load change set of the target business process;

[0010] By comparing the load increase value of the nodes in the full node load change set with the available resource capacity of the nodes, the resource overload risk set of the business process is identified.

[0011] The preliminary optimization scheme set is adjusted based on the resource overload risk set to obtain the collaborative optimization execution sequence of the target business process.

[0012] In a preferred embodiment, the collection of multi-source business data, resource association tables, and task time series sets of the target business process includes:

[0013] Separate multi-source business data from the material records and interaction data in the target business process;

[0014] By integrating the transaction trigger records and resource status change records in the interaction data, a resource association table for the target business process is obtained;

[0015] Extract the timestamps of task creation and completion events from the interaction data to obtain the task time sequence set of the target business process.

[0016] In a preferred embodiment, the step of determining the resource allocation adjustment range and the tasks to be improved for the target business process based on the steady-state performance characteristics and abnormal fluctuation characteristics in the multi-source business data, respectively, to obtain a preliminary optimization scheme set for the target business process, includes:

[0017] Identify the steady-state performance characteristics and abnormal fluctuation characteristics of the target business process from the multi-source business data;

[0018] The resource utilization range corresponding to the steady-state performance characteristics shall be used as the resource configuration adjustment range for the target business process;

[0019] Based on the timestamps of the abnormal fluctuation characteristics, the task items to be improved in the target business process are selected from the task time series set;

[0020] By associating the resource configuration adjustment range with the task items to be improved, a preliminary set of optimization solutions for the target business process is directly generated.

[0021] In a preferred embodiment, separating the steady-state performance characteristics and abnormal fluctuation characteristics of the target business process from the multi-source business data includes:

[0022] Trend separation is performed on the multi-source business data to obtain the steady-state performance characteristics of the target business process;

[0023] The residual sequence of the target business process is generated by subtracting the multi-source business data from the steady-state performance characteristics point by point.

[0024] The standard deviation of the residual sequence is used as the discrimination threshold of the target business process, and points in the residual sequence that exceed the discrimination threshold are identified as abnormal fluctuation features.

[0025] In a preferred embodiment, the step of constructing a resource collaboration graph for the target business process, using the resource entities in the multi-source business data as objects and based on the execution order of the task time sequence set and the dependencies in the resource association table, includes:

[0026] By integrating the material data, equipment data, and personnel data from the multi-source business data, the node set of the target business process is obtained;

[0027] The execution order of the task time sequence is connected to the node pairs corresponding to the task time sequence in the node set to obtain the time sequence edge set of the target business process;

[0028] Based on the dependencies in the resource association table, connect the node pairs in the node set that correspond to the resource association table to obtain the dependency edge set of the target business process;

[0029] By integrating the node set, the temporal edge set, and the dependency edge set, a resource collaboration graph of the target business process is obtained.

[0030] In a preferred embodiment, mapping the preliminary optimization scheme set to the resource collaboration graph to obtain the full-node load change set of the target business process includes:

[0031] The optimized configuration parameters in the preliminary optimization scheme set are used as the expected load increment of the target business process;

[0032] The expected load increment is weighted and allocated based on the relationship strength of the edges in the resource collaboration graph to obtain the indirect load increment of the target business process;

[0033] The total load increase of the target business process is obtained by summing the expected load increase and the indirect load increase.

[0034] By summing up the total load increase, the full node load change set of the target business process is obtained.

[0035] In a preferred embodiment, the step of comparing the load increase value of the nodes in the full node load change set with the available resource capacity of the nodes to identify the resource overload risk set of the business process includes:

[0036] Obtain the available resource capacity and load rate of the corresponding node in the resource body from the resource association table;

[0037] Based on the total load increase of the node, a relative load assessment is performed on the available resource capacity and load rate to obtain the overload risk index of the target business process;

[0038] The nodes whose overload risk index exceeds the critical threshold are defined as the resource overload risk set of the target business process.

[0039] In a preferred embodiment, the overload risk index is calculated using the following formula:

[0040]

[0041] The overload risk index is... For node indexing, Indexing neighboring nodes, For nodes Total load increase value, For nodes Available resource capacity, For nodes load rate, This is the dynamic buffer coefficient. The collaboration impact coefficient. For nodes The set of neighboring nodes, To the neighbor node Pointing to node The strength weight of the edge relationship, Neighboring nodes The load rate.

[0042] In a preferred embodiment, adjusting the preliminary optimization scheme set according to the resource overload risk set to obtain the collaborative optimization execution sequence of the target business process includes:

[0043] The nodes of the resource overload risk concentration are sorted in descending order according to the overload risk index to obtain the risk priority handling order of the target business process.

[0044] Based on the aforementioned risk priority order, the preliminary optimization schemes are selected to target adjustment instructions for the nodes with concentrated resource overload risks.

[0045] Based on the resource collaboration graph, the adjustment instruction is matched with the replacement resource entity in the target business process;

[0046] Based on the adjustment of the preliminary optimization scheme set by replacing the resource subject, the collaborative optimization execution sequence of the target business process is obtained.

[0047] To address the above problems, the present invention also provides a process parameter iterative optimization system based on multi-source data, the system comprising:

[0048] The data acquisition module collects multi-source business data, resource association tables, and task time series sets from the target business process;

[0049] The preliminary optimization scheme set module, based on the steady-state performance characteristics and abnormal fluctuation characteristics in the multi-source business data, determines the resource allocation adjustment range and the task items to be improved for the target business process, and obtains the preliminary optimization scheme set for the target business process.

[0050] The resource collaboration graph module takes the resource entities in the multi-source business data as objects and constructs the resource collaboration graph of the target business process according to the execution order of the task time sequence set and the dependency relationship in the resource association table.

[0051] The full node load change set module maps the preliminary optimization scheme set to the resource collaboration graph to obtain the full node load change set of the target business process;

[0052] The resource overload risk set module compares the load increase value of the nodes in the full node load change set with the available resource capacity of the nodes to identify the resource overload risk set of the business process.

