A limited space operation approval process intelligent pushing method and system

Abstract

This invention belongs to the field of intelligent push notifications, and particularly relates to an intelligent push notification method and system for confined space operation approval processes. The method includes: obtaining the identifier of the pending approval node and the associated set of approval parameters for the target operation application; mapping and formatting the operation application information according to the mandatory approval field requirements to generate a dedicated approval form for each node; refreshing the real-time approval task cache with the dedicated form when a push notification is detected; collecting the approver's running status feedback value, matching it with the permission list to obtain the permission verification result, and comparing it with the timeliness threshold to obtain the timeliness detection result; if the permission verification is successful and the timeout has not occurred, and if the task is completed and the information is complete, generating a push priority based on the next node approver's permission level and historical response time, and simultaneously pushing it to their APP and PC terminals and updating the records; otherwise, generating an approval timeout reminder and pushing it to relevant parties and marking it as abnormal. This invention achieves dynamic tracking and intelligent push notifications, improving the timeliness and traceability of approvals.

Description

Technical Field

[0001] This invention belongs to the field of intelligent push, and in particular relates to an intelligent push method and system for approval processes in confined spaces. Background Technology

[0002] In the field of confined space operation safety management, the approval process is a critical control link before the operation begins. According to regulations such as the "Interim Provisions on Safety Management and Supervision of Confined Space Operations in Industrial and Commercial Enterprises," multiple levels of approval, including operation application, qualification verification, and safety measure confirmation, must be completed before operation. Currently, most enterprises still use a combination of paper-based document circulation and offline manual communication to complete the approval process. That is, the person initiating the operation fills out a paper application form, which is then submitted to the initial reviewer, secondary reviewer, and final reviewer for signature via manual delivery or email. In this complex process involving multiple departments and multiple levels of approval, the approval nodes are scattered, personnel have different authority levels, and there is a lack of unified process tracking methods. This easily leads to problems such as missed approvers, incomplete approval opinions, and no follow-up after approval time limits, resulting in delays in the operation and affecting normal production plans. At the same time, the approval records are stored in scattered paper form, making it impossible to form an effective traceability chain. In the event of an accident, it is difficult to quickly retrieve a complete chain of approval evidence.

[0003] The existing technologies mentioned above all suffer from the problems described in this background: the lack of a dynamic tracking and intelligent push mechanism for the approval process leads to low approval efficiency, opaque process status, and prominent issues of approval timeouts and missing records, which cannot meet the management requirements of timeliness and traceability for confined space operations. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention proposes an intelligent push method and system for confined space operation approval processes. The method includes: acquiring the identifier of the pending approval node and the associated set of approval parameters for the target operation application; mapping and formatting the operation application information according to the mandatory approval field requirements to generate a dedicated approval form for each node; refreshing the real-time approval task cache using the dedicated approval form when the task push time is detected; collecting the appraiser's running status feedback value and verifying it against the appraiser's permission list to obtain the permission verification result, while comparing it with the approval timeliness threshold to obtain the timeliness detection result; when the permission verification is successful and the timeout has not expired, if the approval task has been completed and the mandatory information has been submitted, generating a push priority based on the next node appraiser's permission level and historical response time, and simultaneously pushing it to their APP and PC terminals and updating the records; otherwise, generating an approval timeout reminder and pushing it to relevant parties, while marking the abnormal status. This invention achieves dynamic tracking and intelligent push of the approval process, significantly improving the timeliness, transparency, and traceability of confined space operation approvals.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for intelligently pushing approval processes for confined space operations includes:

[0007] Obtain the set of pending approval node identifiers corresponding to the target job application number. The set of pending approval node identifiers contains the unique identifiers of the initial review node, secondary review node and final review node that the target job application needs to go through in the approval process.

[0008] Obtain the set of approval parameters associated with the identifier of the node to be approved, wherein the set of approval parameters includes the list of approver permissions for each approval node, the required fields for approval, and the approval time limit threshold;

[0009] The target job application information is mapped and formatted according to the required fields in the approval parameter set to generate a unique approval form corresponding to each approval node. The unique approval form contains the approval opinion field and the required approval information to be filled in by the approval node.

[0010] When the task push time of the current approval node is detected, the real-time approval task cache maintained for the current approval node is obtained, and the approval task content in the real-time approval task cache is refreshed using the exclusive approval form of the current approval node.

[0011] Collect operational status feedback values ​​for the current approval node, including the online status of the approver, the progress of the approval task processing, and the submission status of the approval opinions;

[0012] The approver's identity information in the operation status feedback value is matched and verified with the current approver's real-time permission data to obtain the permission verification result;

[0013] The timeliness is compared with the approval task processing progress in the operation status feedback value and the approval timeliness threshold to obtain the timeliness detection result.

[0014] Specifically, the intelligent push method for approval processes also includes:

[0015] When the permission verification result indicates that the user has approval authority and the timeliness detection result indicates that the timeout has not expired, the processing progress of the approval task and the submission status of the approval comments are obtained.

[0016] If the processing progress of the approval task is completed and the submission status of the approval opinion is that all required information has been submitted, then the conditions for process advancement are met, the next approval node identifier is determined, and the approval task containing the exclusive approval form and work application materials is pushed to the approver's APP or PC terminal of the next approval node according to the approval authority level and historical approval response time. At the same time, the current node status and the timestamp of entering the next node in the approval process progress record are updated.

[0017] Otherwise, if the conditions for process advancement are not met, an approval timeout reminder containing the reason for the timeout or missing information is generated based on the timeliness detection result and the approval task processing progress. The approval timeout reminder is pushed to the approver's end and the job initiator's end of the current approval node, and the abnormal status of the current node is marked in the approval process progress record.

[0018] Specifically, the target job application information is mapped and formatted according to the mandatory approval field requirements in the approval parameter set, including:

[0019] Obtain the node type identifier of the current approval node and the job type identifier of the current operation; the node type identifier is a preset type code used to uniquely distinguish between the initial review node, the secondary review node and the final review node; the job type identifier is a preset category code used to uniquely distinguish between daily inspection operations, maintenance operations and cleaning operations.

[0020] Using the combination of the node type identifier and the job type identifier as a query index, a target form template uniquely associated with the query index is retrieved from a pre-built form template library. The form template library stores multiple form templates, each of which is pre-configured with a corresponding combination of node type identifier and job type identifier. Each form template also defines the set of required fields for that node type under that job type, as well as the field mapping relationship between each required field and the data fields in the target job application information.

[0021] Parse the target form template to extract the set of required fields and the field mapping relationship corresponding to each required field in the set of required fields;

[0022] Based on the field mapping relationship, the field data corresponding to each required field is read from the target job application information, and the read field data is written into the required field position of the target form template to generate the exclusive approval form for the current approval node; the exclusive approval form contains the set of required fields and the filled field data.

