Process modeling management method

Abstract

The invention relates to a management method for technological process modeling, is suitable for industrial fields such as semiconductor manufacturing, and aims to solve the problem of low query efficiency in existing technological process modeling. The method comprises the following steps: defining a Main Flow object to represent a complete process flow of a product, wherein the Main Flow object comprises a plurality of Sub Flow objects; each Sub Flow object comprises a plurality of Step objects, and the Sub Flow objects and the Step objects are connected through Path objects; the method comprises the following steps: creating a PO object based on a Step object in Main Flow, and realizing flattening by extracting the Step and removing a Sub Flow hierarchy; creating a PO Path object to be directly connected with a PO object so as to represent a process sequence; the PO link relationship is constructed in the memory, and the process data is queried based on the current PO primary key ID, so that the database operation frequency is greatly reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and in particular relates to a management method for process modeling. Background Art

[0002] In existing process flow modeling systems, such as those in semiconductor manufacturing or similar industrial processes, a hierarchical object structure is typically used to manage process flows. Specifically, existing technologies use a Main Flow object to represent the complete process flow of a product. The Main Flow contains multiple Sub Flow objects as segmented management units, and each Sub Flow further contains multiple Step objects to represent specific process sites (such as processing steps). These objects are connected by Path objects to represent the process sequence, such as the processing sequence between Sub Flows or Steps. Although this hierarchical modeling method can achieve structured organization when managing complex processes (such as processes containing thousands of steps), it has significant drawbacks when performing real-time query operations: when it is necessary to query the next site (next step) of a process site or to batch retrieve the previous and next process steps (fetchstep), the system must traverse multiple layers of the structure. For example, if the current site is the last stop of a sub-flow, querying the next step requires first returning to the sub-flow layer to find the next sub-flow and then locating its starting step, which involves multiple database accesses and SQL queries. Similarly, a fetch step query requires recursively traversing the path relationship to obtain adjacent steps, causing the number of database operations to increase exponentially with process complexity. This not only delays query response time (especially affecting production efficiency during the wiptracking process), but also increases the database burden and reduces system concurrent throughput. Therefore, the main technical problem of the existing technology is that the query path caused by hierarchical modeling is lengthy, making process site queries (such as nextstep or fetch step) inefficient and unable to meet the needs of high-concurrency industrial environments. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a management method for process modeling in view of the above-mentioned deficiencies in the prior art, thereby solving the problem of low efficiency in querying process sites in process flow modeling.

[0004] To solve the above technical problems, the present invention adopts a technical solution: a process modeling management method, comprising the following steps: Define a Main Flow object, which represents the complete process flow of the product and contains multiple Sub Flow objects; Each Sub Flow object contains multiple Step objects, each of which represents a process station. Sub Flow objects and Step objects are connected by Path objects to represent the process sequence. Based on the Step object in the Main Flow, create a PO (Process Object) object. The PO object is generated by extracting the Step object and removing the Sub Flow level. Create a PO Path object, which directly connects PO objects to represent the process sequence between POs, wherein the PO Path object includes: retaining the original Path connection in the Sub Flow and creating a new connection to connect the end Step object of one Sub Flow to the start Step object of the next Sub Flow; In response to the query request, a next step query or a fetch step query is directly executed based on the PO Path object, wherein the query includes: building a link relationship of the PO object in the memory, and querying the process data based on the primary key ID of the current PO object to reduce the number of database operations.

[0005] In the above method, the creation of the PO object includes: parsing the attributes of each Step object from the Main Flow and mapping the Step object to a PO object, wherein the attributes include a site ID and process parameters.

[0006] In the above method, the step of creating a PO Path object includes automatically generating a new PO Path connection when there is no direct Path between the end Step object of a Sub Flow and the start Step object of the next SubFlow.

[0007] In the above method, the fetch step query includes: based on the current PO object, building a forward or backward PO link in the memory, and the link is used to obtain a specified number of process steps; wherein the method of building the link includes using a linked list data structure to store the PO object sequence.

[0008] In the above method, the link building method further includes calculating a query efficiency improvement factor, which is represented by the following formula: ;in: η represents the efficiency improvement factor (dimensionless); Qoriginal represents the number of database queries in the original method; Qnew represents the number of database queries in the new method; In the formula, Qoriginal and Qnew are dynamically calculated based on the number of process steps.

[0009] In the above method, the next step query includes: directly obtaining the next PO object based on the PO Path of the current PO object without accessing the Sub Flow level.

[0010] The above method further includes configuring a Process Spec object, wherein the Process Spec object is associated with the Step object and is used to set process parameters; wherein the Process Spec object loads parameters through templated rules.

