Process operating system and process operating method

Abstract

A process operation method, comprising: receiving operation information by a process control platform, extracting tasks in the operation information by a semantic analysis method, and publishing the tasks; receiving the tasks by a process operation platform and storing the tasks in a task queue; wherein the process operation platform, after receiving the tasks, processes the tasks according to task definitions and sorts the order of the tasks in the task queue; and an endpoint task robot actively obtains the tasks from the process operation platform, executes the tasks, writes the execution results into a log queue, and transmits the execution results to the process control platform.

Description

Technical Field

[0001] This invention relates to an operating system, and more particularly to a process operating system and process operation method for automation. Background Technology

[0002] In recent years, with the significant improvement in hardware computing power, machine learning algorithms have also enhanced their ability to recognize speech and images. In addition, the successful training of multi-layer neural networks has opened up unlimited possibilities for artificial intelligence applications. Enterprises have also invested a lot of manpower and resources in the field of artificial intelligence technology. For example, they are developing robotic process automation (RPA) with artificial intelligence to complete the daily tasks that were previously done by humans.

[0003] However, in the process of robotic process automation, when setting up endpoint task robots, execution permissions must be granted to the deployment management center. High-tech enterprises often have regulations requiring regular changes to machine passwords, necessitating frequent maintenance by operations personnel, which is quite troublesome. Furthermore, endpoint task robots are typically used on the user's end, and users may overlook version updates, resulting in outdated versions and task failures, increasing the burden on operations personnel to assist users in updating endpoint task robots.

[0004] Therefore, how to develop a lightweight process operating system that allows people without an engineering background to edit automated processes and significantly reduce the error rate caused by the tedious work that used to require a lot of employee effort has become one of the problems that need to be solved in this field. Summary of the Invention

[0005] To address the aforementioned issues, this disclosure provides an example of a process operating system. The process operating system includes a process control platform, an endpoint task robot, and an endpoint task robot. The process control platform receives operation information, extracts tasks from the operation information using semantic analysis, and publishes the tasks. The process operating platform receives tasks and stores them in a task queue. Upon receiving a task, the process operating platform processes it according to the task definition and sorts the tasks in the task queue. The endpoint task robot proactively retrieves tasks from the process operating platform, executes the tasks according to the processing flow, writes the execution results to a log queue, and transmits the execution results back to the process control platform.

[0006] Another example of the present invention relates to a process operation method, comprising: receiving operation information through a process control platform, extracting tasks from the operation information through semantic analysis, and publishing the tasks; receiving tasks through the process operation platform and storing the tasks in a task queue; wherein, after receiving the tasks, the process operation platform processes the tasks according to the task definition and sorts the tasks in the task queue; and actively obtaining tasks from the process operation platform through an endpoint task robot, executing the tasks according to the processing flow, writing the execution results to a log queue, and transmitting the execution results to the process control platform.

[0007] The process operating system and process operation method described in this case can use a library to store the components and parameters required to execute tasks. If there are new components or changes to old components, only the library needs to be updated, without affecting the overall system and logic. Users can generally compile their task processes on a configuration webpage according to their task requirements, saving users the cost of installing recording tools and learning new skills. The process control platform publishes tasks to the process operation platform, which then sends the tasks to the task queue. Tasks are received and executed by idle endpoint task robots. Therefore, any entity or virtual host can act as an endpoint task robot to execute tasks. There is no need to maintain endpoint task robot account passwords on the process operation platform, thus transferring the control responsibility of endpoint task robots to the user end, saving maintenance personnel management effort. After the endpoint task robot completes execution, it will notify both maintenance personnel and the user end, regardless of whether the execution is successful or not, bridging the gap between them and allowing for immediate and optimal follow-up measures. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of a process operating system according to an embodiment of the present invention.

[0009] Figure 2 This is a flowchart illustrating a process operation method according to an embodiment of the present invention.

[0010] Figure 3 This is a schematic diagram illustrating an example of a process operation method according to an embodiment of the present invention.

[0011] Figure 4 This is a schematic diagram of a management platform according to an embodiment of the present invention.

[0012] Figure 5 This is a schematic diagram of a host information page illustrated according to an embodiment of the present invention.

[0013] Figure 6 This is a schematic diagram illustrating a history record interface according to an embodiment of the present invention.

[0014] Figure 7This is a schematic diagram illustrating an example of a robot execution interface according to an embodiment of the present invention.