[0023] Specifically, when the task push time for the current approval node is detected, it includes:

[0024] Obtain the node identifier of the current approval node, the approver identifier of the current approver, and the approval time limit threshold of the current approval node;

[0025] Obtain historical online behavior data and real-time online status value associated with the approver's identifier. The historical online behavior data includes the approver's daily login time sequence and logout time sequence within a preset historical period. The real-time online status value is used to characterize whether the approver is currently online or offline.

[0026] Statistical analysis is performed on the historical online behavior data to extract the online probability distribution characteristics of the approver in each time period;

[0027] Based on the online probability distribution characteristics, starting from the current time and using the approval timeliness threshold as the prediction window length, predict a expected online time periods for the approver within the prediction window. Each expected online time period includes an estimated start time and an estimated duration.

[0028] Obtain the current system time and match it with the real-time online status value:

[0029] If the real-time online status value is online, then the current moment is determined to be the time when the task is pushed.

[0030] If the real-time online status value is offline, then the estimated start time of the estimated online period with the smallest difference from the current time is selected from the at least one estimated online period, and the estimated start time of the estimated online period is determined as the task push time.

[0031] When the system clock reaches the task push time, the approval task push instruction is triggered, and the exclusive approval form of the current approval node is pushed to the approver's end.

[0032] Specifically, the permission verification results include:

[0033] Send a permission verification request containing the approver's identifier to the permission center to obtain real-time permission data corresponding to the approver's identifier;

[0034] Receive real-time permission data returned by the permission center. The real-time permission data contains m permission records. Each permission record contains a permission object field, an effective timestamp, and an expiration timestamp.

[0035] Extract the effective timestamp and expiration timestamp of each permission record in the real-time permission data, and compare them one by one with the current system timestamp to filter out a set of valid permission records that satisfy the condition: effective timestamp ≤ current system timestamp ≤ expiration timestamp;

[0036] Match and verify the node identifier with the permission object field in the set of valid permission records:

[0037] If there is at least one permission record in the set of valid permission records whose value of the permission object field matches the node identifier, determine that the permission verification result is having approval permission;

[0038] Otherwise, determine that the permission verification result is not having approval permission. When the permission verification result is not having approval permission, generate a permission insufficient prompt message and push the permission insufficient prompt message to the approver side and the job initiation side of the current approval node.

[0039] Specifically, obtain the timeliness detection result, including:

[0040] Obtain the preset work calendar data. The work calendar data includes the date type of each date and the start time S_work and end time E_work of the effective working period of each working day. The date type is used to distinguish working days, rest days, and legal holidays;

[0041] Based on the work calendar data, identify all time sub-intervals that satisfy the condition that the date type is a working day and the time is within the range of [S_work, E_work] in the time interval from T_start to T_end, calculate the length of each time sub-interval and accumulate them to obtain the effective approval duration T_valid;

[0042] Compare the effective approval duration T_valid with the approval timeliness threshold T_limit:

[0043] If T_valid ≥ T_limit, determine that the timeliness detection result is timeout;

[0044] If T_valid < T_limit, determine that the timeliness detection result is not timeout.

[0045] Specifically, generate a push priority according to the approver permission level of the next approval node and the historical approval response duration, including:

[0046] Obtain the set of pending approval tasks to be pushed to the same approver side at the current moment. Each pending approval task in the set of pending approval tasks is associated with a next approval node identifier, an emergency level value of the current job task, and job application materials;

[0047] For each approval task to be pushed for approval, obtain the approver identifier corresponding to the next approval node identifier, and obtain the authority level value, historical average approval response duration, and current number of pending tasks corresponding to the approver identifier;

[0048] Set a basic priority coefficient according to the emergency level value, and the basic priority coefficient has a positive correlation with the emergency level value;

[0049] Set an authority weight according to the authority level value, and the authority weight has a positive correlation with the authority level value;

[0050] Set a response speed weight according to the historical average approval response duration, and the response speed weight has a negative correlation with the historical average approval response duration;

[0051] Set a load weight according to the current number of pending tasks, and the load weight has a negative correlation with the current number of pending tasks;

[0052] Perform a weighted sum of the basic priority coefficient, authority weight, response speed weight, and load weight according to a preset weight coefficient to obtain the comprehensive priority value of each approval task to be pushed for approval;

[0053] Push each approval task to be pushed for approval to the approver side in order from high to low according to the comprehensive priority value.

[0054] Specifically, generate an approval timeout reminder including the reason for timeout or missing information items, including:

[0055] Obtain the node identifier of the current approval node, the current approver identifier, the job initiation side identifier, and the superior management side identifier;

[0056] Obtain the task push time T_start of the current approval node, the current time T_end, and the approval time limit T_limit;

[0057] Calculate the current timeout duration T_over = T_end - (T_start + T_limit) according to T_start, T_end, and T_limit;

[0058] Calculate the timeout ratio R_over = T_over / T_limit;

[0059] Obtain the preset first-level timeout threshold R1 and second-level timeout threshold R2, where R1 < R2;

[0060] Compare R_over with R1 and R2 respectively:

[0061] If R1 ≤ R_over < R2, determine that the timeout level is a first-level timeout;

[0062] If R_over≥R2, then the timeout level is determined to be a level 2 timeout.

[0063] Specifically, generating approval timeout reminders that include reasons for timeouts or missing information also includes:

[0064] Obtain the running status feedback value of the current approval node, and extract missing information items and timeout reasons based on the approval task processing progress and approval opinion submission status in the running status feedback value;

[0065] Generate corresponding timeout reminder content based on the timeout level:

[0066] If it is a Level 1 timeout, a Level 1 timeout reminder instruction containing the timeout reason, missing information items and suggested handling methods will be generated and pushed to the current approver's terminal;

[0067] If it is a Level 2 timeout, a Level 2 timeout reminder instruction is generated, which includes the timeout reason, missing information items and suggested handling methods, and is pushed to the current approver's terminal, the superior management terminal and the job initiator terminal at the same time;

[0068] In the approval process progress record, mark the timeout level of the current node as Level 1 timeout or Level 2 timeout, and record the timeout reminder push timestamp.

[0069] A confined space operation approval process intelligent push system includes:

[0070] The process configuration module is configured to: obtain a set of pending approval node identifiers corresponding to the target job application number, wherein the set of pending approval node identifiers contains unique identifiers of the initial review node, secondary review node and final review node that the target job application must pass through in the approval process; obtain a set of approval parameters associated with the pending approval node identifiers, wherein the set of approval parameters includes a list of approver permissions for each approval node, mandatory approval field requirements and approval time threshold;

[0071] The form generation module is configured to: map and convert the target job application information according to the required fields in the approval parameter set, and generate a unique approval form corresponding to each approval node. The unique approval form contains the approval opinion field and the required approval information to be filled in by the approval node.

[0072] The push scheduling module is configured to: when the task push time of the current approval node is detected to arrive, obtain the real-time approval task cache maintained for the current approval node, and refresh the approval task content in the real-time approval task cache using the exclusive approval form of the current approval node;

[0073] The status acquisition module is configured to: collect the online status of the approver, the progress of the approval task processing, and the status of the submission of approval opinions for the current approval node.

