MES automatic task allocation method and system based on dynamic process matching and workload balancing

Abstract

The invention discloses an MES automatic task allocation method and system based on dynamic process matching and workload balancing, and the method comprises the steps: S1, a standard production task process set is manually created in an MES system, and the processes comprise at least one of sawing, drilling, milling, riveting, welding, boring, turning and grinding; s2, creating a process flow template library corresponding to the standard production task, wherein the process flow template library comprises a process sequence and a work type requirement required by part production; s3, generating a BOM list through the PLM system, wherein the BOM list at least comprises part names, specifications, models and raw material information; s4, importing the BOM list into an MES (Manufacturing Execution System); s5, key fields in the BOM list are matched with a process flow template library, and a process flow needed for producing the part is determined; s6, dynamically judging a matching result; s7, a task allocation stage; and S8, the task is pushed to an employee account through an MES system mobile terminal interface. The method has the advantages that a production line can be helped to better and more reasonably allocate resources, unfairness is avoided, new tasks can be helped to execute data statistics, management cost can be saved, and efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of MES system processing, and particularly to an MES automatic task allocation method and system based on dynamic process matching and workload balancing. Background Art

[0002] In the prior art, the allocation of tasks from design to workshop implementation is manually dispatched by the planning center or production workshop administrator according to their work experience. mainly, the planning center or production workshop administrator determines the production processes required according to experience, fills in the production transfer form corresponding to the drawing; then sends the production transfer form to the workshop management personnel, and the management personnel assign the work items of the specific implementing workers according to their feelings. The traditional task allocation has the following disadvantages: 1. Experienced scheduling personnel are required to systematically plan the production sequence of each part in the drawing; 2. Scheduling personnel are required to operate. If something is delayed, scheduling cannot be carried out in time; 3. Close cooperation among scheduling personnel, workshop administrators, etc. is required. If any link is disjointed, the production task cannot be ordered in time; the more production tasks, the more people are required, and the purpose of mass-expanded reproduction cannot be achieved.

[0003] In view of the problems in the related art, the present invention proposes an MES automatic task allocation method and system based on dynamic process matching and workload balancing. Summary of the Invention

[0004] In order to achieve the above object, the present invention adopts the following technical solutions: An MES automatic task allocation method based on dynamic process matching and workload balancing, comprising the following steps: S1. Manually create a set of standard production task processes in the MES system, where the processes include at least one of sawing, drilling, milling, riveting, welding, boring, turning, and grinding; S2. Create a process flow template library corresponding to the standard production task, where the process flow template library contains the processes and man-hour requirements for producing parts; S3. Generate a BOM list through the PLM system, where the BOM list contains at least part name, specification, model, and raw material information; S4. Import the BOM list into the MES system; S5. Match the keyword fields (part name, specification, model, raw material information) in the BOM list with the process flow template library to determine the process flow required for producing the part; S6. Dynamically determine the matching result: If there are completely matching keyword fields, automatically generate the corresponding process flow, automatically generate a production task, and notify the manager; If there is no exact match, fuzzy matching is performed in the order of priority (part name > specification > model > raw material) to generate candidate process flows for the planning team to confirm. After confirmation, it is updated and saved to the process flow template library of S2. S7. Task allocation stage: S7.1 Calculate the real-time workload load value W of employee i according to the following formula i : , where is the standard working hours of the kth unfinished process, is the process priority coefficient (value range 1.0 - 2.0), is the number of unfinished tasks of employee i, is the task number weight coefficient (default 0.5); S7.2 Select employees who meet and are jointly matched, and sort them from low to high according to the current workload (real-time workload load value). Give priority to allocating tasks to the employee with the smallest load to achieve load balancing; where W i represents the current workload, and W threshold represents the preset threshold; S8. The production task is pushed to the employee account through the MES system mobile terminal interface.

[0005] Furthermore, in S7.1, the calculation method of the process priority coefficient is: , where is the current date, is the planned completion date of process k, , are the latest task deadline and the earliest task deadline in the system respectively.

[0006] Furthermore, in the fuzzy matching of S6, if the matching similarity is lower than the preset threshold (60%), the process of creating a new process flow template is triggered, and the system records the manual confirmation result in the process flow template library.

