A componentized design and dynamic arrangement method of visualizing process flow transfer nodes
By using depth-first traversal and normalization, loop structures are identified and node coordinates are optimized, solving the problem of component arrangement disorder caused by loop structures in flowcharts and improving the visualization effect of flowcharts.
Patent Information
- Application Number
- CN202611318227.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-28
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies, when dealing with loop structures such as retry loops, iterative approval loops, or polling in flowcharts, result in chaotic node component arrangement, severe line crossing, and difficulty in intuitively presenting the flow pattern.
By using depth-first traversal to identify loop structures, calculating iteration intensity indicators, generating a normalized probability weight graph, merging node coordinates, and optimizing connection paths, the identification of the main path reflects the true importance of the business.
It enables dynamic arrangement of node components in the flowchart to match business logic, reduces the canvas space occupied by loop structures, and improves the visualization effect of the flowchart.
Smart Images

Figure CN123117941A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of process visualization and orchestration technology, and more specifically, to a component-based design and dynamic orchestration method for visualized process flow nodes. Background Technology
[0002] In a visual workflow orchestration system, a flowchart consists of multiple node components, with each component expressing its flow relationship through directed connections. Existing technologies calculate the cumulative number of transitions for each connection from execution history logs and directly use this to drive the identification of the main branch of the node components and the arrangement of their canvas positions. However, when the workflow contains loop structures such as retry loops, iterative approval loops, or polling, the connections within these loops continuously accumulate transition counts over multiple iterations, reaching values tens of times higher than normal forward connections. When this skewed cumulative count is used to directly drive the identification of the main branch path, loop components are misjudged as core nodes, occupying a large amount of orchestration space in the center of the canvas, while node components carrying the main business logic are compressed to the edge areas. This results in a severe discrepancy between the dynamic orchestration results and the logical importance of the actual business process. Summary of the Invention
[0003] This invention provides a component-based design and dynamic arrangement method for visualizing process flow nodes, solving the technical problems in related technologies where the directed graph of the process has a loop structure, making it impossible to accurately quantify iterative behavior, resulting in chaotic node arrangement coordinates, serious line intersections, and difficulty in intuitively presenting the process flow pattern.
[0004] This invention discloses a component-based design and dynamic orchestration method for visual workflow nodes, comprising the following steps: Perform a depth-first traversal on the set of node components in the directed graph of the process, maintain the current recursion stack, mark the directed edges of the target node that already exist in the current recursion stack as back edges, identify all loop structures based on all back edges, extract the set of node components, entry node identifier, exit node identifier and back edge identifier of each loop, and generate a list of loop structure descriptions. Obtain the entry count of each loop entrance node from the execution history log. Number of cycles with the back edge ,according to Calculate the iterative intensity index And store it in the loop structure description list; Normalize the original number of connections for all links: For non-loop links, normalize the probability weights. Take the original number of transfers For the internal connections of the loop, according to Calculate the normalized probability weights, where For the logarithmic damping coefficient, generate a normalized probability weight map; Based on the normalized probability weight graph, the backbone node components are determined along the path sequence with the highest cumulative normalized probability weight, and vertical orchestration coordinates are assigned. The loop structure is treated as a compact sub-orchestration unit, and then... Calculate the size of the region enclosed by the loop body, where The standard arrangement area size for a single node is used to allocate the coordinates of the nodes inside the loop according to the loop topology order within the enclosed area. Merge all node coordinates, perform line intersection minimization processing, generate arc-shaped connection paths for the back edges, output the final set of node coordinates and connection paths, and update the position attributes and connection rendering styles of each node component on the canvas.
[0005] Furthermore, the calculation of the iteration strength index also includes: when a loop has multiple back edges, taking the sum of the number of cycles of each back edge as the value. .
[0006] Furthermore, the calculation of the size of the region enclosed by the loop body also includes: when When the sum of the minimum spacing required for the number of nodes within the loop is less than the sum of the minimum spacings, the sum of the minimum spacings is taken; when When the maximum allowable area of the main trunk is exceeded, the size value of the maximum allowable area of the main trunk shall be used.
[0007] Furthermore, the connection intersection minimization process takes the hierarchical position and the order of arrangement of nodes in the same layer in the complete coordinate set as input, and outputs the arrangement scheme of nodes in the same layer that minimizes the number of connection intersections.
[0008] Furthermore, the logarithmic damping coefficient Dynamic adjustments based on historical programming quality feedback, and .