[0015] Figure 8 This is a schematic diagram of a robot execution interface according to an embodiment of the present invention.

[0016] Figure 9 This is a schematic diagram illustrating the process writing result according to an embodiment of the present invention.

[0017] Figure 10 This is a schematic diagram of an information reply interface according to an embodiment of the present invention. Detailed Implementation

[0018] The following description illustrates a preferred embodiment of the invention and is intended to describe the basic spirit of the invention, but is not intended to limit the invention. The actual scope of the invention must be determined by referring to the claims that follow.

[0019] It must be understood that the words “comprising” and “including” used in this specification are used to indicate the presence of specific technical features, values, method steps, work processes, components and / or components, but do not preclude the addition of more technical features, values, method steps, work processes, components, components, or any combination thereof.

[0020] The use of terms such as "first," "second," and "third" in claims to modify components in the claims is not intended to indicate a priority order, a precedence relationship, or that one component precedes another, or the chronological order of the execution of method steps; it is only used to distinguish components with the same name.

[0021] Please refer to Figure 1 , Figure 1 This is a schematic diagram of a process operating system 100 according to an embodiment of the present invention. In one embodiment, as... Figure 1 As shown, the process operating system 100 includes a process control platform 10, a process operation platform 20, and an endpoint task robot 30. In one embodiment, the process control platform can be implemented using a server, laptop, or desktop computer. In one embodiment, the process operation platform 20 can be implemented using a laptop, desktop computer, or virtual machine. In one embodiment, the endpoint task robot 30 can be implemented using a laptop, desktop computer, or virtual machine.

[0022] In one embodiment, the process operation platform 20 has the functionality of Robotic Process Automation (RPA).

[0023] In one embodiment, an entity (e.g., a server) or a virtual machine (e.g., a virtual machine installed on a server) can both serve as an endpoint task robot 30. Generally, an entity or a virtual machine serves as an endpoint task robot 30.

[0024] In one embodiment, the process control platform 10, the process operation platform 20, and the endpoint task robot 30 each include a processor, a microcontroller, a microprocessor, a digital signal processor, an application-specific integrated circuit (ASIC), or a logic circuit.

[0025] In one embodiment, the process control platform 10 includes a storage device, which may be implemented using a read-only memory, flash memory, floppy disk, hard disk, optical disk, USB flash drive, magnetic tape, a network-accessible database, or other storage media with the same function that can be easily conceived by those skilled in the art.

[0026] Please refer to Figures 1 through 4. Figure 2 This is a flowchart illustrating a process operation method 200 according to an embodiment of the present invention. The process operation method 200 can be... Figure 1 The process is implemented in the operating system 100. Figure 3 This is a schematic diagram illustrating an example of a process operation method 200 according to an embodiment of the present invention. Figure 4 This is a schematic diagram of a management platform 400 according to an embodiment of the present invention.

[0027] In step 210, the process control platform 10 receives operation information, extracts the tasks from the operation information through semantic analysis, and publishes the tasks.

[0028] In one embodiment, such as Figure 3 As shown, the process control platform 10 is responsible for providing a user interface that allows users to start a series of task initiation operations. The user interface can be achieved by opening a webpage CL1, an application CL2, or a scheduled action (e.g., email) through a user device (e.g., mobile phone, laptop, or desktop computer). After receiving the request, it will mainly be used as an information transmission with the process operation platform 20 (step S1).

[0029] In one embodiment, the process control platform 10 receives an email and retrieves a file from it. The process control platform 10 then extracts the task from the file. In one embodiment, this file is, for example, an Excel file. This Excel file is created by purchasing personnel who record various part numbers, quantities, and models to be ordered during procurement. This Excel file is then attached to an email and sent to the process control platform 10. Upon receiving the email, the process control platform 10 automatically detects the attachment, for example, by recognizing the email content (including attachment fields) or by recognizing the icons in each field of the email. It then downloads or opens the file, extracts the task (e.g., a procurement task) from the operation information using known semantic analysis methods, and publishes the task. Known semantic analysis methods include, for example, Natural Language Processing (NLP) algorithms, machine learning algorithms, text detection algorithms, parsing algorithms, etc.