[0053] The collaborative optimization execution sequence module adjusts the preliminary optimization scheme set according to the resource overload risk set to obtain the collaborative optimization execution sequence of the target business process.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] 1. This technology comprehensively collects multi-source business data, resource association tables, and task time series sets from the target business process, accurately identifies steady-state performance characteristics and abnormal fluctuation characteristics, and constructs a complete resource collaboration graph, achieving comprehensive control over the operational status of the business process. Based on the load assessment and risk identification mechanism of the resource collaboration graph, it can accurately quantify the load changes and overload risks of each node. The generated preliminary optimization scheme set and collaborative optimization execution sequence closely match actual business needs, effectively improving the accuracy and relevance of process parameter optimization and ensuring that resource allocation is always adapted to the operational rhythm of the business process.

[0056] 2. This technology, by establishing a risk-priority handling order and a precise resource replacement matching mechanism, enables dynamic adjustment and efficient implementation of optimization solutions. This ensures load balancing across all nodes in the business process while fully leveraging resource synergy. The systematic optimization process not only quickly identifies tasks requiring improvement and resource allocation adjustment ranges but also forms a logically coherent and directly executable collaborative optimization sequence. This significantly improves the operational efficiency and stability of the business process, driving it to achieve the dual goals of maximizing resource utilization and refining risk management. Attached Figure Description

[0057] Figure 1 A flowchart illustrating an embodiment of the process parameter iterative optimization method based on multi-source data provided by the present invention;

[0058] Figure 2 A functional block diagram of a process parameter iterative optimization system based on multi-source data provided in an embodiment of the present invention;

[0059] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0060] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0061] This application provides a method for iterative optimization of process parameters based on multi-source data. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for iterative optimization of process parameters based on multi-source data can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0062] Reference Figure 1 The diagram shown is a flowchart illustrating an iterative optimization method for process parameters based on multi-source data according to an embodiment of the present invention. In this embodiment, the iterative optimization method for process parameters based on multi-source data includes:

[0063] In this embodiment of the invention, the process of collecting multi-source business data, resource association tables, and task time series sets of the target business process is specifically used for:

[0064] Separate multi-source business data from the material records and interaction data in the target business process;

[0065] By integrating the transaction trigger records and resource status change records in the interaction data, a resource association table for the target business process is obtained;

[0066] Extract the timestamps of task creation and completion events from the interaction data to obtain the task time sequence set of the target business process.

[0067] Specifically, basic information such as the type, quantity, specifications, and flow path of materials are extracted from the material records of the target business process. At the same time, equipment operation records, personnel collaboration records, information transmission records, and other content related to business execution are filtered from the interaction data.

[0068] Specifically, each transaction trigger record in the interaction data is reviewed to clarify the triggering conditions, triggering subject, triggering time, and corresponding business operation content of each transaction. At the same time, resource status change records are compiled in detail to accurately record the status changes of various resources during business execution, including specific information on all status transitions such as resources changing from idle to occupied, from normal operation to abnormal shutdown.

[0069] Specifically, it comprehensively traverses all task-related records in the interaction data, accurately locates the creation and completion events of each task, and extracts the specific timestamp of each task's creation event to determine the exact moment when the task preparation began.

[0070] Furthermore, by classifying and organizing the data, these information from different sources and formats are integrated and categorized, and multi-source business data covering multiple dimensions such as materials, equipment, personnel, and information are directly obtained for the target business process.

[0071] Furthermore, the transaction trigger records are matched one by one with the corresponding resource status change records, and the target business process resource association table is formed by association and integration to fully record the relationship between transactions and resource status.

[0072] Furthermore, for the completion event of each task, the specific timestamp at the end is extracted to determine the accurate time of the task's final completion. The creation timestamps and completion timestamps of all tasks are arranged in an orderly manner according to the logical association of task execution to form a target business process task sequence set that can reflect the chronological order of tasks.

[0073] In summary, the separated multi-source business data comprehensively integrates core information from different dimensions such as materials, equipment, and personnel in the target business process. It breaks down the information silos where various types of data are stored separately and are not related to each other, and enables key data in the business process to be presented in a centralized and systematic manner. This provides complete and accurate data support for subsequent management work such as resource allocation optimization and process bottleneck identification, and significantly improves the scientific nature and pertinence of business management decisions.

[0074] In summary, the resulting resource association table clearly presents the correspondence between transaction triggers and resource status changes, allowing business managers to intuitively grasp the specific impact of each transaction execution on resource status, clarify the flow logic and usage trajectory of resources in the business process, effectively avoid the coexistence of resource idleness and shortage, provide a direct basis for the rational allocation of resources and optimization of resource utilization efficiency, and help achieve dynamic balance and maximization of benefits in resource allocation.

[0075] In summary, the constructed task time series accurately records the time nodes from creation to completion of each task, fully presenting the order of task execution and the time connection relationship. It can clearly expose problems such as time delays and process bottlenecks during task execution, helping business managers quickly locate efficiency bottlenecks in the process, providing data support for optimizing task scheduling, shortening business cycles, improving the smoothness of cross-departmental collaboration, and promoting the efficient operation of business processes.

[0076] In this embodiment of the invention, when determining the resource allocation adjustment range and the tasks to be improved for the target business process based on the steady-state performance characteristics and abnormal fluctuation characteristics in the multi-source business data, and obtaining the preliminary optimization scheme set for the target business process, it is specifically used for:

[0077] Identify the steady-state performance characteristics and abnormal fluctuation characteristics of the target business process from the multi-source business data;

[0078] The resource utilization range corresponding to the steady-state performance characteristics shall be used as the resource configuration adjustment range for the target business process;

[0079] Based on the timestamps of the abnormal fluctuation characteristics, the task items to be improved in the target business process are selected from the task time series set;

[0080] By associating the resource configuration adjustment range with the task items to be improved, a preliminary set of optimization solutions for the target business process is directly generated.

[0081] Specifically, by observing the trend of data over a longer period of time, the data segments that remain relatively stable without significant fluctuations are separated out. The characteristics of these data are the steady-state performance characteristics of the target business process.

[0082] Specifically, for the identified steady-state performance characteristics, a comprehensive statistical analysis of the usage of various resources corresponding to them during the operation of the business process is conducted, and detailed information such as the usage amount and duration of each resource in the steady-state state is recorded.

[0083] Specifically, the occurrence timestamps corresponding to each identified abnormal fluctuation feature are extracted one by one, and these timestamps are compared with the creation timestamps and completion timestamps of all tasks in the task time series to identify all tasks in the task time series whose creation timestamps are before the abnormal fluctuation feature timestamps and whose completion timestamps are after the abnormal fluctuation feature timestamps.