[0074] The permission verification module is configured to match and verify the approver's identity information in the running status feedback value with the current approver's real-time permission data to obtain the permission verification result.

[0075] The timeliness detection module is configured to: compare the approval task processing progress in the running status feedback value with the approval timeliness threshold to obtain the timeliness detection result;

[0076] The process advancement decision module is configured to: when the permission verification result indicates that the user has approval authority and the timeliness detection result indicates that the timeout has not expired, obtain the processing progress of the approval task and the submission status of the approval opinion;

[0077] If the processing progress of the approval task is completed and the submission status of the approval opinion is that all required information has been submitted, then the conditions for process advancement are met, the next approval node identifier is determined, and the approval task containing the exclusive approval form and work application materials is pushed to the approver's APP or PC terminal of the next approval node according to the approval authority level and historical approval response time. At the same time, the current node status and the timestamp of entering the next node in the approval process progress record are updated.

[0078] Otherwise, if the conditions for process advancement are not met, an approval timeout reminder containing the reason for the timeout or missing information is generated based on the timeliness detection result and the approval task processing progress. The approval timeout reminder is pushed to the approver's end and the job initiator's end of the current approval node, and the abnormal status of the current node is marked in the approval process progress record.

[0079] Compared with the prior art, the beneficial effects of the present invention are:

[0080] This invention achieves refined pre-configuration of the confined space operation approval process by constructing an association mapping between the identifier of the node to be approved and the set of approval parameters, and generating a dedicated approval form by combining field mapping. It automatically detects the compliance and timeliness of approval node permissions by collecting real-time feedback values ​​of the approver's running status and dynamically matching and verifying them with the permission list and timeliness thresholds. When the progress of the approval task and the submission of opinions meet the conditions for process advancement, a push priority is generated based on the permission level and historical response time of the approver at the next node, and pushed synchronously to the APP and PC, effectively shortening the task flow gap. When permission mismatch, timeout, or missing information is detected, an approval timeout reminder containing specific reasons is immediately generated and pushed to relevant parties, and the abnormal status is simultaneously marked, achieving rapid response to approval anomalies and full-process traceability. Therefore, this invention significantly improves the flow efficiency, status transparency, and timeliness control capabilities of confined space operation approval, meeting the management requirements for approval timeliness and traceability in high-risk operation scenarios. Attached Figure Description

[0081] Figure 1 This is a flowchart of an intelligent push method for confined space operation approval process according to the present invention;

[0082] Figure 2 This is a module diagram of an intelligent push system for confined space operation approval process according to the present invention. Detailed Implementation

[0083] Example 1

[0084] Please see Figure 1 The present invention provides an embodiment of a method for intelligently pushing approval processes for confined space operations, comprising the following steps:

[0085] S1. Obtain the set of pending approval node identifiers corresponding to the target job application number. The set of pending approval node identifiers includes the unique identifiers of the initial review node, secondary review node and final review node that the target job application needs to go through in the approval process.

[0086] S2. Obtain the set of approval parameters associated with the identifier of the node to be approved. The set of approval parameters includes the list of approver permissions for each approval node, the required fields for approval, and the approval time limit threshold.

[0087] For example, taking the dredging operation of storage tank No. 3 as an example, the operation application was initiated by the on-site safety officer Li Si in the system at 14:00 on March 13, 2024, with the operation application number APP-20240313-005. The following describes in detail the specific implementation process of steps S1 and S2 in this scenario.

[0088] {

[0089] "applicationId":"APP-20240313-005",

[0090] "approvalNodes":["01","02","03"]

[0091] }

[0092] Based on node identifiers 01, 02, and 03 in the aforementioned set of node identifiers awaiting approval, the system queries the approval parameter configuration database for the approval parameters associated with each node. Each node's approval parameters include three core components: a list of approver permissions, mandatory approval field requirements, and an approval time limit threshold.

[0093] For the initial review node (identifier 01):

[0094] Approver Permission List: The default approver for this node is workshop director Zhao Gong, whose approver identifier is ZG001; at the same time, considering the temporary authorization scenario, the list also includes backup approver Li Gong (approver identifier LG001). The permission list is stored in array form: ["ZG001","LG001"].

[0095] Required fields for approval: During the initial review phase of the cleanup operation, personnel qualifications and on-site preparedness must be verified. Required fields include: gas detection results, guardian confirmation, and isolation measure confirmation. This requirement is stored as a JSON object, and the specific field definitions will be used for form generation in step S3.

[0096] Approval time limit: The approval time limit for the initial review node is 30 minutes. That is, after a task enters the initial review node, the approval must be completed within 30 minutes, otherwise a timeout reminder will be triggered.

[0097] For the review node (identifier 02):

[0098] Approver Permission List: The default approver for this node is Security Director Wang, with approver ID WG001; Additionally, due to the possibility of Wang traveling for work, a proxy approver, Liu (approver ID LG002), is configured. The permission list is: ["WG001","LG002"].

[0099] Required fields for approval: The review stage requires checking the equipment binding status and gas detection verification. Required fields include: confirmation of equipment binding status, verification of gas detection, and approval comments.

[0100] Approval time limit: The approval time limit for the review node is 60 minutes.

[0101] For the final review node (identifier 03):

[0102] Approver Permission List: The default approver for this node is Vice President Sun, whose ID is SG001; there is no backup approver, and the permission list is: ["SG001"].

[0103] Required fields for approval: The final review stage requires confirmation of overall risk measures. Required fields include: overall risk confirmation and final approval opinion.

[0104] Approval time limit: The approval time limit for the final review node is 120 minutes.

[0105] The system integrates the approval parameter sets from the three nodes into a single, complete data structure, which is stored in the task context data packet for subsequent steps to access. The integrated approval parameter set is as follows:

[0106] {

[0107] "applicationId":"APP-20240313-005",

[0108] "nodes":[

[0109] {

[0110] "nodeId":"01",

[0111] "approvers":["ZG001","LG001"],

[0112] "requiredFields":["Gas detection results","Guardian confirmation","Isolation measures confirmation"],

[0113] "timeoutThreshold":30

[0114] },

[0115] {

[0116] "nodeId":"02",

[0117] "approvers":["WG001","LG002"],

[0118] "requiredFields":["Device binding status confirmation","Gas detection verification","Approval comments"],

[0119] "timeoutThreshold":60

[0120] },

[0121] {

[0122] "nodeId":"03",

[0123] "approvers":["SG001"],

[0124] "requiredFields":["Overall Risk Confirmation","Final Approval Opinion"],

[0125] "timeoutThreshold":120

[0126] } ]

[0128] }

[0129] At this point, the system has completed the acquisition and configuration loading of basic information for the target job application, providing data support for subsequent steps such as generating dedicated approval forms, dynamic push notifications, and permission verification. For example, when the process flows to the review node, the system will extract the approver permission list, required fields, and time limit thresholds for node 02 from this parameter set, which will be used to generate a dedicated approval form for the review node and control the push logic.