[0009] This invention discloses a component-based design and dynamic orchestration system for visual workflow nodes, comprising: The loop detection module is used to perform a depth-first traversal of the node component set in the directed graph of the process, maintain the current recursion stack, mark the directed edges of the target node that already exist in the current recursion stack as back edges, identify all loop structures based on all back edges, extract the node component set, entry node identifier, exit node identifier and back edge identifier of each loop, and generate a loop structure description list. The iteration intensity calculation module is used to obtain the number of times each loop entry node is entered from the execution history log. Number of cycles with the back edge ,according to Calculate the iterative intensity index The normalized weight generation module is used to maintain the original number of cycles for non-loop connections and to normalize the number of cycles for loop connections. Calculate the normalized probability weights and generate a normalized probability weight map; The coordinate allocation module is used to determine the backbone node components based on the normalized probability weight map and allocate vertical orchestration coordinates, treating the loop structure as a compact sub-orchestration unit, according to... Calculate the size of the region enclosed by the loop body and assign the coordinates of the nodes inside the loop; The rendering output module is used to merge all node coordinates, perform line intersection minimization processing, generate arc-shaped connection paths for back edges, output the final set of node coordinates and connection paths, and update the position attributes and connection rendering styles of each node component on the canvas.
[0010] Furthermore, when the loop has multiple back edges, the iterative intensity calculation module takes the sum of the number of cycles of each back edge as the value. .
[0011] Furthermore, the coordinate allocation module is also used to: when When the sum of the minimum spacing required for the number of nodes within the loop is less than the sum of the minimum spacings, the sum of the minimum spacings is taken; when When the maximum allowable area of the main trunk is exceeded, the size value of the maximum allowable area of the main trunk shall be used.
[0012] The beneficial effects of this invention are as follows: This invention separates cyclic structures from ordinary flow paths by performing loop detection on the directed graph of the process, and calculates an iteration intensity index to quantify the average loop depth of a single process instance. Simultaneously, by dividing the original number of flows of the internal connections of the loop by the total number of iterations and applying logarithmic damping, the skew effect of accumulated loop iterations on the probability distribution is eliminated, ensuring that the main path identification results reflect the true importance of the business. This invention also links the canvas occupancy ratio of the loop body to the iteration intensity index, preventing low-business-value loop bodies from occupying a large amount of orchestration space. This invention solves the problem of the misalignment between component orchestration space allocation and business logic importance caused by cyclic structures, ensuring that the dynamic orchestration position of each node component in a flowchart containing a cyclic topology matches its actual weight in the business process. Attached Figure Description
[0013] Figure 1 This is a flowchart of the component-based design and dynamic arrangement method for visual process flow nodes provided in the embodiments of the present invention. Detailed Implementation
[0014] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, some features described in the examples may be combined in other examples.
[0015] At least one embodiment of the present invention discloses a component-based design and dynamic orchestration method for visual workflow nodes, see [link to relevant documentation]. Figure 1 This includes the following steps: Step 1: Perform loop detection on the directed graph of the process and generate a list of loop structure descriptions.
[0016] Perform a depth-first traversal of the node component set in the directed graph of the process. During the traversal, maintain the current recursion stack. When the target node of a directed edge already exists in the current recursion stack, mark the edge as a back edge. Based on the complete set of back edges, identify all loop structures in the flowchart, extract the node component set, entry node identifier, exit node identifier, and back edge identifier for each loop, and generate a list of loop structure descriptions.
[0017] It should be noted that an entry node is the node pointed to by a back edge, that is, a node in a loop that can be directly reached from outside the loop; an exit node is a node inside the loop that has a connection pointing to the outside of the loop. A loop can contain multiple exit nodes, all of which are recorded in the loop structure description list.
[0018] Step 2: Calculate the iteration intensity index for each loop.
[0019] For each loop structure in the loop structure description list, obtain the number of times its entry node is entered from the execution history log. Number of cycles corresponding to the back edge of the loop Calculate the iteration intensity index : ; in, To determine the total number of times the entry node was reached in the historical log, This represents the total number of times the back edge was triggered. This represents the average number of loop iterations that occur after a single process instance enters the loop. Store the corresponding loop structure description list entry.
[0020] In this embodiment of the application, when a loop has multiple back edges, the sum of the number of cycles of each back edge is taken as the loop length. This is to fully reflect the iterative behavior of the loop.
[0021] Step 3: Normalize the original number of connections for all connections to generate a normalized probability weight map.
[0022] Normalize the original number of turns for all connections in the directed graph. Let the connections... The original number of transfers is The total number of iterations for its associated loop is The logarithmic damping coefficient is ( ).
[0023] For non-loop connections, their normalized probability weights Keep the original number of transfers unchanged: ; For connections within a loop, their normalized probability weights for: ; in, It is the natural logarithm function. This is a logarithmic damping term used to scale the weights of connections within a loop to the same order of magnitude as connections outside the loop. For connecting lines The normalized probability weights are then calculated. After processing, a normalized probability weight graph is obtained, where each line corresponds to a normalized probability weight value. .