[0007] An MES automatic task allocation system for implementing the MES automatic task allocation method based on dynamic process matching and workload balance described in any one of the above, including: a process flow template library management module for storing and updating the process flow template library; a BOM parsing engine for extracting key fields and performing multi-level matching; a workload calculation unit for calculating the employee load value in real time; a task allocation optimization module for generating an optimal allocation plan based on the genetic algorithm; a mobile terminal interface for task pushing and status feedback.

[0008] Furthermore, the algorithm of the BOM parsing engine satisfies: , where S represents the comprehensive similarity score, is the weight factor of the i-th level material node, represents the BOM coding industrial knowledge graph embedding vector, represents the relevant label vector of the material. When S > 0.85, it triggers the automatic association of the process flow template library.

[0009] Furthermore, the fitness function F of the genetic algorithm is defined as: , where is the estimated production cycle, is the benchmark cycle, represents the load balance degree, is the equipment changeover cost, , , , where α represents a weight coefficient used to adjust the influence of the first term ; β represents another weight coefficient used to adjust the influence of the second term ; and γ represents a third weight coefficient used to adjust the influence of the third term .

[0010] Furthermore, the BOM coding industrial knowledge graph embedding vector satisfies three-dimensional constraints: represents the comprehensive constraint of equipment, personnel, and time. Among them, ⊗ is the tensor product, representing a certain combination of vectors; is the equipment coding vector with a length of n, and each element is 0 or 1, indicating whether the equipment is available; is the personnel coding vector with a length of m, and each element is also 0 or 1, indicating whether the personnel is available; is the start and end time of the process time window, indicating that the task needs to be completed within this time range.

[0011] Furthermore, the data transmission compression ratio n of the mobile terminal interface is calculated as: , where is the information entropy of the process operation code, is the parameter differential coding amount, is the original data amount. When the data transmission compression ratio n < 0.6, the lossless compression mode is enabled.

[0012] Compared with the existing technologies, the advantages of the present invention are as follows: The present invention proposes an MES automatic task allocation method and system based on dynamic process matching and workload balance, which can automatically update and calculate the process flow according to the process flow template library, allocate tasks to appropriate employees for task execution. It can not only help the production line better and more reasonably allocate resources, avoid injustice, but also help with the statistical analysis of new task execution data, and more importantly, save management costs and improve efficiency. Description of the Drawings

[0013] Figure 1 It is a flow framework diagram of the MES automatic task allocation method in the present invention; Detailed Embodiments

[0014] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0015] Example 1, please refer to Figure 1 In [reference], an MES automatic task allocation method based on dynamic process matching and workload balance includes the following steps: S1. Manually create a set of standard production task processes in the MES system, and the processes include at least one of sawing, drilling, milling, riveting, welding, boring, turning, and grinding; S2. Create a process flow template library corresponding to the standard production task, and the process flow template library contains the processes and man-hour requirements for producing parts; S3. Generate a BOM list through the PLM system, and the BOM list at least includes part name, specification, model, and raw material information; S4. Import the BOM list into the MES system; S5. Match the keyword fields (part name, specification, model, raw material information) in the BOM list with the process flow template library to determine the process flow required for producing the part; S6. Dynamically determine the matching result: If there are completely matching keyword fields, automatically generate the corresponding process flow, automatically generate production tasks, and notify the manager; If there is no complete match, perform fuzzy matching in the order of priority (part name > specification > model > raw material), generate candidate process flows for the planning team to confirm, and update and save them to the process flow template library in S2 after confirmation; S7. Task allocation stage: S7.1 Calculate the real-time workload load value W of employee i according to the following formula i : , where is the standard man-hour of the kth unfinished process, is the process priority coefficient (value range 1.0 - 2.0), is the number of unfinished tasks of employee i, is the task number weight coefficient (default 0.5); S7.2 Select employees who meet and are jointly matched, sort them in ascending order according to the current workload (real-time workload load value), and preferentially assign tasks to the employees with the smallest load to achieve load balancing; where W i represents the current workload, and W threshold represents the preset threshold; S8. The production task is pushed to the employee account through the MES system mobile terminal interface.