[0130] The system first obtains the target job application number APP-20240313-005, and then queries the corresponding approval process template from the approval process definition database based on this number. The query reveals that the approval process for this job application is preset to a three-level approval process, requiring sequential processing through an initial review node, a secondary review node, and a final review node. The system extracts the unique identifier for each node from the process template, generating a set of node identifiers to be approved.

[0131] Specifically, the system returns the following set of node identifiers awaiting approval:

[0132] Preliminary review node identifier: 01 (representing the preliminary review node)

[0133] Review node identifier: 02 (representing the review node)

[0134] Final review node identifier: 03 (representing the final review node)

[0135] This collection is stored as a JSON array:

[0136] S3. Map and convert the target job application information according to the required fields in the approval parameter set to generate a unique approval form corresponding to each approval node. The unique approval form contains the approval opinion field and the required approval information to be filled in by the approval node.

[0137] In this embodiment, after generating the dedicated approval form for the current approval node in S3, the system stores this form along with the basic task data (such as job application materials, node identifier, approver list, etc.) in the real-time approval task cache maintained for that node, as the initial task content. When the task push time is detected, the system first retrieves the cache and replaces the original form content in the cache with the dedicated approval form regenerated at the current time (ensuring that the field mapping and data are up-to-date), to ensure that the form pushed to the approver is in real-time status.

[0138] S4. When the task push time of the current approval node is detected to arrive, the real-time approval task cache maintained for the current approval node is obtained, and the approval task content in the real-time approval task cache is refreshed using the exclusive approval form of the current approval node. In this embodiment, after the task flow is transferred to the current approval node, the system immediately starts a timed monitoring thread to continuously compare the current system time with the predicted task push time T_push. If the current time has not yet reached T_push, the system remains in a waiting state and does not perform any push operation. At the same time, the real-time online status of the approver can continue to be collected.

[0139] S5. Collect the running status feedback values ​​of the approver's online status, approval task processing progress, and approval opinion submission status of the current approval node;

[0140] S6. Match and verify the approver's identity information in the running status feedback value with the current approver's real-time permission data to obtain the permission verification result;

[0141] S7. Compare the approval task processing progress in the operation status feedback value with the approval timeliness threshold to obtain the timeliness detection result;

[0142] S8. When the permission verification result indicates that the user has approval permission and the timeliness detection result indicates that the timeout has not expired, obtain the processing progress of the approval task and the submission status of the approval opinion.

[0143] S9. If the processing progress of the approval task is completed and the submission status of the approval opinion is that all required information has been submitted, then it is determined that the process advancement conditions are met, the next approval node identifier is determined, and the approval task containing the exclusive approval form and work application materials is pushed to the approver's APP or PC terminal of the next approval node according to the approval authority level and historical approval response time of the next approval node. At the same time, the current node status and the timestamp of entering the next node in the approval process progress record are updated.

[0144] S10. Otherwise, if the process progress conditions are not met, an approval timeout reminder containing the timeout reason or missing information items is generated based on the timeliness detection result and the approval task processing progress. The approval timeout reminder is pushed to the approver's end and the job initiator end of the current approval node, and the abnormal status of the current node is marked in the approval process progress record.

[0145] For example, taking the dredging operation of Tank No. 3 as an example, after receiving the application initiated by safety officer Li Si, the system automatically obtains the identifiers of the initial review, secondary review, and final review nodes and the associated approval parameters. When the process flows to the secondary review node, the system dynamically generates a unique approval form containing mandatory fields such as equipment binding status and gas detection verification based on the node type and operation type. It also predicts that the approver, Engineer Wang, is most likely to be online at 15:00 based on his historical online activity, and pushes the form to his APP and PC at that time. After Engineer Wang opens the form online, the system collects his online status, processing progress, and opinion submission status in real time, and verifies that he has the secondary review authority through the permission center. At the same time, it calculates the valid approval based on the work calendar. The timeout period has not expired. After Mr. Wang submits all the required information, the system determines that the conditions for the process to proceed are met. Then, based on the approval level of Mr. Sun, the final review node approver, the historical response time, and the current pending workload, combined with the task urgency level (the operation starts at 16:00), the system calculates the priority and pushes the final review form to Mr. Sun's end first. If the subsequent final review node expires, the system will issue a tiered reminder based on the timeout ratio. The second-level timeout will be pushed to the approver, the superior, and the operation initiator simultaneously to ensure timely intervention for urgent tasks. This achieves intelligent push throughout the entire process, from form generation and intelligent push to dynamic permission verification, precise timeliness control, and tiered timeout reminders, significantly improving the timeliness, transparency, and traceability of confined space operation approvals.

[0146] In this embodiment, during the confined space operation approval process, different approval nodes (such as preliminary review, secondary review, and final review) have different focuses on the operation application information. The preliminary review requires verification of personnel qualifications, the secondary review requires checking the equipment binding status, and the final review requires confirming the overall risk measures. If a fixed field mapping is used to generate the approval form, it cannot adapt to the personalized needs of each node, resulting in redundant information or missing key approval items in the approval form, affecting the accuracy and efficiency of the approval. In addition, the approval fields required for different operation types (such as daily inspection and maintenance operations) also differ, and static forms are difficult to adapt flexibly, further reducing the pertinence and effectiveness of the approval process. Therefore, this embodiment maps and converts the target operation application information according to the mandatory approval field requirements in the approval parameter set, including:

[0147] S301. Obtain the node type identifier of the current approval node and the job type identifier of the current job. The node type identifier is a preset type code used to uniquely distinguish between the initial review node, the secondary review node, and the final review node. For example, it uses a two-digit code: 01 represents the initial review node, 02 represents the secondary review node, and 03 represents the final review node. The job type identifier is a preset category code used to uniquely distinguish between daily inspection jobs, maintenance jobs, and cleaning jobs. For example, it uses a two-digit code: 01 represents daily inspection jobs, 02 represents maintenance jobs, and 03 represents cleaning jobs. The above identifiers are written into the task context data packet when the job application is initiated, and the system can obtain them by parsing the task context.