[0024] It should be noted that the logarithmic damping coefficient It can be preset according to the size of the flowchart and the depth of the loop, or it can be dynamically adjusted based on historical orchestration quality feedback. The larger the value, the higher the weight of the internal connections of the loop, and the stronger the influence of the loop as a whole in the main structure identification.
[0025] Step 4: Perform main path identification and coordinate allocation based on the normalized probability weight map.
[0026] Based on the normalized probability weight graph, the main node components are determined along the path sequence with the highest cumulative normalized probability weight. The main node components are assigned vertical arrangement coordinates, and the non-main branch node components are assigned coordinates at the horizontal offset positions of the corresponding main nodes.
[0027] For the loop structure, the entire structure is treated as a compact sub-arrangement unit. The dimensions of the region enclosed by the loop body are calculated using the inlet and outlet nodes as anchor points. : ; in, This refers to the standard orchestration area size for a single node. The iterative intensity index calculated in step 2, The size of the area enclosed by the loop body. and Proportional to the iteration intensity, loops with lower iteration intensity occupy a smaller canvas space. Node components within a loop are assigned coordinates according to the loop's topological order within the enclosing region.
[0028] In this embodiment of the application, in order to maintain visual readability when the arrangement of coordinates has a large number of loop nodes, the following can be done: Set upper and lower bound constraints: when When the sum of the minimum spacing required for the number of nodes within the loop is less than the sum of the minimum spacings, the sum of the minimum spacings is taken; when When the maximum allowable area of the main trunk is exceeded, it is clipped to the upper limit value.
[0029] Step 5: Merge the coordinate sets and complete the dynamic arrangement and rendering of the flowchart.
[0030] The coordinates of the main nodes, branch nodes, and loop nodes are merged into a complete coordinate set. A connection intersection minimization process is then performed on the merged complete coordinate set. The input to this process is the hierarchical position and order of nodes within the complete coordinate set; the output is the node arrangement scheme that minimizes the number of connection intersections. Arc-shaped connection paths are generated for loop back edges to differentiate them from forward connections in terms of rendering style. The final set of node coordinates and connection paths are output, and the position attributes of each node component on the canvas and the connection rendering style are updated, completing the dynamic arrangement and rendering of the flowchart.
[0031] On the other hand, this invention also proposes a component-based design and dynamic orchestration system for visualized workflow nodes, including: The loop detection module is used to perform a depth-first traversal of the node component set in the directed graph of the process, identify all loop structures based on the back edge set, extract the node component set, entry node identifier, exit node identifier and back edge identifier of each loop, and generate a loop structure description list. The iteration intensity calculation module is used to obtain the number of entries and the number of back edges of each loop entry node from the execution history log, calculate the iteration intensity index and store it in the loop structure description list. The normalized weight generation module is used to normalize the original number of iterations of all connections, keep the original value of non-loop connections, and scale the internal loop connections according to the total number of iterations and the logarithmic damping coefficient to generate a normalized probability weight map. The coordinate allocation module is used to perform backbone path identification based on the normalized probability weight map, allocate the longitudinal arrangement coordinates of backbone node components, treat the loop structure as a compact sub-arrangement unit, calculate the size of the enclosing region according to the iterative intensity index, and allocate the coordinates of the nodes inside the loop. The rendering output module is used to merge all node coordinate sets, perform line intersection minimization processing, generate arc-shaped connection paths for back edges, output the final node coordinate set and connection path set, and update the position attributes and connection rendering styles of each node component on the canvas.
[0032] By way of example, the system proposed in this invention may also include various features and combinations thereof in the method embodiments, which will not be elaborated here.
[0033] Step 1, loop detection, separates the loop structure from the ordinary flow path, enabling subsequent normalization processing to accurately locate the internal connections of the loop. Step 2, the iteration intensity index, quantifies the average loop depth of a single process instance, providing a reference benchmark based on actual execution behavior for step 3, weight scaling. Step 3 eliminates the skew effect of accumulated loop iterations on the probability distribution by dividing the original number of flows of the internal connections of the loop by the total number of iterations and applying logarithmic damping. This ensures that loop connections and non-loop connections in the normalized probability weight map are on the same order of magnitude, thus allowing the backbone path identification result in step 4 to reflect the true business importance, rather than the iteration frequency. Step 4, the compact sub-orchestration unit processing, links the canvas occupancy ratio of the loop body to the iteration intensity index, preventing low-business-value loop bodies from occupying a large amount of orchestration space. Combining the above steps, the dynamic orchestration position of each node component in the flowchart containing loop topology matches its actual weight in the business process, solving the problem of the discrepancy between component orchestration space allocation and business logic importance caused by loop structures.