[0084] Specifically, the determined resource allocation adjustment range is associated one-to-one with the selected task items to be improved. For each task item to be improved, the various resources required during its operation are identified, and the corresponding resource allocation adjustment range is used as the adjustment standard for the resource use of that task item.

[0085] Furthermore, the multi-source business data is compared point by point with the established steady-state performance characteristics. The difference between each data point and the corresponding steady-state characteristic point is calculated to form a complete residual sequence. Then, all the data in the residual sequence are statistically analyzed to determine the dispersion index of the sequence. This index is used as a discrimination threshold to identify all data points in the residual sequence that exceed the threshold. The characteristics corresponding to these data points that exceed the threshold are the abnormal fluctuation characteristics of the target business process.

[0086] Furthermore, by summarizing and analyzing this information, the minimum and maximum utilization rates of each resource under steady-state performance characteristics are determined. The range formed by these two utilization rate values ​​is the resource allocation adjustment range for the target business process.

[0087] Furthermore, tasks whose creation timestamps, completion timestamps, and abnormal fluctuation feature timestamps overlap are the target business process improvement tasks selected from the task time sequence set.

[0088] Furthermore, it is clearly stipulated that after optimization, the utilization rate of various resources should be controlled within the corresponding resource configuration adjustment range for the task items to be improved. Through this association method, a preliminary optimization scheme set of target business processes containing all task items to be improved and their corresponding resource configuration adjustment requirements can be directly integrated.

[0089] In summary, accurately identifying the steady-state performance characteristics and abnormal fluctuation characteristics of target business processes clearly defines the baseline state of normal business process operation and abnormal situations that deviate from the standard. This allows business managers to intuitively distinguish the reasonable range and risk range of process operation, avoiding management misconduct caused by ambiguous judgment of process status, and providing clear problem guidance and judgment basis for subsequent targeted optimization.

[0090] In summary, defining the resource utilization range corresponding to steady-state performance characteristics as the resource allocation adjustment interval clarifies the reasonable usage boundaries of various resources when the business process is running stably. This provides a quantitative standard that can be directly followed for resource allocation, while avoiding waste caused by overly lenient resource allocation or process blockage caused by overly tight allocation. It ensures that resource allocation always meets the steady-state needs of the business process and improves the accuracy and rationality of resource utilization.

[0091] In summary, the timestamp-based filtering of tasks to be improved based on abnormal fluctuation characteristics enables precise binding of abnormal issues with corresponding tasks. This allows business managers to quickly identify the specific task links causing process fluctuations, avoiding efficiency losses caused by blind investigations. At the same time, it clarifies the key targets for optimization work, enabling subsequent optimization actions to directly address the core issues and improve the pertinence and effectiveness of process optimization.

[0092] In summary, by associating the resource allocation adjustment range with the tasks to be improved, a preliminary set of optimization solutions is generated. Specific resource allocation optimization standards are directly matched to each task to be improved, forming a feasible and executable optimization direction. This avoids the problem of optimization solutions being out of touch with actual business needs. At the same time, it integrates scattered optimization points to form a systematic optimization framework, laying the foundation for subsequent in-depth optimization based on resource collaboration relationships, and promoting the orderly and efficient development of business process optimization.

[0093] In this embodiment of the invention, the step of separating the steady-state performance characteristics and abnormal fluctuation characteristics of the target business process from the multi-source business data is specifically used for:

[0094] Trend separation is performed on the multi-source business data to obtain the steady-state performance characteristics of the target business process;

[0095] The residual sequence of the target business process is generated by subtracting the multi-source business data from the steady-state performance characteristics point by point.

[0096] The standard deviation of the residual sequence is used as the discrimination threshold of the target business process, and points in the residual sequence that exceed the discrimination threshold are identified as abnormal fluctuation features.

[0097] Specifically, multi-source business data is integrated according to the time dimension, and the data is arranged in the order of business execution. By observing the changes in the data over a continuous period of time, short-term, sudden, and irregular fluctuations are eliminated.

[0098] Specifically, the specific data value corresponding to each time node in the multi-source business data is compared one by one with the stable data value corresponding to the same time node in the steady-state performance characteristics. The difference between the two at each time node is calculated by subtracting the corresponding data point value in the steady-state performance characteristics from the data point value in the multi-source business data.

[0099] Specifically, first, statistical analysis is performed on all differences in the residual sequence to calculate the average level of all differences. Then, each difference is compared with the average level to determine the degree of deviation of each difference from the average level.

[0100] Furthermore, the core data that remains relatively stable and changes very little over a longer period of time is retained. The stable operating characteristics exhibited by the retained data after screening are the steady-state performance characteristics of the target business process.

[0101] Furthermore, all these differences are arranged sequentially in chronological order to form a complete sequence. This sequence, composed of the differences at each time point, is the residual sequence of the target business process.

[0102] Furthermore, by summarizing and calculating all deviations, the overall dispersion of the residual sequence is determined. The quantification result of this dispersion is the discrimination threshold of the target business process. Then, each data point in the residual sequence is checked one by one, and all data points whose values ​​exceed the discrimination threshold are selected. The characteristics corresponding to these selected data points are the abnormal fluctuation characteristics of the target business process.

[0103] In summary, by performing trend separation on multi-source business data, the steady-state performance characteristics of the target business process are accurately extracted. The core data patterns and operating benchmarks of the business process under normal operating conditions are clearly defined, enabling business managers to clearly understand the reference standards for stable process operation. This avoids misjudging the process status due to the lack of clear benchmarks and provides a unified and reliable basis for comparison for all subsequent optimization work, ensuring that the optimization direction always revolves around the core requirement of stable process operation.

[0104] In summary, by generating residual sequences through point-by-point subtraction between multi-source business data and steady-state performance characteristics, the system quantifies the degree of deviation of multi-source business data from steady-state performance characteristics. It presents the fluctuations at the data level in an intuitive sequence form, fully preserving the deviation details of the data at each time point, avoiding general judgments on data fluctuations, and enabling business managers to accurately capture subtle differences in data changes. This provides comprehensive and detailed data support for the subsequent accurate identification of abnormal fluctuations.