[0011] In the above method, the query request includes an error handling mechanism: when the PO Path query fails, it falls back to the SubFlow layer to execute the query.

[0012] The above method further includes dynamically updating the PO Path object: in response to the process change signal, regenerating the PO Path connection.

[0013] The above method is applied to the supply chain management process, wherein the PO object represents a logistics site and the PO Path represents a transportation path.

[0014] Compared with the prior art, the present invention has the following advantages: 1. This invention doesn't simply use Step objects directly. Instead, it extracts and removes the Sub Flow hierarchy (i.e., the intermediate structure of Sub Flow objects) to generate a new abstraction layer (PO object). In existing technologies (such as the conventional hierarchical modeling of Main Flow, Sub Flow, and Step), process sites (Steps) are nested within the Sub Flow hierarchy. Querying requires traversing each Sub Flow layer, resulting in low efficiency. By removing the Sub Flow hierarchy, this invention flattens the data, simplifies the data structure, and provides a foundation for efficient subsequent queries.

[0015] 2. This invention does not simply reuse existing Path objects. Instead, it builds a path network directly connecting PO objects by combining "preserving existing Path connections" with "creating new connections." Specifically, in the prior art, Path objects are only used to connect the order between sub-flows or steps. To query the next step, one must first return to the sub-flow layer to find the next sub-flow and then locate its starting step, resulting in multiple database accesses. This invention, however, establishes a direct path between POs by "creating new connections" (e.g., directly connecting the ending step to the starting step), achieving a "one-step" query.

[0016] 3. This invention shifts query operations from traditional database-intensive processing to in-memory processing. Specifically, in the prior art, fetch step queries (e.g., searching for a certain number of process steps before and after) require multiple SQL queries to traverse the Path and Step relationships, resulting in heavy database overhead and low concurrent throughput. This invention, however, simplifies multiple database operations into a small number of primary key ID queries by "building link relationships between PO objects in memory" (i.e., generating a complete link using pre-established direct connections within the PO Path) and "querying process data based on the primary key ID of the current PO object."

[0017] In summary, the beneficial effects of the present invention are concentrated in "creation of PO objects", "construction of PO Path objects" and "memory-based query execution". These features work together to solve the query efficiency problem in process flow modeling. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Schematic diagram of the Main Flow hierarchical relationship of the present invention.

[0019] Figure 2 Schematic diagram of the PO Path connection logic of the present invention. DETAILED DESCRIPTION

[0020] Explanation of terms: 1. Main Flow: Represents the complete process flow of a product and is the top-level management object. It contains multiple Sub Flow objects and is used to structure and organize the process chain (such as the entire process of semiconductor chip production).

[0021] 2. Sub Flow: A segmented management unit of the Main Flow; each Sub Flow contains multiple Step objects and is connected to other Sub Flows or Steps through Path objects, representing a local process sequence (e.g., a "welding sub-flow" contains multiple welding sites).

[0022] 3.Step (process site): The basic operation unit of Sub Flow, representing a specific process step (such as "temperature control site"); attributes include the site ID (unique identifier) ​​and process parameters (such as temperature range); Steps are connected by Path objects.

[0023] 4. Path: connects the sequential relationship objects between sub flows or steps, and defines the processing sequence logic (such as from the end step of sub flow A to the start step of sub flow B).

[0024] 5. PO (Process Object): A flattened object generated by extracting the Step object and removing the Sub Flow level; the original properties of the Step (such as the site ID and process parameters) are retained to simplify queries (for example, mapping a "weld point" to an independent PO).

[0025] 6. PO Path (Process Object Path): Directly connects to the path network of the PO object; retains the original path connection and adds a new connection between the end step and the starting step of different sub flows to ensure the continuity of the process sequence (such as directly connecting the "end station of the welding section" to the "start station of the test section").

[0026] 7.Process Spec (Process Specification Object): A parameter template associated with the Step object, used to set process parameters (such as the "weld point" loading voltage and time template); ensure parameter consistency through regular configuration.

[0027] 8. WIP tracking: A process for tracking product process status in real time. This method optimizes query efficiency through the PO model and solves the delay problem caused by traditional hierarchical modeling.

[0028] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments.