[0148] S302. Using the combination of the node type identifier and the job type identifier as a query index, for example, concatenating the identifier 01 of the initial review node and the identifier 03 of the cleanup job into a combination key 01-03, the target form template uniquely associated with the query index is retrieved from the pre-built form template library; the form template library is stored in a relational database table, which contains at least the following fields: template ID (primary key), node type identifier, job type identifier, template content (stored in JSON format), creation time, and version number; by setting a joint unique index of the node type identifier and the job type identifier, it is ensured that each (node ​​type, job type) tuple corresponds to only one valid template in the library, avoiding retrieval conflicts; the template content of each form template defines two core data parts: the first part is a set of required fields, which stores the list of field names that must be filled in when approving the node type under the job type in the form of a JSON array, for example, the required fields for the initial review node under the cleanup job. The first part can be represented as ["Gas Detection Result", "Guardian Confirmation", "Isolation Measures Confirmation"]. The second part is the field mapping relationship, which stores the mapping rules between each required field and the data fields in the target job application information in the form of a JSON object. Each mapping rule contains a list of data source field names and a format template string. For example, the mapping rule for the gas detection result field can be defined as: {"sourceFields":["Oxygen Content Value", "Hydrogen Sulfide Content Value"], "formatTemplate":"Oxygen{0} / Hydrogen Sulfide{1}"}, which means that the value of this field needs to be read from the oxygen content value and hydrogen sulfide content value fields in the application information and concatenated according to the format oxygen{0} / hydrogen sulfide{1}. The mapping rule for the guardian confirmation field can be defined as: {"sourceFields":["Guardian Name"], "formatTemplate":"{0}"}, which means that the value of the guardian name field is directly taken. Through the above structure, the template library can provide customized approval form configurations for combinations of different node types and job types, ensuring that the exclusive approval form presented by each approval node only contains the necessary review items for the current scenario, and that the data source is accurate and traceable.

[0149] S303. Parse the target form template, extract the set of required fields and the field mapping relationships corresponding to each required field in the set of required fields; in specific implementation, the system first reads the complete template file content from the database or file system according to the storage path or template ID of the determined target form template; then calls the JSON parser to deserialize the template content into a JSON object model in memory. During the parsing process, the system reads data according to the predefined template structure specification: first, it locates the required field list node through the fixed node path requiredFields, reads the list of field names stored under this node, and converts it into a memory object in the form of a string array; then it traverses the string array, and for each required field name, locates the mapping relationship definition node through the fixed node path fieldMappings, and searches for the corresponding mapping rule object under this node with the current field name as the key; the mapping rule object contains at least a list of data source field names (sourceFields) and a format template string (formatTemplate). After parsing, the system constructs a field-mapping rule lookup table in memory. This lookup table uses a HashMap data structure, with the key being the name of the required field and the value being a mapping rule object (containing an array of strings named `sourceFields` and a string named `formatTemplate`). This lookup table serves as the sole basis for subsequent data population steps, ensuring that each required field can accurately locate its data source and processing method. If, during parsing, any required field is found to lack a corresponding mapping rule in the mapping rule node, or if the mapping rule is missing the necessary `sourceFields` or `formatTemplate` attributes, the system will log an error and throw a template format exception, indicating that the template configuration is incomplete. For example, assuming the matched target form template is in JSON format, its content is as follows:

[0150] {

[0151] "templateId":"TMPL-PRI-CLN-001",

[0152] "requiredFields":["Gas detection results", "Guardian confirmation", "Isolation measures confirmation"],

[0153] "fieldMappings":{

[0154] Gas detection results:{

[0155] "sourceFields":["Oxygen content value", "Hydrogen sulfide content value"],

[0156] "formatTemplate":"Oxygen{0} / Hydrogen Sulfide{1}"

[0157] },

[0158] "Guardian Confirmation":{

[0159] "sourceFields":["Guardian's Name"],

[0160] "formatTemplate":"{0}"

[0161] },

[0162] "Isolation measures confirmed":{

[0163] "sourceFields":["Isolation status"],

[0164] "formatTemplate":"{0}"

[0165] }

[0166] }

[0167] }

[0168] After the system calls the JSON parser to read the template, it first extracts the string array ["Gas Detection Result", "Guardian Confirmation", "Isolation Measures Confirmation"] from the requiredFields node. Then, it iterates through the array. For the gas detection result field, it finds the corresponding mapping rule object under the fieldMappings node with the gas detection result as the key: {"sourceFields":["Oxygen Content Value", "Hydrogen Sulfide Content Value"], "formatTemplate":"Oxygen {0} / Hydrogen Sulfide {1}"}; for the guardian confirmation field, it finds {"sourceFields":["Guardian Name"], "formatTemplate":"{0}"}; for the isolation measures confirmation field, it finds {"sourceFields":["Isolation Status"], "formatTemplate":"{0}"}. After parsing, the system constructs the following field-mapping rule lookup table:

[0169] Key: Gas detection result; Value: sourceFields=["Oxygen content", "Hydrogen sulfide content"], formatTemplate="Oxygen{0} / Hydrogen sulfide{1}"

[0170] Key: Guardian Confirmation, Value: sourceFields=["Guardian Name"], formatTemplate="{0}"

[0171] Key: Isolation measures confirmed; Value: sourceFields=["Isolation status"], formatTemplate="{0}"

[0172] The lookup table is then passed to step S304, which reads data from the job application information and generates a specific approval form.

[0173] S304. Based on the field mapping relationship, read the field data corresponding to each required field from the target job application information and generate a dedicated approval form for the current approval node. Specifically, the system first obtains a complete application information dataset from the job application database or the job application data object in memory. This dataset stores the original values ​​of all entered fields in key-value pairs. Then, it iterates through each required field in the generated field-mapping rule lookup table and performs the following operations for the current field: According to the list of data source field names (sourceFields) in the mapping rule, read the original values ​​of the corresponding fields sequentially from the application information dataset. If the original value does not exist or is empty, it is processed according to the default value strategy preset in the mapping rule (such as filling in "no value" or throwing a missing field exception). After reading, the original value is formatted according to the format template (formatTemplate) in the mapping rule: The format template uses placeholders (such as {0}, {1}, etc.). The system replaces the placeholders with the read original values ​​in the order of the data source field name list to generate the formatted value. Field data; for directly mapped fields that do not require concatenation (the formatted template is only {0}), the first original value is directly used as the field data; after the field data is processed, the field data of the current required field is written to the corresponding position in the target form template: if the target form template is a JSON structure, the system constructs a JSON object in memory, using the field name as the key and the formatted field data as the value, and stores it in the JSON object; if the target form template is an HTML structure, the system replaces the pre-set field placeholders in the template (such as {{gas detection results}}) with the formatted field data through string replacement or DOM operations; after all required fields are filled, the system merges the complete filled form data with the static layout of the template to generate the exclusive approval form for the current approval node; when this exclusive approval form is returned to the front end in the form of an HTML page, it includes the tags and values ​​of all required fields, the approval opinion input box, and the pass / reject operation buttons; if interaction with other systems is required, it can also be returned in the form of a JSON data structure, including field names, field values, form metadata, and other information.

[0174] For example, following the template and lookup table in example S303, assume that the application information for the dredging operation of tank No. 3 stores the following key-value pairs: Oxygen content = 20.9%, Hydrogen sulfide content = 0ppm, Guardian's name = Zhang San, Isolation tag status = Locked. System traversal field-mapping rule lookup table:

[0175] For the gas detection results, read the oxygen content value = 20.9% and the hydrogen sulfide content value = 0ppm from the application information, and replace the placeholders according to the format template oxygen{0} / hydrogen sulfide{1} to get oxygen 20.9% / hydrogen sulfide 0ppm;

[0176] For guardian confirmation, read the guardian's name = Zhang San, and directly output Zhang San using the format template {0}.