[0016] In the above embodiment, the following data can be set: The number of tasks n = 3 that employee i is responsible for, and the standard working hours of each task and the priority coefficient are as follows: Task 1: t1 = 4 hours, p1 = 1.5, Task 2: t2 = 6 hours, p2 = 1.8, Task 3: t3 = 2 hours, p3 = 1.2, Set the number of unfinished tasks of employee i currently , the task number weight coefficient , Calculate the load contribution of each task: , Task 1: , Task 2: , Task 3: , Calculate the total load contribution of all tasks: , Calculate the load contribution of the number of unfinished tasks of employee i currently: , Calculate the real-time workload value of employee i , The real-time workload load value of employee i is 21.7.

[0017] According to S7.2, we need to meet and the jointly matched employees are sorted in ascending order according to the current workload (real-time workload load value), and tasks are preferentially assigned to the employees with the smallest load. Suppose it is 25, then the load value of this employee meets the conditions and can be selected for task assignment.

[0018] Embodiment 2, on the basis of the above Embodiment 1, in S7.1, the calculation method of the priority coefficient is: , where is the current date, where is the current date, is the planned completion date of process k, 、 are respectively the latest task deadline and the earliest task deadline within the system.

[0019] In the above embodiment, the following data are set: Current date: , Planned completion date of process k: , Latest task deadline within the system: , Earliest task deadline within the system: , Calculate the difference between the dates: , (Note: Take the absolute value for calculation here), Calculate the priority coefficient : , Priority coefficient of process k is 2.1.

[0020] From this result, it shows that the priority coefficient of process k is 2.1, which means its priority in the current task is the highest because its planned completion date is later than the current date and close to the latest task deadline within the system. This indicates that this process needs to be completed as soon as possible to avoid affecting the progress of the entire project. In practical applications, the calculation of dates may need to consider specific date processing methods, such as using a date difference function to directly calculate the number of days difference instead of manual calculation. In addition, if is greater than , it means that the current date has exceeded the planned completion date and special handling or adjustment is required, which means that the current date has exceeded the planned completion date and special handling or adjustment is required.

[0021] Embodiment 3, based on the above Embodiment 1, when performing fuzzy matching in S6, if the matching similarity is lower than the preset threshold (60%), a new process flow template library creation process is triggered, and the system records the manual confirmation result in the knowledge base.

[0022] Example 4. A MES automatic task allocation system includes: a process template management module for storing and updating standard process flows; a BOM parsing engine for extracting keyword fields and performing multi-level matching; a workload calculation unit for calculating the employee load value in real time; a task allocation optimization module for generating an optimal allocation plan based on a genetic algorithm; a mobile terminal interface for task push and status feedback. On the basis of the above embodiments, the data transmission compression rate of the mobile terminal interface is calculated as: , where is the information entropy of the process operation code, is the parameter differential coding amount, is the original data amount. When the data transmission compression rate n < 0.6, the lossless compression mode is enabled.

[0023] In this embodiment, we have a mobile terminal interface in a MES automatic task allocation system and need to calculate its data transmission compression rate. We use a simple example to show how to apply this formula: , , , Calculate the sum of the information entropy and the differential coding amount: , Convert bits to bytes (because D is in bytes): , Calculate the data transmission compression rate: .

[0024] Determine whether to enable the lossless compression mode: Since the data transmission compression rate n = 0.95625 is greater than 0.6, the lossless compression mode is not enabled. In the above embodiment, the data transmission compression rate is 95.625%, which means that the data can be compressed to about 4.375% of the original size during transmission. Since the compression rate is greater than 60%, according to the given conditions, we do not enable the lossless compression mode. The embodiment shows how to use the given formula to calculate the data transmission compression rate and determine whether to enable the lossless compression mode according to the data transmission compression rate. In practical applications, the parameters can be adjusted according to the specific data volume and information entropy.