[0105] In summary, using the standard deviation of the residual sequence as the discrimination threshold sets an objective and unified quantitative standard for identifying abnormal fluctuation characteristics, completely eliminating the bias caused by subjective judgment, ensuring that the identification of all abnormal fluctuations is based on consistent rule execution, and accurately identifying points in the residual sequence that exceed the threshold as abnormal fluctuation characteristics. This achieves precise location and screening of abnormal situations, avoiding the problem of missing key anomalies or misjudging normal fluctuations, and allowing abnormal nodes in the business process to be clearly and accurately locked, providing clear targets for subsequent targeted optimization.

[0106] In this embodiment of the invention, when constructing the resource collaboration graph of the target business process based on the resource entities in the multi-source business data and the dependencies in the resource association table according to the execution order of the task time sequence set, the specific usage is as follows:

[0107] By integrating the material data, equipment data, and personnel data from the multi-source business data, the node set of the target business process is obtained;

[0108] The execution order of the task time sequence is connected to the node pairs corresponding to the task time sequence in the node set to obtain the time sequence edge set of the target business process;

[0109] Based on the dependencies in the resource association table, connect the node pairs in the node set that correspond to the resource association table to obtain the dependency edge set of the target business process;

[0110] By integrating the node set, the temporal edge set, and the dependency edge set, a resource collaboration graph of the target business process is obtained.

[0111] Specifically, the material data in the multi-source business data is sorted out, and the core information such as the unique identifier, specifications, and storage location of each material is identified and classified separately. At the same time, the key contents such as the number, functional parameters, and operating status of various equipment are extracted from the equipment data, as well as the detailed information such as the identity information, job responsibilities, and skills and qualifications of the personnel involved in the business process from the personnel data.

[0112] Specifically, the task sequence set is fully analyzed to clarify the execution order of each task and the participating nodes corresponding to each task. The start and end nodes related to each task are accurately identified from the node set, and the tasks are executed in the order specified in the task sequence set.

[0113] Specifically, extract all resource dependencies recorded in the resource association table, including various dependencies such as material supply dependencies, equipment usage dependencies, and personnel cooperation dependencies. Based on these dependencies, determine the corresponding associated node pairs and find each pair of nodes with dependencies from the node set.

[0114] Specifically, the constructed node set, temporal edge set, and dependency edge set are systematically integrated. First, all nodes in the node set are reasonably arranged in the graph according to the actual operation scenario of the business process. Then, each edge in the temporal edge set is accurately drawn in the graph according to the corresponding node connection relationship.

[0115] Furthermore, these separately categorized material data, equipment data, and personnel data are fully integrated, duplicate information is removed, and related information is added to form a set containing all core elements related to materials, equipment, and personnel. This set is the node set of the target business process, and each material, equipment, and personnel becomes an independent node in the node set.

[0116] Furthermore, the starting and ending nodes of each task are connected sequentially with lines. Each connection accurately corresponds to the sequential collaboration logic between nodes during task execution. All these node connection relationships formed according to the task execution order are summarized and organized to form the temporal edge set of the target business process.

[0117] Furthermore, each pair of dependent nodes is connected by lines, with the connection method strictly following the dependency direction and dependency type specified in the resource association table. All node connection relationships that reflect resource dependencies are centrally integrated to form the dependency edge set of the target business process.

[0118] Furthermore, each edge in the dependency edge set is also mapped to the corresponding node in the graph, ensuring that the drawing of temporal edges and dependency edges does not conflict and is clearly distinguishable. Through this integration method, a visual graph that can comprehensively display all nodes in the target business process, the temporal relationships and dependencies between nodes is formed. This graph is the resource collaboration graph of the target business process.

[0119] In summary, by integrating material data, equipment data, and personnel data from multiple business sources to form a node set, the scattered core resource elements in the business process are systematically integrated. This transforms the originally independent information of various resources into unified node identifiers, clearly presenting all resource entities involved in the business process. This avoids management blind spots caused by fragmented resource information and lays a complete and standardized data foundation for subsequent analysis of the relationships between resources, enabling business managers to fully control the resource composition in the process.

[0120] In summary, by connecting corresponding node pairs according to the execution order of the task time sequence set to obtain the time sequence edge set, the sequential collaboration logic of resource nodes during task execution is accurately restored. The abstract task process is transformed into an intuitive node connection relationship, clearly showing the resource flow path of each task from start to finish. This avoids collaboration confusion caused by unclear task order, helps business managers quickly clarify the time context of task execution, and provides a clear structural basis for optimizing task scheduling and improving process efficiency.

[0121] In summary, by connecting corresponding node pairs based on the dependency relationships in the resource association table to obtain the dependency edge set, the system presents the inherent relationships such as supply and demand dependencies and collaborative cooperation between resource nodes, clarifies the resource input source and output destination of each node, breaks the implicit state of resource dependencies, avoids resource shortages or idleness caused by unclear dependencies, enables business managers to accurately identify resource dependency bottlenecks, and provides key support for the rational allocation of resources and ensuring the smooth operation of processes.

[0122] In summary, by integrating node sets, temporal edge sets, and dependency edge sets to form a resource collaboration graph, the resource composition, task temporal relationships, and resource dependencies in the business process are visualized. This provides a panoramic view of the business process operation logic, making the originally complex and scattered process information intuitive and easy to understand. It avoids management decision-making delays caused by opaque process logic, helps business managers to grasp the process operation status from a global perspective, and provides a comprehensive and systematic visualization tool for subsequent load assessment, risk identification, and optimization scheme formulation.

[0123] In this embodiment of the invention, when mapping the preliminary optimization scheme set to the resource collaboration graph to obtain the full node load change set of the target business process, it is specifically used for:

[0124] The optimized configuration parameters in the preliminary optimization scheme set are used as the expected load increment of the target business process;

[0125] The expected load increment is weighted and allocated based on the relationship strength of the edges in the resource collaboration graph to obtain the indirect load increment of the target business process;

[0126] The total load increase of the target business process is obtained by summing the expected load increase and the indirect load increase.

[0127] By summing up the total load increase, the full node load change set of the target business process is obtained.

[0128] Specifically, each optimization item in the preliminary optimization plan set is broken down one by one, and the specific requirements for resource configuration adjustment are clarified. The specific configuration standards corresponding to these requirements are the optimization configuration parameters. These optimization configuration parameters are directly defined as the target business process after the optimization plan is executed.