[0177] For confirmation of isolation measures, read the isolation tag status = "Locked Tag", and directly output "Locked Tag".

[0178] The system constructs a populated data object in memory: {"Gas Detection Result":"Oxygen 20.9% / Hydrogen Sulfide 0ppm", "Guardian Confirmation":"Zhang San", "Isolation Measures Confirmation":"Locked and Tag-Off"}. Subsequently, the system combines this data object with the template's static HTML layout to generate a custom approval form HTML fragment; this custom approval form HTML fragment is ultimately pushed to the task list front-end page of the initial reviewer for display. For example, the custom approval form HTML fragment is:

[0179] {

[0180] Gas detection results: Oxygen 20.9% / Hydrogen sulfide 0 ppm

[0181] Guardian's confirmation: Zhang San

[0182] Quarantine measures confirmed: Locked up and labeled.

[0183] <textarea placeholder="审批意见">< / textarea>

[0184] <button> pass< / button>

[0185] <button> turn down< / button>

[0186] }

[0187] This embodiment introduces a two-dimensional combined retrieval mechanism of node type identifier and job type identifier, combined with a pre-built form template library, to achieve dynamic generation and precise adaptation of approval forms. Addressing the different concerns of the initial review, secondary review, and final review nodes regarding personnel qualifications, equipment status, and risk measures, as well as the differentiated approval needs of different scenarios such as daily inspections, maintenance operations, and cleaning operations, the system can automatically match the corresponding set of required fields and field mapping rules to generate a dedicated approval form containing only the necessary review items for the current node and current job type, effectively avoiding the information redundancy or missing key items problems caused by traditional static forms. Simultaneously, through... The field mapping relationship is automatically read from the job application information and formatted to fill in the data, reducing the workload of manual data entry for approvers and improving approval efficiency and data accuracy. The template library uses a joint unique index to ensure that each (node ​​type, job type) tuple corresponds to a unique template, and clearly defines the required fields and mapping rules through JSON structure, ensuring the flexibility and maintainability of form generation. Therefore, this embodiment significantly improves the pertinence and effectiveness of the approval process for jobs in confined spaces, meets the personalized approval needs under different nodes and different job types, and further ensures the integrity and accuracy of approval information in high-risk job scenarios.

[0188] This embodiment, while addressing the issue of dynamic form generation, still faces adaptability challenges in the approval push process: If pushes are only sent at fixed times based on node entry, without considering the approver's real-time online status and individual processing habits, pushes may fail due to the approver being offline or busy, leading to task backlog. Simultaneously, a uniform timeliness threshold is difficult to match the response patterns of different approvers, failing to trigger reminders at the optimal time, resulting in frequent process delays and idle timeouts. Therefore, a dynamic push mechanism based on the approver's real-time status and historical behavioral characteristics is needed to further improve the response speed and timeliness controllability of the approval workflow; to this end, this embodiment, when detecting the arrival of the task push time for the current approval node, includes:

[0189] S401. Obtain the node identifier of the current approval node, the approver identifier of the current approver, and the approval time threshold of the current approval node; wherein, the approval time threshold in this embodiment is a fixed duration parameter preset according to business management requirements for each approval node (such as preliminary review, secondary review, and final review) during the process configuration stage, such as 30 minutes for preliminary review and 60 minutes for secondary review, which is stored in the approval parameter set together with the approver permission list and the required approval fields, and is obtained by associating with the node identifier to be approved.

[0190] S402. Obtain historical online behavior data and real-time online status value associated with the approver's identifier. The historical online behavior data includes the approver's daily login time sequence and logout time sequence within a preset historical period. The real-time online status value is used to characterize whether the approver is currently online or offline.

[0191] S403. Perform statistical analysis on the historical online behavior data to extract the online probability distribution characteristics of the approver in each time period;

[0192] S404. Based on the online probability distribution characteristics, starting from the current time and using the approval timeliness threshold as the prediction window length, predict a expected online time periods for the approver within the prediction window, where each expected online time period includes an estimated start time and an estimated duration.

[0193] S405. Obtain the current system time and match it with the real-time online status value:

[0194] S406. If the real-time online status value is online, then the current time is determined to be the task push time;

[0195] S407. If the real-time online status value is offline, then select the estimated online time period from the at least one estimated online time period whose estimated start time value is greater than the current time and whose difference from the current time value is the smallest, and determine the estimated start time value of the estimated online time period as the task push time.

[0196] S408. When the system clock reaches the task push time, the approval task push instruction is triggered, and the exclusive approval form of the current approval node is pushed to the approver's end.

[0197] For example, suppose the approval process for the dredging operation of storage tank No. 3 has passed the initial review node and is now moving to the secondary review node. The current approval node is the secondary review node, with node identifier 02. The current approver is Safety Director Wang (approver identifier WG001). The approval time limit T_limit for the secondary review node is 60 minutes, meaning that the approval task must be completed within 60 minutes of entering the secondary review node; otherwise, a timeout reminder will be triggered.

[0198] The system first retrieves the following information from the preceding data of the current approval task:

[0199] The node identifier for the current approval node is 02 (representing the review node).

[0200] The current approver's identifier: WG001 (Engineer Wang's unique employee ID);

[0201] The current approval time threshold is 60 minutes (this threshold is pre-configured in the approval parameter set).

[0202] The system reads historical online behavior data associated with the approver identifier WG001 from the user behavior log database. This historical online behavior data includes the approver's login and logout time sequences for each day within a preset historical period (in this embodiment, the past four natural weeks). The login time refers to the specific point in time when the approver successfully logs into the system via the APP or PC, and the logout time refers to the point in time when the approver actively logs out or is passively offline due to session timeout.

[0203] For example, the system reads the following partial historical records of Mr. Wang over the past 4 weeks (stored in JSON format):

[0204] {

[0205] "userId":"WG001",

[0206] "historyData":[

[0207] {"date":"2024-02-12","logins":["08:55:23","13:20:45"],"logouts":["11:58:12","17:45:30"]},

[0208] {"date":"2024-02-13","logins":["09:02:17","13:15:22"],"logouts":["12:02:45","18:02:10"]},

[0209] ... / / Data for approximately 4 weeks and 20 working days ]

[0211] }

[0212] Meanwhile, the system obtains Wang Gong's real-time online status value through the real-time heartbeat interface or session management service. Assuming the current system time T_current is 14:30:00 on March 13, 2024, the system detects that Wang Gong's real-time online status value is offline (e.g., his APP is not connected, and there are no active sessions on the PC).

[0213] The system divides a 24-hour day into 96 time slices, each lasting 15 minutes (e.g., 00:00-00:15 is the first time slice, 00:15-00:30 is the second time slice, and so on). For each time slice, the system calculates the percentage of days the approver has been online in the past four weeks, which is used as the online probability for that time slice.