[0030] In one embodiment, the process control platform 10 receives files transmitted from an application and retrieves tasks from those files. In one embodiment, the application may be, for example, a chatbot or a virtual voice assistant. The user inputs files through the application; for example, the user inputs "order processor, 10,000 units, model i7," and the process control platform 10 receives this information transmitted from the application. Alternatively, the user can directly input Excel spreadsheets, text files, or forms with specific formats into the application, and the process control platform 10 receives these files transmitted from the application.

[0031] In one embodiment, users or maintenance personnel communicate with the process control platform 10 through an interactive text-based application CL2, a webpage CL1 (e.g., the user enters text in the webpage), and / or email. The process control platform 10 uses known semantic analysis techniques to understand the user's or maintenance personnel's needs for the task and presents the list of tasks to be executed, the robot's relevant status, and the daily execution records on the management platform 400.

[0032] In one embodiment, such as Figure 4 As shown, the management platform 400 can be presented via a webpage. The management platform 400 displays information about each host N1 to N4. Each host N1 to N4 can be named in advance, for example, host N2 is named "DaVinci". This information includes the status of each host ST1 to ST4, the robot-related status, and the execution records IFO1 to IFO4 for the day.

[0033] In one embodiment, the management platform 400 further includes a viewing interface for each host. For example, if the host information buttons BT1 to BT4 are pressed, the corresponding host information page is displayed. For instance, if the host information button BT2 of host N2 is pressed, the host information page 450 for host N2 is displayed.

[0034] Figure 5 This is a schematic diagram of a host information page 450 according to an embodiment of the present invention. In one embodiment, when the host information button BT2 of host N2 is pressed, the host information page 450 corresponding to host N2 is displayed. The host information page 450 can be a pop-up window or a webpage. In one embodiment, the host information page 450 displays the host name, for example, "DaVinci Host Information" displayed in the upper left corner of the window or webpage, and displays that host N2 can obtain two endpoint task robot information A1 and endpoint task robot information A2 provided by the process operation platform 20. Next to endpoint task robot information A1, there is a robot status indicator STA1, and next to endpoint task robot information A2, there is a robot status indicator STA2. When robot status indicator STA1 is green, it means that endpoint task robot information A1 is in an idle state; when robot status indicator STA2 is green, it means that endpoint task robot information A2 is in an idle state; when robot status indicator STA1 is red, it means that endpoint task robot information A1 is in an execution state; when robot status indicator STA2 is red, it means that endpoint task robot information A2 is in an execution state. Figure 5 In the example, both endpoint task robot information A1 and endpoint task robot information A2 are in an idle state. Therefore, the robot status prompts STA1 and STA2 are both green.

[0035] However, this invention is not limited to using red and green to represent robot status prompts; other colors, texts, or symbols can also be used as robot status prompts.

[0036] In one embodiment, the two endpoint task robot information A1 and endpoint task robot information A2 provided by the process operation platform 20 also include the name of the endpoint task robot corresponding to each endpoint task robot information A1 and endpoint task robot information A2, the work being performed, and / or the running time.

[0037] In one embodiment, the host information page 450 may display a worksheet BOM1 to be executed, which contains information such as user name, event source, event name, start time, status, etc.

[0038] exist Figure 5In the example shown, since endpoint task robot information A1 and endpoint task robot information A2 are both in an idle state, the content of the pending execution worksheet BOM1 is empty. Conversely, if endpoint task robot information A1 and endpoint task robot information A2 are both in an idle state, but there are pending tasks in the pending execution worksheet BOM1, it means that the endpoint task robot corresponding to endpoint task robot information A1 and / or endpoint task robot information A2 is damaged, or there is an error in task queue 25.

[0039] In one embodiment, the management platform 400 further includes a viewing interface for each host. When the position of the robot-related status and the daily execution record IFO1 to IFO4 is pressed, the corresponding historical record interface is accessed. For example, when the position of the robot-related status and the daily execution record IFO2 of host N2 is pressed, the historical record interface 460 of host N2 is displayed.