[0129] Specifically, we comprehensively analyze the relationships among all edges in the resource collaboration graph, and assign a corresponding relationship strength level to each edge based on the actual situation such as the closeness of business collaboration, the frequency of data interaction, and the importance of dependencies between the nodes connected by the edge.

[0130] Specifically, for each node in the target business process, the expected load increment and indirect load increment are extracted respectively, and the expected load increment value and indirect load increment value of the same node are combined for calculation.

[0131] Specifically, collect the total load increase value of all nodes in the target business process, systematically organize these values, and classify and sort them according to node type, number, or business relationship logic.

[0132] Furthermore, the expected increase in load for each relevant node is the expected load increment of the target business process. This ensures that each optimized configuration parameter accurately corresponds to the expected load change of a specific node, without any mismatch or omission between parameters and nodes.

[0133] Furthermore, the relationship strength level directly reflects the degree of mutual influence between nodes. Based on these relationship strength levels, a corresponding weight value is assigned to each edge. The closer the collaboration and the greater the influence of the edge, the higher the weight value. Then, the expected load increment is distributed to the corresponding associated nodes according to the weight values ​​of each edge. These load increments distributed to the associated nodes are the indirect load increments of the target business process.

[0134] Furthermore, the total load increase of the node after the optimization plan is implemented is obtained by adding the values. This total load increase is the total load increase of the target business process. Each node has a unique total load increase, which accurately reflects the overall load change of a single node due to the implementation of the optimization plan.

[0135] Furthermore, the total load increase of all nodes after sorting is integrated into a complete set, which is the set of all node load changes for the target business process.

[0136] In summary, by directly using the optimized configuration parameters from the initial optimization scheme set as the expected load increment, the direct impact of optimization measures on the load of each node is clarified. This makes the load change effect of the optimization scheme quantifiable and traceable, avoiding the situation where the load change is unclear after the implementation of optimization measures. It provides clear basic data for subsequent accurate evaluation of the feasibility of optimization schemes, ensuring that the load evaluation work can be carried out in an orderly manner closely aligned with the optimization objectives.

[0137] In summary, the indirect load increment is obtained by weighting the expected load increment based on the relationship strength of the edges in the resource collaboration graph. This fully considers the transmission effect of the collaborative association between nodes on load changes, breaks the limitation of only focusing on the direct load of a single node and ignoring the indirect impact of related nodes, accurately captures the chain load changes caused by optimization measures, avoids load assessment bias caused by omitting indirect load, and makes the assessment of load changes more comprehensive and more in line with the actual collaborative scenarios of business processes.

[0138] In summary, the total load increase is obtained by summing the expected load increment and the indirect load increment. This comprehensively integrates the direct and indirect load impacts of optimization measures on nodes, forming a complete quantitative result of node load changes. This avoids the one-sidedness of assessment caused by considering only a certain type of load, accurately reflects the overall load change of each node after the implementation of the optimization scheme, and provides accurate and comprehensive numerical basis for subsequent judgment on whether there is an overload risk of nodes.

[0139] In summary, the total load increase is aggregated to obtain a set of load changes across all nodes. The system integrates load change data from all nodes in the business process, achieving a centralized and systematic presentation of load change information. This avoids the risk identification omissions caused by scattered load data, allowing business managers to fully grasp the load distribution changes of the entire business process after the implementation of optimization solutions. It provides complete and unified data support for subsequent systematic overload risk identification and optimization solution adjustments, ensuring that optimization work can take into account the load balance of all nodes.

[0140] In this embodiment of the invention, when comparing the load increase value of the nodes in the full node load change set with the available resource capacity of the nodes to identify the resource overload risk set of the business process, it is specifically used for:

[0141] Obtain the available resource capacity and load rate of the corresponding node in the resource body from the resource association table;

[0142] Based on the total load increase of the node, a relative load assessment is performed on the available resource capacity and load rate to obtain the overload risk index of the target business process;

[0143] The nodes whose overload risk index exceeds the critical threshold are defined as the resource overload risk set of the target business process.

[0144] Specifically, we conduct in-depth analysis of all resource-related information recorded in the resource association table, clarify the resource type corresponding to each node in the resource entity, including materials, equipment, personnel, etc., and accurately locate the corresponding entry in the resource association table based on the node's unique identifier.

[0145] Specifically, by combining the increase in the total load of the node, we analyze how this increase affects the available resource capacity of the node, determine whether the increase in the total load will lead to insufficient available resource capacity, and at the same time, by referring to the current load rate of the node, we evaluate the trend of the load rate after the total load increase is added.

[0146] Specifically, based on the actual operational needs of the target business process, resource allocation standards, and historical operational data, a unified overload risk threshold is determined. This threshold defines the maximum overload risk limit that a node can withstand.

[0147] Furthermore, the total amount of resources that can be allocated to each node is extracted. This total amount is the available resource capacity of the node. At the same time, the ratio of the amount of resources currently used to the available resource capacity of each node is extracted. This ratio is the load rate of the node.

[0148] Furthermore, by comprehensively considering the collaborative relationship between nodes and their neighboring nodes, as well as the indirect impacts that may arise during the collaboration process, these factors are incorporated into the overall assessment. Through a comprehensive weighing, a quantitative result is obtained that reflects the possibility of a node's load exceeding the safe range. This quantitative result is the overload risk index of the target business process. Each node corresponds to a unique overload risk index, which intuitively reflects the degree of overload risk faced by the node.

[0149] Furthermore, the overload risk index of each node is compared with the critical threshold one by one. All nodes with overload risk index values ​​exceeding the critical threshold are screened out. The set of these screened nodes is the resource overload risk set of the target business process.

[0150] In summary, accurately obtaining the available resource capacity and load rate of the corresponding node from the resource association table directly provides core data on the node's current resource supply capacity and actual occupancy. This avoids subsequent assessment biases caused by missing or inaccurate basic resource information, providing a true and reliable basis for load assessment. It enables business managers to clearly understand the resource usage status of each node, laying a solid data foundation for subsequent risk assessment.