[0214] The specific statistical method is as follows: For each historical date, iterate through the login / logout records for that day, marking the time period between the login and logout times as online, and covering the corresponding time slice. For example, on February 12, 2024, if Mr. Wang logs in at 08:55:23 and logs out at 11:58:12, then the time slices 08:45-09:00 (because the login time of 08:55 falls within this time slice), 09:00-09:15, and so on up to 11:45-12:00 are all marked as online. The system accumulates all historical dates to obtain the number of online days for each time slice, divides it by the total number of valid working days (excluding holidays), and obtains the online probability for that time slice.

[0215] For example, the statistical results are as follows (only some relevant time slices are listed):

[0216] Time slot 14:15-14:30: Over the past 4 weeks (20 working days), Engineer Wang was online for 1 day during this time slot, with an online probability of 1 / 20 = 5%.

[0217] Time slot 14:30-14:45: Online status days: 0 days, online probability = 0%;

[0218] Time slot 14:45-15:00: Online status for 2 days, online probability = 10%;

[0219] Time slot 15:00-15:15: Online status for 18 days, online probability = 90%;

[0220] Time slot 15:15-15:30: Online status for 19 days, online probability = 95%;

[0221] Time slot 15:30-15:45: Online status for 20 days, online probability = 100%.

[0222] The above statistical results constitute the online probability distribution characteristics of Engineer Wang. The system stores them as a floating-point array of length 96, with index i corresponding to the online probability value of the i-th time slice.

[0223] The system uses the current time T_current (14:30:00) as the starting point and the approval time limit threshold of 60 minutes as the prediction window length, determining the time range of the prediction window to be from 14:30:00 to 15:30:00. The prediction window is mapped to the time slice index range: 14:30 belongs to time slice 14:30-14:45 (index 58), and 15:30 belongs to time slice 15:30-15:45 (index 62). Therefore, the window covers five time slices from index 58 to 62.

[0224] The system uses a sliding window algorithm to identify expected online periods: an online probability threshold P_threshold is set at 70% (this threshold can be adjusted according to the actual business scenario). When the online probability of several consecutive time slices exceeds this threshold, these consecutive time slices are merged into one expected online period. In this embodiment, the online probability threshold P_threshold is a fixed percentage value pre-set according to the actual business scenario (such as 70% in this embodiment), used to determine whether the online probability of consecutive time slices meets the conditions for constituting an expected online period. This threshold can be dynamically adjusted according to different approvers, job requirements, or the urgency of the task.

[0225] The specific calculation process is as follows:

[0226] Time slot 58 (14:30-14:45): Online probability 0% < 70%, not satisfied;

[0227] Time slot 59 (14:45-15:00): Online probability 10% < 70%, not satisfied;

[0228] Time slot 60 (15:00-15:15): Online probability 90% ≥ 70%, meets the condition, and is marked as the start of the online time slot;

[0229] Time slot 61 (15:15-15:30): Online probability 95% ≥ 70%, meets the condition, online time slot continues;

[0230] Time slice 62 (15:30-15:45): Online probability 100% ≥ 70%, which meets the condition, but this time slice has exceeded the prediction window (15:30 is the end of the window, and the start time of time slice 62 is 15:30. On the window boundary, the time slice with the start time inside the window is taken).

[0231] Therefore, the system predicts that there is an expected online period within the prediction window, with an estimated start time of 15:00:00 (the start time of time slice 60) and an estimated duration of 30 minutes (15:00-15:30) across two time slices. The system records this expected online period as: {start:15:00:00, duration:30 minutes}.

[0232] The system obtains the current time T_current=14:30:00 and matches it with the real-time online status value offline for judgment:

[0233] If the real-time online status value is online, then the current time T_current is directly determined as the task push time, and the task is pushed immediately;

[0234] Since the real-time online status is offline, the system selects the expected online time period from the predicted online time periods obtained in step four, where the estimated starting time value is greater than T_current and the difference between T_current and the estimated starting time value is the smallest.

[0235] Currently, there is only one estimated online time period, with the start time 15:00:00 being later than 14:30:00, a difference of ΔT = 30 minutes. Therefore, the system sets the task push time T_push to 15:00:00.

[0236] The system initiates a delayed scheduled task, setting the trigger time to T_push=15:00:00. When the system clock reaches 15:00:00 on March 13, 2024, the scheduled task is triggered, executing the approval task push instruction. The system retrieves the dedicated approval form corresponding to the review node from the real-time approval task cache (this form has been dynamically generated according to the method in the aforementioned embodiments, including mandatory fields such as device binding status, gas detection verification, and personal protective equipment confirmation, and has been filled with the corresponding data for this operation), and pushes this form along with the operation application materials to Wang Gong's approver's APP and PC.

[0237] Mr. Wang habitually logs into the system around 3:00 PM to process approval tasks. At this time, the tasks are delivered, allowing him to see the pending approval items immediately, avoiding message delays and response latency caused by invalid push notifications when he is offline at 2:30 PM. If Mr. Wang logs online earlier between 2:30 PM and 3:00 PM, the system can also dynamically adjust the push notification time through a real-time status polling mechanism (such as checking online status every 5 minutes), triggering a push notification immediately upon detecting his online status, further improving response speed.

[0238] Through the aforementioned dynamic push time detection mechanism, this embodiment achieves the following technical effects: based on the intelligent fusion of the approver's historical online patterns and real-time status, the approval task is pushed to the time window when the approver is most likely to process it in a timely manner. This not only avoids invalid pushes in offline states, but also ensures that the task is processed within the approval time threshold, significantly improving the timeliness and success rate of the approval process for confined space operations.

[0239] For example, in this embodiment S5, it is assumed that the dredging operation for tank No. 3 was pushed to the APP and PC of the review approver, Mr. Wang (apprant identifier WG001), at 15:00 on March 13, 2024. Mr. Wang opened the APP at 15:05 to view the tasks pending approval. At this time, the system started the running status acquisition module to monitor the entire process of Mr. Wang handling the task in real time. The specific acquisition process is as follows:

[0240] The system continuously monitors Wang's online status through a heartbeat mechanism on the app (reporting every 30 seconds) and a long-lived WebSocket connection on the PC. When Wang opens the app at 15:05, the client reports an online heartbeat to the server, and the system records his online status as online. If Wang switches the app to the background or locks the screen during the approval process, the system will update his status to offline after two consecutive heartbeat losses (i.e., 60 seconds of no response). In this scenario, Wang operated online continuously from 15:05 to 15:08, therefore the collected online status remained consistently online.

[0241] The system maintains a processing progress state machine for each approval task, including three states: Not Started, Processing, and Completed. When Mr. Wang first opened the dedicated approval form for the review node at 15:05:23, the system detected that the task had never been accessed before, automatically updating the task processing progress from Not Started to Processing and recording the start timestamp 15:05:23. The system uses front-end tracking technology to monitor Mr. Wang's actions on the form in real time, including clicking input boxes, checking options, and scrolling. When Mr. Wang filled in all required fields and clicked the submit button at 15:07:45, the front-end triggered a task completion event, and the system updated the task processing progress to Completed and recorded the completion timestamp 15:07:45. Therefore, the sequence of task processing progress changes collected by the system is: Not Started → Processing → Completed.