[0040] Figure 6 This is a schematic diagram of a history record interface 460 according to an embodiment of the present invention. In one embodiment, when the robot-related status and daily execution record IFO2 of host N2 are pressed, the history record interface 460 of the corresponding host N2 is displayed. The history record interface 460 can be a pop-up window or presented as a webpage. In one embodiment, the history record interface 460 can query historical records by selecting the event source, status, entering the event name, and / or entering the username. For example, when the event start date is set to 11 / 25 / 2020, the event end date is set to 11 / 25 / 2020, the event source is set to all, the status is set to completed, and the event name and username are not specified, the history record table BOM2 displays the username, event source, event name, start time, run time, status, and / or action of the task performed under these conditions. The user name can be the user's name or the name of the host N2, "DaVinci". The event source can be a file sent by the user via email, allowing the process control platform 10 to retrieve the task from the file, or a scheduled task pre-set by the host N2 for a specific time period (such as searching for news, retrieving forms from a specific website periodically, etc.). The action field can be a file that records the entire task execution process, provided to the operation and maintenance personnel of the process control platform 10 for reference when needed.

[0041] In step 220, the process operation platform 20 is used to receive tasks and store them in the task queue 25; wherein, after receiving a task, the process operation platform 20 processes the task according to the task definition and sorts the order of the tasks in the task queue 25.

[0042] Please see Figure 3After retrieving the task from the file, the process control platform 10 transmits the task to the process operation platform 20 (step S2). In one embodiment, the process control platform 10 transmits the endpoint task robots 30-32 that need to execute the task, along with task-related information (such as historical records and host information), to the process operation platform 20. After confirming that everything is correct, the process operation platform 20 sends the task to the task queue 25 and continues to wait for the next task notification (step S3). When the process operation platform 20 receives multiple tasks, it calculates the priority execution value and sorts the execution priority of each task according to the priority execution value before storing it in the task queue 25. The endpoint task robots 30-32 of the process operation platform 20 find the current task to be executed in the task queue 25 according to this priority and actively retrieve the relevant information of this current task from the process operation platform 20 (step S4).

[0043] In one embodiment, endpoint task robots 30-32 may belong to different units within the company. Endpoint task robots 30-32 can see and retrieve all tasks from the process operation platform 20. However, endpoint task robots 30-32 will only perform tasks within their own defined scope. For example, endpoint task robot 31, belonging to the procurement unit, will only perform procurement-related tasks and will not execute tasks related to news gathering, even if it receives such a task. In one embodiment, if endpoint task robot 31 malfunctions, endpoint task robot 32 will automatically take over and temporarily handle procurement-related tasks.

[0044] In one embodiment, the process operation platform 20 processes tasks according to defined procedures. Specifically, the process operation platform 20 follows defined standard operating procedures (SOPs) for each task. For example, for procurement-related tasks, the process operation platform 20 automatically (or pre-sets) executes the task using components and parameters already present in a library. More specifically, the process operation platform 20 schedules the endpoint task robot 30 to perform the following steps for procurement-related tasks: opening a Chrome browser, clicking the login button, navigating to the main menu, selecting material price query, selecting export to Excel, uploading files, and sending back execution results, etc. The library can be pre-stored in the process operation platform 20.

[0045] However, this is only an example, and the detailed process will not be described in detail. In addition, the process operation platform 20 can define different processing flows according to different tasks (such as searching for news, regularly obtaining forms from specific websites, etc.).

[0046] In one embodiment, the process operation platform 20 employs the highest response ratio first (HRRN) scheduling algorithm, along with an importance parameter, to calculate a priority execution value. The higher the priority execution value, the faster the endpoint task robot will execute it. This allows the execution order of tasks to be defined based on the characteristics of the tasks.

[0047] The mathematical formula for the high response ratio scheduling algorithm is: Priority = (Waiting Time + Required Service Time) / Required Service Time. This means that for tasks with the same "waiting time," the shorter the "required service time," the easier it is to be executed first, accelerating the number of tasks completed; conversely, for tasks with the same "required service time," the longer the "waiting time," the easier it is to be executed first, avoiding long waiting times. This method effectively balances service time for tasks, but it doesn't consider the importance of tasks. Therefore, an importance parameter α is added for weight adjustment, allowing more important tasks to be executed faster. Thus, the priority execution value is the priority multiplied by the corresponding importance parameter α, where the importance parameter α is calculated as 2 / (L+1), and L represents a user-defined importance of 1 to 4, where 1 represents the most important, 2 the next most important, and so on, with larger numbers indicating less important tasks. Importance can be defined by the user beforehand, based on historical material shortage records, or the frequency of executing the same task within a certain period.