[0151] In summary, the system performs a relative load assessment of available resource capacity and load rate based on the total load increase of nodes, and obtains an overload risk index. It comprehensively considers the load changes brought about by the existing resource status of nodes and optimization schemes, and realizes a quantitative assessment of node overload risk. This breaks the limitation of relying on a single data point to judge risk, accurately reflects the actual load pressure of each node after optimization, avoids vague judgments of risk level, and allows business managers to intuitively compare the risk levels of different nodes.

[0152] In summary, identifying nodes whose overload risk index exceeds the critical threshold as the resource overload risk set enables precise screening and centralized classification of overload risk nodes. It clarifies the key nodes that need to be focused on and adjusted after the implementation of the optimization plan, avoiding the inefficient behavior of blindly checking for risks among numerous nodes. This makes the risk management objectives clearer and the direction more defined, providing precise targeted guidance for subsequent targeted adjustments to the initial optimization plan, and ensuring that resource allocation optimization can directly address the core risks.

[0153] In this embodiment of the invention, the formula for calculating the overload risk index is specifically used for:

[0154]

[0155] The overload risk index is... For node indexing, Indexing neighboring nodes, For nodes Total load increase value, For nodes Available resource capacity, For nodes load rate, This is the dynamic buffer coefficient. The coefficient representing the impact of collaboration. For nodes The set of neighboring nodes, To the neighbor node Pointing to node The strength weight of the edge relationship, For neighboring nodes The load rate.

[0156] Specifically, the node index and neighbor node index are determined based on the distribution and relationships of nodes in the resource collaboration graph. Each node corresponds to a unique index identifier, while the neighbor node index is the identifier of the node directly connected to it through an edge in the resource collaboration graph. The total load increase value comes from the total node load change set and is the node load increase value obtained by summing the expected load increase and the indirect load increase. Available resource capacity and load rate are extracted from the resource association table, corresponding to the total amount of resources currently available to the node and the ratio of used resources to available resource capacity, respectively. The dynamic buffer coefficient is determined based on the resource redundancy configuration standards of the target business process, the load fluctuations in historical operation, and the anti-interference capability of resources, and is used to buffer the impact of load changes on node operation. The collaboration impact coefficient is set according to the degree of collaboration between nodes in the resource collaboration graph, the frequency of data interaction, and the importance of dependencies, reflecting the degree of influence of neighbor nodes on the current node. The neighbor node set is the set of all nodes directly connected to the current node through an edge in the resource collaboration graph, and is directly determined by the node connection relationships in the resource collaboration graph. The edge relationship strength weight is assigned based on actual business scenario factors such as the frequency of business collaboration, data transmission volume, and degree of dependence between nodes connected by the edge in the resource collaboration graph. It is used to quantify the influence weight of neighboring nodes on the current node. The load rate of neighboring nodes is also obtained from the resource association table, which is the ratio of the amount of resources used by the neighboring node to its own available resource capacity.

[0157] Furthermore, this formula is used to comprehensively assess the degree of overload risk faced by each node in the target business process after implementing the initial optimization plan. By combining the node's own load changes with the impact of neighboring nodes, it comprehensively reflects whether the node's load may exceed its carrying capacity, providing accurate quantitative basis for identifying resource overload risk sets.

[0158] In general, the greater the increase in a node's total load, the smaller the difference between its available resource capacity and current load rate, and the smaller its dynamic buffer coefficient, the greater the node's overload risk index and the higher the risk of overload. Conversely, the greater the increase in the total load of neighboring nodes, the higher the strength weight of the edges between them and the current node, and the greater the cooperation influence coefficient, the smaller the current node's available resource capacity, and the greater its overload risk index and the higher the risk of overload. Conversely, the smaller the increase in the total load of neighboring nodes, the higher the strength weight of the edges between them and the current node, and the greater the cooperation influence coefficient, the lower the current node's available resource capacity and the higher its overload risk index and the safer the node's operation.

[0159] In this embodiment of the invention, when adjusting the preliminary optimization scheme set according to the resource overload risk set to obtain the collaborative optimization execution sequence of the target business process, it is specifically used for:

[0160] The nodes of the resource overload risk concentration are sorted in descending order according to the overload risk index to obtain the risk priority handling order of the target business process.

[0161] Based on the aforementioned risk priority order, the preliminary optimization schemes are selected to target adjustment instructions for the nodes with concentrated resource overload risks.

[0162] Based on the resource collaboration graph, the adjustment instruction is matched with the replacement resource entity in the target business process;

[0163] Based on the adjustment of the preliminary optimization scheme set by replacing the resource subject, the collaborative optimization execution sequence of the target business process is obtained.

[0164] Specifically, the overload risk index corresponding to each node in the resource overload risk set is extracted one by one, the specific value of each index is determined, and all nodes are arranged in descending order of value.

[0165] Specifically, based on the priority of risk handling, all adjustment instructions in the preliminary optimization scheme set are checked one by one, the target node corresponding to each adjustment instruction is identified, and all adjustment instructions whose target nodes belong to the resource overload risk set are selected.

[0166] Specifically, we conduct in-depth analysis of the relationships, functional attributes, and resource configurations of each node in the resource collaboration graph, clarify the core business needs and resource usage requirements of each overload risk node, and based on the optimization goals and resource adjustment directions for the overload risk node in the adjustment instructions, we search for nodes in the resource collaboration graph that have the same or similar functions, sufficient available resource capacity, and a reasonable collaboration path with the original node.

[0167] Specifically, the matched replacement resource entities are mapped one by one to the corresponding adjustment instructions, replacing the overload risk nodes pointed to in the original adjustment instructions, redefining the execution entity of each adjustment instruction, and adjusting the resource configuration details in the adjustment instructions according to the available resource capacity, load rate, and cooperation relationship with other nodes of the replacement resource entities.

[0168] Furthermore, the node sequence formed after orderly arrangement is the risk priority order of the target business process. This order clarifies the order in which overload risks need to be addressed first. The higher the overload risk index, the earlier the node is in the sequence, and the more priority it receives optimization and adjustment resources.

[0169] Furthermore, during the screening process, the nodes in the risk priority order are strictly compared, and irrelevant instructions that do not belong to the resource overload risk set are excluded. The selected adjustment instructions are then organized according to the node order of the risk priority order to form a set of adjustment instructions that matches the risk handling order.