[0029] See Figure 1 and Figure 2 As shown, a process modeling management method includes the following steps: Step 1: Define a Main Flow object, which represents the complete process flow of the product and contains multiple Sub Flow objects. Each Sub Flow object contains multiple Step objects, each of which represents a process station. Sub Flow objects and Step objects are connected by Path objects to represent the process sequence. Step 2: Create a PO (Process Object) object based on the Step object in the Main Flow. The PO object is generated by extracting the Step object and removing the Sub Flow level. Step 3: Create a PO Path object. The PO Path object directly connects PO objects to represent the process sequence between POs. The PO Path object includes: retaining the original Path connection in the Sub Flow and creating a new connection to connect the end Step object of one Sub Flow to the start Step object of the next Sub Flow; Step 4: In response to the query request, directly execute a next step query or a fetch step (batch fetching) query based on the PO Path object. The query includes: building a link relationship of the PO object in memory and querying the process data based on the primary key ID of the current PO object to reduce the number of database operations.

[0030] See Figure 1 As shown: 1. Main Flow contains multiple Sub Flows; 2. Sub Flow contains multiple steps; 3. After the last station of the Sub Flow is completed, the sub flow will jump out and find the next SubFlow based on the main flow, and then continue to find the first station of the Subflow to start running the goods.

[0031] See Figure 2 As shown: 1. The Main Flow PO directly flattens the Sub Flow, and the PO number is the original Sub Flow number + ". " + the original Step number; 2.PO Path retains the path connections within Sub Flow; For example, 0010.0010->0010.0020; 3. Cross-Sub Flow connection: connect the last step of a Sub Flow to the first step of the next Sub Flow; For example: 0010.0020->0020.0010.

[0032] It's important to note that while the above method constructs a hierarchical process model by defining a Main Flow object (containing multiple Sub Flow objects) and Step objects, the core innovation lies in the creation of PO and PO Path objects, which achieve a flat representation of the process sequence. This directly addresses the issue of inefficient queries: in "next step" queries, there's no need to access each Sub Flow layer, reducing the number of database operations from multiple to one. In "fetch step" queries, in-memory PO links significantly reduce query time and improve software throughput, especially for processes with up to 5,000 steps. In terms of implementation details, the Main Flow object represents the entire process flow. For example, an electronics assembly process might include multiple Sub Flows (such as "Soldering Sub-Flow" and "Testing Sub-Flow"). Each Sub Flow contains multiple Steps (such as "Solder Point 1" and "Solder Point 2"). Steps are connected by Path objects to represent their sequence. When creating a PO object, all Step objects are extracted from the Main Flow, removing the Sub Flow hierarchy. For example, "Solder Point 1" in the "Welding Sub-Process" is mapped to a separate PO object, whose attributes include a site ID (e.g., a unique identifier) ​​and process parameters (e.g., temperature and time). The PO Path object directly connects the PO object, retaining the original path (e.g., the line from "Solder Point 1" to "Solder Point 2") and adding new lines to connect the end and start steps of different Sub Flows (e.g., linking the last solder point in the "Welding Sub-Process" directly to the first test point in the "Testing Sub-Process"). When processing a query request, such as a user querying the "next step" of the current site, the system constructs a PO link in memory: based on the primary key ID of the current PO object (e.g., a database index), the next PO object is directly found through the PO Path, without querying the Sub Flow table. For example, suppose the Main Flow has three Sub Flows, each containing 10 Steps. Traditionally, searching for the next site requires first querying the Sub Flow layer (an average of two database queries) and then the Step layer (one query). This method, however, uses the Point-of-View (PO) Path to perform a single in-memory query, improving efficiency by over 50%. Fetching a step query is similar. For example, to retrieve five steps backward from the current point, the system constructs a linked list structure to store the sequence and directly retrieve data in batches.

[0033] In this embodiment, creating the PO object includes: parsing the attributes of each Step object from the Main Flow, and mapping the Step object to a PO object, wherein the attributes include a site ID and process parameters.

[0034] By parsing Step object attributes (such as site ID and process parameters) and mapping them to PO objects, data consistency is ensured, avoiding the risk of attribute loss or incorrect mapping, thereby improving model accuracy. In implementation details, when parsing Step object attributes from the Main Flow, for example, a Step representing a "Temperature Control Site" would have attributes including a site ID (such as "TC-001") and process parameters (such as "Temperature Range: 100-150°C"). When mapping to a PO object, the system automatically extracts these attributes and stores them as fields in the PO. For example, in supply chain logistics, a Step representing a "Warehouse Sorting Point" would have its ID (such as "WH-Sort-01") and parameters (such as "Sorting Speed: 50 pieces / minute") directly included in the mapped PO object. This step is implemented as a data parsing function in the software, and those skilled in the art can use JSON or XML format to store attributes for ease of implementation.