[0034] The embodiments of the present invention have been described above. However, the embodiments are not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make more equivalent embodiments under the guidance of the present embodiments, and all of them are within the protection scope of the present embodiments.
Claims
1. A method for visualizing componentized design and dynamic orchestration of process flow nodes, comprising: Includes the following steps: Perform a depth-first traversal on the set of node components in the directed graph of the process, maintain the current recursion stack, mark the directed edges of the target node that already exist in the current recursion stack as back edges, identify all loop structures based on all back edges, extract the set of node components, entry node identifier, exit node identifier and back edge identifier of each loop, and generate a list of loop structure descriptions. Obtain the entry count of each loop entrance node from the execution history log. Number of cycles with the back edge ,according to Calculate the iterative intensity index And store it in the loop structure description list; Normalize the original number of connections for all links: For non-loop links, normalize the probability weights. Take the original number of cycles For the internal connections of the loop, according to Calculate the normalized probability weights, where For the logarithmic damping coefficient, generate a normalized probability weight map; Based on the normalized probability weight graph, the backbone node components are determined along the path sequence with the highest cumulative normalized probability weight, and vertical orchestration coordinates are assigned. The loop structure is treated as a compact sub-orchestration unit, and then... Calculate the size of the region enclosed by the loop body, where The standard arrangement area size for a single node is used to allocate the coordinates of the nodes inside the loop according to the loop topology order within the enclosed area. Merge all node coordinates, perform line intersection minimization processing, generate arc-shaped connection paths for the back edges, output the final set of node coordinates and connection paths, and update the position attributes and connection rendering styles of each node component on the canvas.
2. The component-based design and dynamic orchestration method for visual workflow nodes according to claim 1, wherein, The calculation of the iteration strength index also includes: when a loop has multiple back edges, taking the sum of the number of cycles of each back edge as the value. .
3. The component-based design and dynamic orchestration method for visual workflow nodes according to claim 1, wherein, The calculation of the size of the area enclosed by the loop body also includes: when When the sum of the minimum spacing required for the number of nodes within the loop is less than the sum of the minimum spacings, the sum of the minimum spacings is taken; when When the maximum allowable area of the main trunk is exceeded, the size value of the maximum allowable area of the main trunk shall be used.
4. The component-based design and dynamic orchestration method for visual workflow nodes according to claim 1, wherein, The line intersection minimization process takes the hierarchical position and order of the nodes in the complete coordinate set as input, and outputs the node arrangement scheme with the minimum number of line intersections.
5. The component-based design and dynamic orchestration method for visual workflow nodes according to claim 1, wherein, The logarithmic damping coefficient Dynamic adjustments based on historical programming quality feedback, and .
6. A component-based design and dynamic orchestration system for visualizing workflow nodes, characterized in that, include: The loop detection module is used to perform a depth-first traversal of the node component set in the directed graph of the process, maintain the current recursion stack, mark the directed edges of the target node that already exist in the current recursion stack as back edges, identify all loop structures based on all back edges, extract the node component set, entry node identifier, exit node identifier and back edge identifier of each loop, and generate a loop structure description list. The iteration intensity calculation module is used to obtain the number of times each loop entry node is entered from the execution history log. Number of cycles with the back edge ,according to Calculate the iterative intensity index And store it in the loop structure description list; The normalized weight generation module is used to maintain the original number of cycles for non-loop connections and to normalize the number of cycles for internal loop connections. Calculate the normalized probability weights and generate a normalized probability weight map; The coordinate allocation module is used to determine the backbone node components based on the normalized probability weight map and allocate vertical orchestration coordinates, treating the loop structure as a compact sub-orchestration unit, according to... Calculate the size of the region enclosed by the loop body and assign the coordinates of the nodes inside the loop; The rendering output module is used to merge all node coordinates, perform line intersection minimization processing, generate arc-shaped connection paths for back edges, output the final set of node coordinates and connection paths, and update the position attributes and connection rendering styles of each node component on the canvas.
7. The component-based design and dynamic orchestration system for visual workflow nodes according to claim 6, wherein, When the loop has multiple back edges, the iteration intensity calculation module takes the sum of the number of cycles of each back edge as the value. .
8. The component-based design and dynamic orchestration system for visual workflow nodes according to claim 6, wherein, The coordinate allocation module is also used for: when When the sum of the minimum spacing required for the number of nodes within the loop is less than the sum of the minimum spacings, the sum of the minimum spacings is taken; when When the maximum allowable area of the main trunk is exceeded, the size value of the maximum allowable area of the main trunk shall be used.