[0025] This MES automatic task allocation system aims to optimize the production process and improve efficiency and accuracy by integrating multiple modules. The MES automatic task allocation system mainly includes five modules: the process template management module, the BOM parsing engine, the workload calculation unit, the task allocation optimization module, and the mobile terminal interface. Among them, the functions of the process template management module are: storing and managing standard process flow templates; operation process: users can add, delete, or update the process flow template library; the template includes detailed steps of the process, required materials, estimated working hours, etc.; the system provides version control functions to facilitate tracking and backtracking of changes; Functions of the BOM parsing engine: Extract key fields from the BOM list and perform multi-level matching; operation process: import the BOM list; the engine parses the file and extracts key fields such as material numbers, quantities, specifications, etc.; perform multi-level matching according to preset rules to ensure the correctness and availability of materials; Functions of the workload calculation unit: Calculate the real-time workload value of employees; operation process: collect employees' task allocation information and completion status; calculate the real-time workload value of each employee according to the formula; the workload value is used for subsequent task allocation optimization; Functions of the task allocation optimization module: Generate the optimal task allocation plan based on the genetic algorithm; operation process: input the workload value and task requirements of employees; run the genetic algorithm and iteratively search for the optimal allocation plan; output the optimized task allocation plan; Functions of the mobile terminal interface: Used for task push and status feedback; operation process: the system pushes tasks to employees through the mobile terminal; employees feedback the task completion status through the mobile terminal; the system updates the task status and employees' workload values in real time.

[0026] Through the above embodiments, the system can achieve the following effects: improve the efficiency and accuracy of task allocation; reduce the workload of employees, improve job satisfaction, and the overall efficiency and response speed of the production process.

[0027] In Embodiment 5, the semantic matching algorithm of the BOM parsing engine satisfies: , where S represents the comprehensive similarity score, is the weight factor of the i-th level material node, represents the BOM coding industrial knowledge graph embedding vector, represents the relevant label vector of the material. When S > 0.85, the automatic association of the process flow template is triggered.

[0028] In this embodiment, the following data is set: The number of material nodes n = 3, The weights of the material nodes W1 = 0.5, W2 = 0.3, W3 = 0.2, BOM coding industrial knowledge graph embedding vector: , Label embedding vector: , Calculation process: Calculate the similarity of each node: For , ; For , ; For , , Calculate the total similarity: .

[0029] According to the calculation of the above embodiments, the total similarity S is 0.992, which is greater than 0.85. Therefore, the automatic association of the process template is triggered. The above embodiments show how to calculate the semantic matching similarity of the BOM parsing engine according to the given formula and data, and determine whether to trigger the self - definition association of the process template according to the similarity.

[0030] Embodiment 6, the fitness function F of the genetic algorithm is defined as: , where is the estimated production cycle, is the benchmark cycle, represents the load balance degree, is the equipment change - over cost, , , , In this embodiment, the following data can be set: Estimated production cycle = 10 days, Benchmark cycle = 5 days, Load balance degree , Equipment change - over cost The values of

[0031] Calculation process: Calculate the exponential part of the fitness function F: , Calculate the first term of the fitness function F: , Calculate the second term of the fitness function F: , Calculate the third term of the fitness function F: , Calculate the fitness function F: 。

[0032] Example 7, based on the above embodiments, the BOM code satisfies three-dimensional constraints: , where is the device coding vector, with a length of n, and each element is 0 or 1, indicating whether the device is available; is the personnel coding vector, with a length of m, and each element is also 0 or 1, indicating whether the personnel is available; is the start and end time of the process time window, indicating that the task needs to be completed within this time range.

[0033] In the above embodiments, assume there is a simple production line that includes two devices and three workers. We need to allocate devices and personnel to each process of this production line and determine the process time window boundaries; Parameter definition: The number of devices A = 2, the number of personnel m = 3, the start time t s = 1 of the process time window and the end time t e = 10; Coding vector: Device coding vector , , personnel coding vector , ; Start and end time of the process time window: t s = 1 and t e = 10 respectively indicate that the process starts at the 1st time unit and ends at the 10th time unit; Calculation process: According to the formula , we need to calculate and of the Kronecker product, and then combine it with the process time window boundary ; Calculate the Kronecker product : , Combine the process time window boundary: Since the result is a 6×1 matrix, we can regard it as a grid, where each element corresponds to a possible device-personnel combination.

[0034] Process time window boundary indicates that each device-personnel combination starts at the 1st time unit and ends at the 10th time unit.

[0035] This embodiment demonstrates how to use the given formula to allocate equipment and personnel and determine the process time window boundaries. In practical applications, parameters and coding vectors can be adjusted according to specific production line configurations and requirements.