[0170] Furthermore, these nodes are replacement resource entities that can replace overload risk nodes to perform corresponding business tasks. During the matching process, it is ensured that the replacement resource entities can fully meet the execution requirements of the adjustment instructions, while taking into account the collaboration and compatibility with other nodes, so as to avoid new resource conflicts or business interruptions caused by replacement.

[0171] Furthermore, all adjusted instructions are integrated and sorted according to risk priority and business process execution logic to form a logically coherent and directly executable instruction sequence. This sequence is the collaborative optimization execution sequence of the target business process.

[0172] In summary, sorting by overload risk index in descending order yields the priority order for risk handling, clearly defining the emergency handling level of different overload risk nodes. This allows business managers to quickly focus on high-risk nodes, avoiding the dispersion of resources that would prevent high-risk issues from being addressed in a prioritized manner. It ensures that optimization resources are tilted towards the nodes with the highest risk level, improving the pertinence and efficiency of risk management, and setting a clear priority guide for subsequent optimization work.

[0173] In summary, by prioritizing risk handling, adjustment instructions targeting overload risk nodes are selected, accurately identifying the core instructions that need optimization and adjustment, eliminating redundant instructions unrelated to risk nodes, avoiding blind and scattered optimization actions, and allowing optimization work to focus on solving key risk issues. At the same time, it ensures that adjustment instructions are consistent with risk handling priorities, providing focused and efficient instruction support for subsequent matching of replacement resources and adjustment plans.

[0174] In summary, the resource collaboration graph is used to match and replace resource entities for adjustment instructions. It makes full use of the relationships, functional attributes and resource status information of the nodes in the graph to ensure that the matched replacement resource entities can fully adapt to the execution requirements of the adjustment instructions and maintain good collaboration compatibility with other nodes. This avoids business interruption or new resource conflicts caused by improper replacement resource selection, and achieves the accuracy and feasibility of resource replacement, providing reliable resource guarantee for the implementation of optimization solutions.

[0175] In summary, the collaborative optimization execution sequence is obtained by adjusting the initial set of optimization schemes based on resource replacement. The scattered adjustment instructions are integrated into a logically coherent and directly executable systematic scheme, which not only solves the resource overload problem in the original scheme, but also takes into account the collaborative needs between nodes. It avoids the process disorder caused by the fragmentation of optimization measures and forms a complete optimization path that takes into account both risk control and efficiency improvement. This ensures that the target business process can achieve efficient collaborative operation of each node while avoiding overload risks.

[0176] Compared with the prior art, the present invention has the following beneficial effects:

[0177] 1. This technology comprehensively collects multi-source business data, resource association tables, and task time series sets from the target business process, accurately identifies steady-state performance characteristics and abnormal fluctuation characteristics, and constructs a complete resource collaboration graph, achieving comprehensive control over the operational status of the business process. Based on the load assessment and risk identification mechanism of the resource collaboration graph, it can accurately quantify the load changes and overload risks of each node. The generated preliminary optimization scheme set and collaborative optimization execution sequence closely match actual business needs, effectively improving the accuracy and relevance of process parameter optimization and ensuring that resource allocation is always adapted to the operational rhythm of the business process.

[0178] 2. This technology, by establishing a risk-priority handling order and a precise resource replacement matching mechanism, enables dynamic adjustment and efficient implementation of optimization solutions. This ensures load balancing across all nodes in the business process while fully leveraging resource synergy. The systematic optimization process not only quickly identifies tasks requiring improvement and resource allocation adjustment ranges but also forms a logically coherent and directly executable collaborative optimization sequence. This significantly improves the operational efficiency and stability of the business process, driving it to achieve the dual goals of maximizing resource utilization and refining risk management.

[0179] like Figure 2 The diagram shown is a functional block diagram of a process parameter iterative optimization system based on multi-source data provided in an embodiment of the present invention.

[0180] The process parameter iterative optimization system 100 based on multi-source data described in this invention can be installed in an electronic device. Depending on the functions implemented, the process parameter iterative optimization system 100 based on multi-source data may include a data acquisition module 101, a preliminary optimization scheme set module 102, a resource collaboration graph module 103, a full node load change set module 104, a resource overload risk set module 105, and a collaborative optimization execution sequence module 106. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0181] In this embodiment, the functions of each module / unit are as follows:

[0182] The data acquisition module collects multi-source business data, resource association tables, and task time series sets from the target business process;

[0183] The preliminary optimization scheme set module, based on the steady-state performance characteristics and abnormal fluctuation characteristics in the multi-source business data, determines the resource allocation adjustment range and the task items to be improved for the target business process, and obtains the preliminary optimization scheme set for the target business process.

[0184] The resource collaboration graph module takes the resource entities in the multi-source business data as objects and constructs the resource collaboration graph of the target business process according to the execution order of the task time sequence set and the dependency relationship in the resource association table.

[0185] The full node load change set module maps the preliminary optimization scheme set to the resource collaboration graph to obtain the full node load change set of the target business process;

[0186] The resource overload risk set module compares the load increase value of the nodes in the full node load change set with the available resource capacity of the nodes to identify the resource overload risk set of the business process.

[0187] The collaborative optimization execution sequence module adjusts the preliminary optimization scheme set according to the resource overload risk set to obtain the collaborative optimization execution sequence of the target business process.

[0188] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0189] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0190] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0191] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0192] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0193] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for iterative optimization of process parameters based on multi-source data, characterized in that, The method includes: Collect multi-source business data, resource association tables, and task time series sets for the target business process; Based on the steady-state performance characteristics and abnormal fluctuation characteristics of the multi-source business data, the resource allocation adjustment range and the task items to be improved for the target business process are determined respectively, and a preliminary set of optimization schemes for the target business process is obtained. Using the resource entities in the multi-source business data as objects, and based on the execution order of the task time sequence set and the dependency relationship in the resource association table, a resource collaboration graph of the target business process is constructed. The preliminary optimization scheme set is mapped to the resource collaboration graph to obtain the full node load change set of the target business process; By comparing the load increase value of the nodes in the full node load change set with the available resource capacity of the nodes, the resource overload risk set of the business process is identified, including: Obtain the available resource capacity and load rate of the corresponding node in the resource body from the resource association table; Based on the total load increase of the nodes, a relative load assessment is performed on the available resource capacity and load rate to obtain the overload risk index of the target business process. The overload risk index is calculated using the following formula: The overload risk index is... For node indexing, Indexing neighboring nodes, For nodes Total load increase value, For nodes Available resource capacity, For nodes load rate, This is the dynamic buffer coefficient. The collaboration impact coefficient. For nodes The set of neighboring nodes, To the neighbor node Pointing to node The strength weight of the edge relationship, Neighboring nodes Load rate; The nodes whose overload risk index exceeds the critical threshold are defined as the resource overload risk set of the target business process; The preliminary optimization scheme set is adjusted based on the resource overload risk set to obtain the collaborative optimization execution sequence of the target business process.