[0048] In one embodiment, the missing parts record file most frequently viewed by maintenance personnel indicates that the task is of high importance, and its importance L is set to 1; the missing parts record file less frequently viewed by maintenance personnel indicates that the task is of average importance, and its importance L is set to 2; the missing parts record file rarely viewed by maintenance personnel indicates that the task is of low importance, and its importance L is set to 3; and the historical record file that has never been opened indicates that the task is of very low importance, and its importance L is set to 4.

[0049] In one embodiment, for example, the task that is performed most frequently in a day is considered to be of higher importance and its importance L is set to 1. The task that is performed most frequently in a day is considered to be of higher importance and its importance L is set to 2, and so on.

[0050] The following table 1 is used as an example. When there are 5 tasks to be executed, the priority execution value and result are calculated using the above formula, and the order of tasks in task queue 25 is sorted according to the result.

[0051]

[0052] Table 1

[0053] As shown in Table 1, priority = (waiting time + required service time) / required service time. Therefore, it will not be elaborated here. Multiply the priority of each task by the corresponding importance parameter α to obtain the result. For example, task 1 has a priority of 1.5, multiplied by the importance parameter α = 1, resulting in a result of 1.5; task 2 has a priority of 2, multiplied by the importance parameter α = 0.67, resulting in a result of 1.34; task 3 has a priority of 2.5, multiplied by the importance parameter α = 0.4, resulting in a result of 1; task 4 has a priority of 3, multiplied by the importance parameter α = 0.5, resulting in a result of 1.5; and task 5 has a priority of 2, multiplied by the importance parameter α = 0.6, resulting in a result of 1.2. The result here refers to the priority execution value.

[0054] Based on the example in Table 1, the process operation platform 2 will, according to these 5 results, sequentially pass the tasks to the task queue 25 in the order of task 1, task 4, task 2, task 5, and task 3, to ensure that the more important and shorter the time task is, the faster it will be completed and executed by the endpoint robot (e.g., endpoint robots 30-32).

[0055] The results of both Task 1 and Task 4 are 1.5, but due to the first-in-first-out (FIFO) concept, Task 1 was received by the process operation platform 20 first, so its priority is higher than that of Task 4.

[0056] In one embodiment, the process operation platform 20 receives status and execution information from endpoint task robots 30-32, and transmits this status and execution information to the database for query and use by the process control platform 10.

[0057] In one embodiment, when the maintenance personnel release a new version of the robot, the process control platform 10 will automatically update the new version to the process operation platform 20, and the process operation platform 20 will send the version information to the version queue for endpoint robots 30-32 to listen for.

[0058] In step 230, the endpoint task robot (taking endpoint task robot 30 as an example) actively obtains a task from the process operation platform 20, executes the task according to the processing flow, writes the execution result to the log queue, and transmits the execution result to the process control platform 10.

[0059] In one embodiment, after the endpoint task robot 30 confirms the connection with the process operation platform 20 and the task queue 25, the user can log in to the endpoint task robot 30 with the account and password of the endpoint task robot 30 and start running it. With the installation of additional agent devices, the endpoint task robot 30 can start accepting tasks. This method increases the flexibility of enterprises in executing tasks.

[0060] In this way, users can log in to Endpoint Task Robot 30 with their account and password, and maintenance personnel do not need to ask users for their account and password.

[0061] Among them, the tasks actively obtained by the endpoint task robot 30 from the process operation platform 20 include the following information: the person who initiated the task, the device that initiated the task, the robot that needs to perform the task, the globally unique identifier (GUID) of the email data, the parameter code, the operating environment and task code, etc. Through this data, the endpoint task robot 30 can know what kind of task to perform, where to retrieve data and report the task.

[0062] In one embodiment, endpoint task robots 30-32 will proactively obtain the corresponding tasks from the process operation platform 20 according to the order of the sorted tasks in the task queue 25.

[0063] Please refer to Figures 7 and 8. Figure 7 This is a schematic diagram illustrating an example of a robot execution interface 700 according to an embodiment of the present invention. Figure 8 This is a schematic diagram of a robot execution interface 750 according to an embodiment of the present invention.

[0064] In one embodiment, when the current state of the endpoint task robot 30 is idle and the process operation platform 20 issues an updated version, the endpoint task robot 30 updates according to the updated version.