[0036] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. An MES automatic task allocation method based on dynamic process matching and workload balancing, characterized in that, It includes the following steps: S1. Manually create a set of standard production task processes in the MES system, where the processes include at least one of sawing, drilling, milling, riveting, welding, boring, turning, and grinding; S2. Create a process flow template library corresponding to the standard production task, and the process flow template library contains the processes and man-hour requirements for producing parts; S3. Generate a BOM list through the PLM system, and the BOM list at least includes part name, specification, model, and raw material information; S4. Import the BOM list into the MES system; S5. Match the BOM list with the process flow template library through key fields (part name, specification, model, raw material information) in the BOM list to determine the process flow required for producing the part; S6. Dynamically determine the matching result: If there are completely matching key fields, automatically generate the corresponding process flow, automatically generate a production task, and notify the manager; If there is no complete match, perform fuzzy matching in the order of priority (part name > specification > model > raw material), generate a candidate process flow for the planning team to confirm, and after confirmation, update and save it to the process flow template library in S2; S7. Task assignment stage: S7.1 Calculate the real-time workload value \(W\) of employee \(i\) according to the following formula i : , where is the standard working hours of the \(k\)-th unfinished process, is the process priority coefficient (value range 1.0 - 2.0), is the number of unfinished tasks of employee \(i\), is the task number weight coefficient (default 0.5); S7.2 Select employees who meet and match jointly, sort them in ascending order of the current workload (real-time workload load value), and preferentially assign tasks to the employee with the smallest load to achieve load balancing; where W i represents the current workload, and W threshold represents the preset threshold. S8. Push the production task to the employee account through the MES system mobile terminal interface.

2. The MES automatic task allocation method based on dynamic process matching and workload balance according to claim 1, wherein The process priority coefficient in S7.1 is calculated as follows: , wherein, is the current date, is the planned completion date of process k, , are respectively the latest task deadline and the earliest task deadline within the system.

3. A MES automatic task allocation method based on dynamic process matching and workload balance according to claim 1, characterized in that, During the fuzzy matching in S6, if the matching similarity is lower than the preset threshold (60%), trigger a new process flow template creation process, and the system records the manual confirmation result in the process flow template library.

4. An MES automatic task allocation system for implementing the MES automatic task allocation method based on dynamic process matching and workload balancing according to any one of claims 1-3, characterized in that, It includes: A process flow template library management module for storing and updating the process flow template library; A BOM parsing engine for extracting key fields and performing multi-level matching; A workload calculation unit for calculating the employee load value in real time; A task assignment optimization module for generating an optimal assignment plan based on the genetic algorithm; A mobile terminal interface for task push and status feedback.

5. An MES automatic task allocation system according to claim 4, characterized in that The algorithm of the BOM parsing engine satisfies: , where S represents the comprehensive similarity score, is the weight factor of the i-th level material node, represents the BOM coding industrial knowledge graph embedding vector, represents the relevant label vector of the material. When S > 0.85, it triggers the automatic association of the process flow template library.

6. The MES automatic task allocation system according to claim 4, wherein The fitness function F of the genetic algorithm is defined as: , where is the estimated production cycle, is the benchmark cycle, represents the load balancing degree, is the equipment changeover cost, , , , where α represents a weight coefficient for adjusting the influence of the first term ; β represents another weight coefficient for adjusting the influence of the second term ; and γ represents a third weight coefficient for adjusting the influence of the third term .

7. An MES automatic task allocation system according to claim 5, characterized in that, The BOM-coded industrial knowledge graph embedding vector satisfies three-dimensional constraints: Represents the comprehensive constraints of equipment, personnel, and time, where ⊗ is the tensor product, representing a certain combination of vectors; Is the equipment coding vector, with a length of n, and each element is 0 or 1, indicating whether the equipment is available; Is the personnel coding vector, with a length of m, and each element is also 0 or 1, indicating whether the personnel is available; Is the start and end time of the process time window, indicating that the task needs to be completed within this time range.

8. An MES automatic task allocation system according to claim 4, wherein, The data transmission compression ratio n of the mobile terminal interface is calculated as follows: , where is the information entropy of the process operation code, is the parameter differential coding amount, is the original data amount. When the data transmission compression ratio n < 0.6, the lossless compression mode is enabled.

Images