[0035] like Figure 2 As shown, in this embodiment, the creation of the PO Path object includes: when there is no direct Path between the end Step object of a Sub Flow and the start Step object of the next Sub Flow, automatically generating a new PO Path connection.

[0036] When there's no direct path between the end step of a sub-flow and the start step of the next sub-flow, a new PO path is automatically generated. This resolves process flow interruptions, ensures query continuity, reduces manual intervention, and improves system automation. For example, consider two sub-flows: the "Assembly" segment ends at "Final Inspection" and the "Packaging" segment begins at "Carton Sealing Start." If there's no path between them, the system detects this and automatically creates a new PO path, directly connecting the PO objects of these two steps. Operational process: The software traverses the main flow, identifies sub-flow boundary points, and uses algorithms to generate connections (e.g., based on location or logical relationships). For example, in electronics manufacturing, this automatically links "Test Completion" to "Packaging Start," avoiding the overhead of backtracking to the sub-flow layer during querying.

[0037] In this embodiment, the fetch step query includes: based on the current PO object, building a forward or backward PO link in the memory, and the link is used to obtain a specified number of process steps; wherein the method of building the link includes using a linked list data structure to store the PO object sequence.

[0038] It should be noted that the fetch step query constructs a forward or backward PO link in memory based on the current PO object, using a linked list data structure to quickly retrieve a specified number of process steps, significantly reducing database access times (from O(n) to O(1)), and improving concurrency (as described in Document 2 regarding throughput improvements). In implementation details, when building a link, a linked list is used to store the PO object sequence. For example, to construct a three-step link from the current PO (e.g., "Site A"), the system creates linked list nodes in memory, each pointing to a PO object (Node 1: PO_A, Node 2: PO_B, Node 3: PO_C), and directly traverses the list to retrieve data. For example, if a user queries for the next five transport points from the current logistics site, the system initializes the linked list, populates the PO objects based on the POPath order, and then queries the database at once to retrieve all process data, avoiding the multiple SQL queries required by traditional methods.

[0039] In this embodiment, the link construction method further includes calculating a query efficiency improvement factor, which is represented by the following formula: ;in: η represents the efficiency improvement factor (dimensionless); Qoriginal represents the number of database queries in the original method; Qnew represents the number of database queries in the new method; In the formula, Qoriginal and Qnew are dynamically calculated based on the number of process steps.

[0040] Method improvements are quantified by calculating the query efficiency improvement factor (η). η is expressed as η = (Q_original - Q_new) / Q_original, where Q_original is the number of original queries and Q_new is the number of queries using the new method. This provides a benchmark for performance evaluation and helps optimize system design. In implementation details, the factor calculation is dynamic based on the number of process steps. For example, for a 1000-step process, the traditional next step query Q_original averages 10 times (involving the SubFlow layer), while the new method uses Q_new only once. Therefore, η = (10-1) / 10 = 0.9, representing a 90% efficiency improvement. During operation, the software counts the number of queries before and after the query, calculating η in real time. For example, in supply chain management, the system monitors database calls for a 5000-step process and dynamically outputs the η value for reporting.

[0041] In this embodiment, the next step query includes: directly obtaining the next PO object based on the PO Path of the current PO object without accessing the Sub Flow level.

[0042] It's important to note that the next step query directly retrieves the next PO object based on the current PO object's PO Path, bypassing the Sub Flow hierarchy. This simplifies query logic and reduces response time, making it particularly beneficial for real-time tracking systems such as WIP tracking. In implementation details, for example, if the current PO is a "Quality Inspection Station," its PO Path directly links to the PO object at a "Packaging Station." The system only needs a single memory access to retrieve the next PO's primary key ID, then query the database. Those skilled in the art can implement this using a hash table to store PO Path relationships to ensure fast retrieval.

[0043] This embodiment also includes configuring a Process Spec object, which is associated with a Step object and is used to set process parameters; wherein the Process Spec object loads parameters through template rules.

[0044] It's important to note that by associating a Process Spec object with a Step object to set process parameters, and loading these parameters through templated rules, this enhances the flexibility and consistency of parameter management and reduces configuration errors. In implementation details, the Process Spec object acts as a template. For example, a "Welding Point" Step can be associated with a Process Spec to define parameters (e.g., "Voltage: 220V, Time: 5s"). Templated rules can be automatically loaded based on product type (e.g., "Class A Product" uses a preset template). Operational process: The system calls the Process Spec when creating a Step. For example, in automotive assembly, the "Bolt Tightening" Step loads template parameters to ensure consistent parameters at each station.

[0045] In this embodiment, the query request includes an error handling mechanism: when the PO Path query fails, it falls back to the SubFlow layer to execute the query.