[0065] In one embodiment, the endpoint task robot (e.g., endpoint task robot 30) explicitly displays the current version number 72 in the robot execution interface 700 (e.g., ...). Figure 7 (Version shown is 1.0.0.57). When the endpoint task robot 30 receives the new version information of the process operation platform 20 from the version queue, the new version notification 70 will display a hyperlink (or the new version notification 70 can be designed as a button, whose button status changes from unclickable gray to clickable black text). The endpoint task robot 30 confirms the current version number 72 and confirms that there are no tasks currently running (status is standby). At this time, the endpoint task robot 30 will pause receiving tasks and start automatically downloading the new version robot software and updating the version of the endpoint task robot 30. Finally, after the update is completed, the endpoint task robot 30 will be automatically activated according to the previous settings. Therefore, the maintenance personnel do not need to monitor or operate during the upgrade process. This shortens the upgrade steps and reduces the workload of the maintenance personnel, which is a great help in adopting the automation of the process operation platform 20.

[0066] In one embodiment, the robot execution interface 700 includes a display of the robot name, a start button 73, a stop button 74, and / or a task field 75. In one embodiment, when a task is available, the operator can manually press the start button 73 to start the endpoint task robot 30 to execute the task. In another embodiment, when a task error occurs, the operator can manually press the stop button 74 to stop the endpoint task robot 30 from executing the task.

[0067] exist Figure 7 In the task field 75, there are no tasks, so endpoint task robot 30 is in standby mode.

[0068] In one embodiment, the idle endpoint task robot 30 listens to the corresponding task queue 25 according to the operating environment 71 (development, simulation or production environment). This setting facilitates the operation and maintenance personnel to conduct pre-launch testing of tasks. The endpoint task robots 30-32 listen to and obtain information including task code, necessary parameters and the robot responsible for execution (e.g., endpoint task robot 30).

[0069] In one embodiment, when the endpoint task robot 30 hears a task that is not its responsibility, it ignores it and does not process it. The task is still kept in the queue. When the endpoint task robot 30 receives a task that it is responsible for, it will accept the task and start execution to avoid the task being executed repeatedly. At this time, the endpoint task robot 30 pauses listening to the task queue to prevent a single robot from executing multiple tasks at the same time, which would affect the result.

[0070] In one embodiment, the task is divided into a user-defined webpage operation process and a custom-developed software execution file. After the endpoint task robot 31 and endpoint task robot 32 confirm the task type, the endpoint task robot 31 simulates human operations on the webpage according to the task process, such as opening the webpage, clicking buttons and entering values, and finally completes the predetermined task. The endpoint task robot 32 then passes the parameters to the execution file, which completes the operation and returns the execution result. The similarity between the two is that after the task is completed, the endpoint task robots 31 and 32 switch to a waiting state and listen to the status queue, and finally report the execution result (execution success, execution failure or execution timeout) to the process control platform 10 (step S5).

[0071] In one embodiment, endpoint task robots 31 and 32 may selectively report the execution results to the user device (step S6), or the process control platform 10 may report the execution results to the user device.

[0072] In one embodiment, when a task is successfully executed, the process control platform 10 will transmit the execution result to the process operation platform 20 and then to the status queue. The endpoint task robot (e.g., endpoint task robot 31) in the waiting state will stop waiting after receiving the completion of its own task and change its status to the idle state, and then start a new round of task listening.

[0073] In one embodiment, if the task execution fails, the process control platform 10 notifies the task owner (user) to confirm the problem and then the endpoint task machine (e.g., endpoint task machine 31).

[0074] In one embodiment, the process operating system 100 further includes a log module, which can be virtualized on a server or implemented by the server itself. The log module includes a log queue. To ensure complete records for each task execution without any omissions, in addition to complete logs on the local end of the endpoint task robots 30-32, message queue technology is used. The endpoint task robots send each task execution record to the log queue. The process operating platform 20 also has a task status agent to listen to information in the log queue, receive the information, and correctly write it to the database (which can be stored on the server's storage device). This approach has the advantage of unified database reading and writing by the task status agent compared to each endpoint task robot 30-32 writing to the database individually, as it reduces the risk of database deadlocks and speeds up database processing.

[0075] In one embodiment, the process operating system 100 further includes a notification module, which can be virtualized on a server or implemented by the server. Regardless of whether the endpoint task robots 30-32 complete, fail, or time out, in addition to reporting the execution results to the process operating platform 20, the endpoint task robots 30-32 will also report the execution results to the notification module. The notification module will then send notifications (e.g., emails) to the electronic devices of the task owner (e.g., the user) and relevant managers (e.g., the user's department head and / or colleagues). After the notification module sends the notification, the maintenance personnel can check the log queue and find the corresponding execution process file in the database to understand the cause of the error as soon as possible and take appropriate measures, greatly reducing the time spent on abnormal delays.