2. The process parameter iterative optimization method based on multi-source data as described in claim 1, characterized in that, The multi-source business data, resource association tables, and task time series sets of the target business process collected include: Separate multi-source business data from the material records and interaction data in the target business process; By integrating the transaction trigger records and resource status change records in the interaction data, a resource association table for the target business process is obtained; Extract the timestamps of task creation and completion events from the interaction data to obtain the task time sequence set of the target business process.

3. The process parameter iterative optimization method based on multi-source data as described in claim 1, characterized in that, Based on the steady-state performance characteristics and abnormal fluctuation characteristics of the multi-source business data, the resource allocation adjustment range and the tasks to be improved for the target business process are determined respectively, resulting in a preliminary set of optimization schemes for the target business process, including: Identify the steady-state performance characteristics and abnormal fluctuation characteristics of the target business process from the multi-source business data; The resource utilization range corresponding to the steady-state performance characteristics shall be used as the resource configuration adjustment range for the target business process; Based on the timestamps of the abnormal fluctuation characteristics, the task items to be improved in the target business process are selected from the task time series set; By associating the resource configuration adjustment range with the task items to be improved, a preliminary set of optimization solutions for the target business process is directly generated.

4. The process parameter iterative optimization method based on multi-source data as described in claim 3, characterized in that, The step of separating the steady-state performance characteristics and abnormal fluctuation characteristics of the target business process from the multi-source business data includes: Trend separation is performed on the multi-source business data to obtain the steady-state performance characteristics of the target business process; The residual sequence of the target business process is generated by subtracting the multi-source business data from the steady-state performance characteristics point by point. The standard deviation of the residual sequence is used as the discrimination threshold of the target business process, and points in the residual sequence that exceed the discrimination threshold are identified as abnormal fluctuation features.

5. The method for iterative optimization of process parameters based on multi-source data as described in claim 1, characterized in that, The step of constructing a resource collaboration graph for the target business process, using the resource entities in the multi-source business data as objects and based on the execution order of the task time sequence set and the dependencies in the resource association table, includes: By integrating the material data, equipment data, and personnel data from the multi-source business data, the node set of the target business process is obtained; The execution order of the task time sequence is connected to the node pairs corresponding to the task time sequence in the node set to obtain the time sequence edge set of the target business process; Based on the dependencies in the resource association table, connect the node pairs in the node set that correspond to the resource association table to obtain the dependency edge set of the target business process; By integrating the node set, the temporal edge set, and the dependency edge set, a resource collaboration graph of the target business process is obtained.

6. The method for iterative optimization of process parameters based on multi-source data as described in claim 1, characterized in that, The step of mapping the preliminary optimization scheme set to the resource collaboration graph to obtain the full node load change set of the target business process includes: The optimized configuration parameters in the preliminary optimization scheme set are used as the expected load increment of the target business process; The expected load increment is weighted and allocated based on the relationship strength of the edges in the resource collaboration graph to obtain the indirect load increment of the target business process; The total load increase of the target business process is obtained by summing the expected load increase and the indirect load increase. By summing up the total load increase, the full node load change set of the target business process is obtained.

7. The method for iterative optimization of process parameters based on multi-source data as described in claim 1, characterized in that, The step of adjusting the preliminary optimization scheme set according to the resource overload risk set to obtain the collaborative optimization execution sequence of the target business process includes: The nodes of the resource overload risk concentration are sorted in descending order according to the overload risk index to obtain the risk priority handling order of the target business process. Based on the aforementioned risk priority order, the preliminary optimization schemes are selected to target adjustment instructions for the nodes with concentrated resource overload risks. Based on the resource collaboration graph, the adjustment instruction is matched with the replacement resource entity in the target business process; Based on the adjustment of the preliminary optimization scheme set by replacing the resource subject, the collaborative optimization execution sequence of the target business process is obtained.

8. A process parameter iterative optimization system based on multi-source data, used to implement the process parameter iterative optimization method based on multi-source data as described in claim 1, characterized in that, The system includes: The data acquisition module collects multi-source business data, resource association tables, and task time series sets from the target business process; The preliminary optimization scheme set module, based on the steady-state performance characteristics and abnormal fluctuation characteristics in the multi-source business data, determines the resource allocation adjustment range and the task items to be improved for the target business process, and obtains the preliminary optimization scheme set for the target business process. The resource collaboration graph module takes the resource entities in the multi-source business data as objects and constructs the resource collaboration graph of the target business process according to the execution order of the task time sequence set and the dependency relationship in the resource association table. The full node load change set module maps the preliminary optimization scheme set to the resource collaboration graph to obtain the full node load change set of the target business process; The resource overload risk set module compares the load increase value of the nodes in the full node load change set with the available resource capacity of the nodes to identify the resource overload risk set of the business process, including: Obtain the available resource capacity and load rate of the corresponding node in the resource body from the resource association table; Based on the total load increase of the nodes, a relative load assessment is performed on the available resource capacity and load rate to obtain the overload risk index of the target business process. The overload risk index is calculated using the following formula: The overload risk index is... For node indexing, Indexing neighboring nodes, For nodes Total load increase value, For nodes Available resource capacity, For nodes load rate, This is the dynamic buffer coefficient. The collaboration impact coefficient. For nodes The set of neighboring nodes, To the neighbor node Pointing to node The strength weight of the edge relationship, Neighboring nodes Load rate; The nodes whose overload risk index exceeds the critical threshold are defined as the resource overload risk set of the target business process; The collaborative optimization execution sequence module adjusts the preliminary optimization scheme set according to the resource overload risk set to obtain the collaborative optimization execution sequence of the target business process.

Images