[0046] It's important to note that the error handling mechanism falls back to the Sub Flow layer if a PO Path query fails, providing redundancy, preventing system crashes, and improving robustness. In implementation details, if a PO Path is damaged or missing, the system will fail detection and fall back to the traditional method: first querying the Sub Flow layer to find the next Sub Flow and then obtaining its starting step. During implementation, the software includes an exception handling module, such as a try-catch block, to handle query failures, ensuring a seamless fallback.

[0047] This embodiment also includes dynamically updating the PO Path object: in response to the process change signal, regenerating the POPath connection.

[0048] It's important to note that dynamically updating the PO Path object in response to process change signals and regenerating connections allows the model to adapt to process changes (such as process optimization) in real time, maintaining query efficiency. In implementation details, upon receiving a change signal (such as a user modifying the order of sub-flows), the system triggers an update: it reparses the main flow and generates a new PO Path. For example, if a sub-flow is added during a production line modification, the system automatically rebuilds all PO Path connections to ensure they are up-to-date.

[0049] In this embodiment, the method is applied to a supply chain management process, wherein a PO object represents a logistics site and a POPath represents a transportation path.

[0050] It should be noted that the process modeling management method provided by this invention can also be extended to supply chain management processes. PO objects represent logistics sites (such as warehouses or transshipment points), and PO Path represents transportation routes. This expands application scenarios and resolves query bottlenecks in logistics tracking (e.g., obtaining the next site in real time). In a specific implementation, for example, a Main Flow represents a global logistics chain, with Sub Flows representing "sea transport segments" and "land transport segments," and Steps representing specific sites (e.g., "Port A"). After creating a PO object, the PO Path directly connects these sites, allowing queries to retrieve the route in a single operation. Those skilled in the art can easily transfer this model to similar fields, such as using the same model to manage e-commerce delivery processes.

[0051] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any way. Any simple modification, change and equivalent structural change made to the above embodiment based on the technical essence of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A management method for process modeling, characterized in that: The following steps are involved: Define a Main Flow object, which represents the complete process flow of the product and contains multiple SubFlow objects; Each Sub Flow object contains multiple Step objects, each of which represents a process site. Sub Flow objects and Step objects are connected by Path objects to represent the process sequence. Based on the Step object in the Main Flow, create a PO object, which is generated by extracting the Step object and removing the SubFlow level; Create a PO Path object, which directly connects PO objects to represent the process sequence between POs, wherein the POPath object includes: retaining the original Path connection in the Sub Flow and creating a new connection to connect the end Step object of one Sub Flow to the start Step object of the next Sub Flow; In response to the query request, a next step query or a fetch step query is directly executed based on the PO Path object, wherein the query includes: building a link relationship of the PO object in the memory, and querying the process data based on the primary key ID of the current PO object to reduce the number of database operations.

2. The method according to claim 1, characterized in that The creation of the PO object includes: parsing the attributes of each Step object from the Main Flow and mapping the Step object to the PO object, wherein the attributes include the site ID and process parameters.

3. The method according to claim 1, characterized in that The creation of the PO Path object includes: automatically generating a new PO Path connection when there is no direct Path between the end Step object of a Sub Flow and the start Step object of the next Sub Flow.

4. The method according to claim 1, wherein The fetch step query includes: based on the current PO object, building a forward or backward PO link in the memory, the link is used to obtain a specified number of process steps; wherein the method of building the link includes using a linked list data structure to store a PO object sequence.

5. The method according to claim 4, characterized in that The method for building a link further includes calculating a query efficiency improvement factor, which is represented by the following formula: ;in: η represents the efficiency improvement factor; Qoriginal represents the number of database queries in the original method; Qnew represents the number of database queries in the new method; In the formula, Qoriginal and Qnew are dynamically calculated based on the number of process steps.

6. The method according to claim 1, characterized in that The next step query includes: directly obtaining the next PO object based on the PO Path of the current PO object without accessing the Sub Flow level.

7. The method according to claim 1, characterized in that The method further includes configuring a Process Spec object, which is associated with a Step object and is used to set process parameters; wherein the Process Spec object loads parameters through template rules.

8. The method according to claim 1, characterized in that The query request includes an error handling mechanism: when the POPath query fails, it falls back to the Sub Flow layer to execute the query.

9. The method according to claim 1, characterized in that It also includes dynamic updating of PO Path objects: in response to process change signals, PO Path connections are regenerated.

10. The method according to claim 1, characterized in that The method is applied to the supply chain management process, wherein the PO object represents a logistics site and the PO Path represents a transportation path.

Images