[0076] In one embodiment, the various queues mentioned in this invention (such as task queues, log queues, status queues, version queues, etc.) can all be stored on the same server's storage device or in the virtual machine's storage space.

[0077] Please refer to Figures 9 and 10. Figure 9 This is a schematic diagram illustrating the process writing result 900 according to an embodiment of the present invention. Figure 10 This is a schematic diagram of the information response interface RPLY according to an embodiment of the present invention.

[0078] In one embodiment, the company can effectively shorten development, testing, and deployment time through this process operating system 100. The specific process is as follows: when a business unit wants to convert the current process into an endpoint task robot, it will contact the developers in the management information center. After confirmation, the developers will apply for robots and skills on the process control platform 10 (also known as the artificial intelligence platform) and start editing the task process. They will extract the operations performed by humans on web pages, desktops, or third-party systems, and match them with the components provided on the process operation platform 20 (also known as the RPA platform). With the logical judgment of the requirements, the written process result 900 is completed. The written process result 900 can be regarded as the actual task of querying material prices and uploading them to the company's internal system.

[0079] Then, the process control platform 10 is used to initiate the task execution requirement. The endpoint task robot (such as endpoint task robot 31) is used to select a test environment for task testing. After receiving the task, the endpoint task robot operates as if it were a human, and performs the planned system operations in sequence. Developers can directly view the task execution process on the screen. After the test is completed, it can be directly transferred to the production environment for deployment.

[0080] Compared to the past processes that involved a lot of manual operation and required manpower and time to develop, test and launch the required processes, taking about 1 to 3 months, the process operating system 100, which uses automated processes, effectively shortens the development, testing and launch time to about half a month. Therefore, the process operating system 100 saves more than 50% of the time.

[0081] During testing, the Process Operating System 100 executed over 3,500 tasks, with each robot averaging 16.6 tasks per day (such as RPA tasks). These tasks covered various departments including procurement, R&D, and sales, saving 1-5 hours of human time per week compared to manual tasks. For example, taking a task involving querying material prices and uploading them to the company's internal system, each column in the file represents the robot's steps on the webpage, including clicking to open the webpage, entering values, and clicking buttons. The process can include logical checks for success or failure. Each step is fully logged, and upon task completion, the system replies to the process control platform and notifies the task owner (e.g., via...). Figure 10 The RPLY message reply interface (shown) notifies the owner of the task, completing a full operation that is completely automated and requires no human intervention or judgment.

[0082] The process operating system 100 and process operation method 200 described in this case can use a library to store the components and parameters required to execute tasks. If there are new components or changes to old components, only the library needs to be updated, without affecting the overall system and logic. Users can generally compile their task processes on a configuration webpage according to their task requirements, saving users the cost of installing recording tools and learning. The process control platform publishes tasks to the process operation platform, which then sends the tasks to the task queue. Tasks are received and executed by idle endpoint task robots. Therefore, any entity or virtual host can act as an endpoint task robot to execute tasks. There is no need to maintain endpoint task robot account passwords on the process operation platform, transferring the control responsibility of endpoint task robots to the user end, saving maintenance personnel management effort. After the endpoint task robot completes execution, it will notify both maintenance personnel and the user end, regardless of whether the execution is successful or not, reducing the gap between them and allowing for immediate and optimal follow-up measures.

[0083] In addition, the process operating system 100 and process operation method 200 provide complete step-by-step log records of task execution for operation and maintenance personnel to query and audit. Furthermore, after operation and maintenance personnel release a new version of the endpoint task robot, the backend system will automatically place the upgrade file on the server and send the version information to the version queue. The endpoint task robot listens for version queue information at any time. Once received, the robot execution interface displays the version upgrade information, allowing the endpoint task robot to automatically upgrade or be upgraded by the user when idle. This method shifts the decision of version update time from operation and maintenance personnel to users, or allows the endpoint task robot to automatically update when idle, reducing the effort required for operation and maintenance personnel to manage endpoint task robot versions.

[0084] The method, or a specific form or part thereof, of the present invention may exist in the form of program code. The program code may be contained in physical media, such as floppy disks, optical discs, hard disks, or any other machine-readable (e.g., computer-readable) storage media, or may be a computer program product, not limited to an external form, wherein when the program code is loaded and executed by a machine, such as a computer, that machine becomes an apparatus for participating in the present invention. The program code may also be transmitted via some transmission medium, such as wires or cables, optical fibers, or any transmission method, wherein when the program code is received, loaded, and executed by a machine, such as a computer, that machine becomes an apparatus for participating in the present invention. When implemented in a general-purpose processing unit, the program code, in conjunction with the processing unit, provides a unique apparatus that operates similarly to application-specific logic circuitry.

[0085] Although the present invention has been disclosed above with reference to embodiments, it is not intended to limit the present invention. Anyone skilled in the art can make various modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

[0086] [Symbol Explanation]

[0087] 100: Process Operating System

[0088] 10: Process Control Platform

[0089] 20: Process Operation Platform

[0090] 30-32: Endpoint Task Robot

[0091] 200: Process Operation Method

[0092] 210~230, S1~S6: Steps

[0093] CL1: Webpage

[0094] CL2: Application

[0095] 25: Task Queue

[0096] 400: Management Platform

[0097] N1~N4: Main unit

[0098] BT1~BT4: Host Information Buttons

[0099] ST1~ST4: Host Status

[0100] IFO1~IFO4: Robot status and daily execution records

[0101] 450: Host Information Page

[0102] A1, A2: Endpoint Task Robot Information

[0103] BOM1: Worksheet to be executed

[0104] STA1, STA2: Robot status prompts

[0105] 460: History Record Interface

[0106] BOM2: Historical Records Table

[0107] 700, 750: Robot Execution Interface

[0108] 70: New Version Notification

[0109] 71: Operating Environment

[0110] 72: Current version number

[0111] 73: Start button

[0112] 74: Stop button

[0113] 75: Task Field

[0114] 900: Process Writing Results

[0115] RPLY: Message Reply Interface

Claims

1. A process operating system, comprising: The process control platform includes a first processor, which is used to receive operation information, extract the task from the operation information through semantic analysis, and issue the task. A process operation platform includes a second processor, which receives the task and stores it in a task queue; upon receiving the task, the second processor defines a processing flow based on the task and sorts the tasks in the task queue; and The endpoint task robot includes a third processor, which actively retrieves the task from the process operation platform, executes the task according to the processing flow, writes the execution result to a log queue, and sends the execution result back to the process control platform. Specifically, when the current state of the endpoint task robot is idle and the second processor of the process operation platform issues an updated version, the endpoint task robot updates according to the updated version.

2. The process operating system as described in claim 1, wherein, The first processor of the process control platform receives an email, retrieves a file from the email, and then retrieves the task from that file.

3. The process operating system as described in claim 1, wherein, The first processor of the process control platform receives a file from the application and retrieves the task from that file.

4. The process operating system as described in claim 1, wherein, The second processor of the process operation platform uses the highest response ratio next (HRRN) scheduling algorithm, along with an importance parameter, to calculate the priority execution value. The higher the priority execution value, the faster the task robot at that point will execute it.

5. A process operation method, comprising: The process control platform receives operation information through its first processor, extracts the task from the operation information using semantic analysis, and then publishes the task. The task is received by the second processor of the process operation platform and stored in the task queue; wherein... After receiving the task, the second processor processes the task according to the task definition and sorts the task in the task queue. as well as The endpoint task robot actively retrieves the task from the process operation platform via its third processor, executes the task according to the processing flow, writes the execution result to the log queue, and sends the execution result back to the process control platform. Specifically, when the current state of the endpoint task robot is idle and the second processor of the process operation platform issues an updated version, the endpoint task robot updates according to the updated version.

6. The process operation method as described in claim 5, further comprising: The first processor of the process control platform receives an email, retrieves a file from the email, and obtains the task from the file.

7. The process operation method as described in claim 5, further comprising: The first processor of the process control platform receives a file transmitted from the application and retrieves the task from that file.

8. The process operation method as described in claim 5, further comprising: The second processor of the process operation platform uses the highest response ratio first (HRRN) scheduling algorithm, along with an importance parameter, to calculate the priority execution value. The higher the priority execution value, the faster the task will be executed by the third processor of the robot at that point.

Images