[0242] The dedicated approval form for the review node (dynamically generated based on node type 02 and job type 03) includes the following required fields:

[0243] Approval comments (text input box, required)

[0244] Device binding status confirmation (radio button, options are bound / not bound, required field)

[0245] Gas detection verification (checkbox; you must check "verified" to confirm; required)

[0246] The system uses a form field change monitoring mechanism to maintain a submission status (initially unsubmitted) for each required field. When Engineer Wang performs an operation on the form:

[0247] At 15:05:30, Engineer Wang typed "agree" in the approval comments input box. The equipment was bound, the gas detection was qualified, and the system detected that the content of this field was not empty, so it updated the submission status to "submitted".

[0248] At 15:06:10, Engineer Wang checked the device binding status and confirmed it as bound. The system then updated the status of this field to "submitted".

[0249] At 15:07:20, Engineer Wang checked the gas detection verification checkbox, and the system updated the status of that field to "submitted".

[0250] The system monitors the submission status of the three required fields in real time. When all fields are marked as submitted (i.e., 15:07:20), the system determines that all required information has been submitted for the approval opinion. When Engineer Wang clicks the submit button at 15:07:45, the system reconfirms the submission status to ensure nothing is missing.

[0251] The system integrates the above three collected results into a data structure of operational status feedback values, which serve as input for subsequent process advancement decisions. In this scenario, the operational status feedback values ​​generated by the system at 15:07:45 are as follows:

[0252] {

[0253] "approvalId":"APP-20240313-005",

[0254] "nodeId":"02",

[0255] "approverId":"WG001",

[0256] "onlineStatus":"Online",

[0257] "taskProgress": "Completed",

[0258] "submissionStatus":"All required fields have been submitted",

[0259] "progressStartTime":"2024-03-1315:05:23",

[0260] "progressCompleteTime":"2024-03-1315:07:45",

[0261] "timestamp":"2024-03-1315:07:45"

[0262] }

[0263] This feedback value will be passed to permission verification and timeliness comparison to determine whether the conditions for process advancement are met. If Mr. Wang exits midway through the approval process (e.g., closes the APP at 15:06 without completing the form), the task processing progress may be "processing" and the approval opinion submission status may be "partially submitted." The system will then trigger a logical branch indicating that the conditions for process advancement are not met and generate a corresponding missing information reminder.

[0264] This embodiment, while addressing the challenges of dynamic form generation and intelligent prediction of push timing, still faces dynamic adaptation challenges in the authorization verification stage of the approval process: Approvers' permissions may frequently change due to temporary authorizations, job rotations, or multiple personnel sharing a role. Relying solely on a static authorization list for matching and verification can lead to two problems: firstly, temporarily authorized approvers may fail static verification, causing process blockages; secondly, if personnel who have left their positions are not promptly removed from the authorization list, the system may mistakenly classify them as having authorization, creating security vulnerabilities. Furthermore, proxy authorizations typically come with expiration dates, which static authorization lists cannot reflect, further increasing the risk of inaccurate authorization verification. Therefore, an authorization verification mechanism supporting real-time synchronization of authorization data and expiration date verification is needed to ensure a strict match between the approver's actual permissions at the current moment and the process progress conditions. To this end, this embodiment obtains the following authorization verification results:

[0265] S601. Send an authorization verification request containing the approver identifier to the authorization center to obtain real-time authorization data corresponding to the approver identifier;

[0266] S602. Receive the real-time permission data returned by the permission center. The real-time permission data contains m permission records. Each permission record contains a permission object field, an effective timestamp, and an expiration timestamp.

[0267] S603. Extract the effective timestamp and expiration timestamp of each permission record in the real-time permission data, compare them one by one with the current system timestamp, and filter out the set of valid permission records that satisfy the condition that effective timestamp ≤ current system timestamp ≤ expiration timestamp.

[0268] S604. Match and verify the node identifier with the permission object field in the set of valid permission records:

[0269] S605. If there is at least one permission record in the set of valid permission records, and the value of its permission object field matches the node identifier, then the permission verification result is determined to be that the user has approval permission.

[0270] S606. Otherwise, if the permission verification result is determined to be that the approval permission is not available, when the permission verification result is that the approval permission is not available, a permission insufficient prompt message is generated and the permission insufficient prompt message is pushed to the approver end and the job initiator end of the current approval node.

[0271] For example, suppose the approval process for the dredging operation of Tank No. 3 has passed the initial review stage and entered the secondary review stage at 14:30 on March 13, 2024. Based on the dynamic push time detection mechanism, the system determines the task push time as 15:00 and pushes the dedicated approval form for the secondary review stage to the approver, Engineer Wang (approver identifier WG001), on both the APP and PC. Engineer Wang opens the APP at 15:05 to view the tasks awaiting approval. At this time, the system collects the approver identity information WG001 from the running status feedback value, and then triggers the permission verification process.

[0272] The system obtains the node identifier 02 (review node) of the current approval node and the approver identifier WG001 of the current approver, constructs an authorization verification request, and sends the request to the independent authorization center service. The authorization center is responsible for maintaining real-time authorization data for all users, including basic job permissions, temporary authorizations, proxy authorizations, etc., and updating them in real time.

[0273] The permission center queries all associated permission records based on the approver identifier WG001 and returns the real-time permission data to the approval system. The returned data is organized in JSON format and contains multiple permission records, each containing at least the following fields:

[0274] Permission object: Identifies the approval node or functional object to which this permission applies, for example, 02 represents the review node permission;

[0275] Effective timestamp: This permission record indicates when it began to take effect;

[0276] Expiration timestamp: This permission record indicates when it expired.

[0277] Assume the real-time permission data returned by the permission center is as follows:

[0278] {

[0279] "userId":"WG001",

[0280] "permissions":[

[0281] {

[0282] "permissionObject":"02",

[0283] "effectiveTime":"2024-03-0100:00:00",

[0284] "expireTime":"2024-03-1023:59:59"

[0285] },

[0286] {

[0287] "permissionObject":"02",

[0288] "effectiveTime":"2024-03-1208:00:00",

[0289] "expireTime":"2024-03-1518:00:00"

[0290] },

[0291] {

[0292] "permissionObject":"03",

[0293] "effectiveTime":"2024-03-0100:00:00",

[0294] "expireTime":"2024-12-3123:59:59"

[0295] } ]

[0297] }

[0298] The above data indicates that: Engineer Wang had review node permissions (old authorization) from March 1 to March 10, and had review node permissions again (new authorization or renewal) from March 12 to March 15 at 18:00. Meanwhile, Engineer Wang has always had final review node permissions (permission object 03).

[0299] Step S603: Filter the set of valid permission records

[0300] The system obtains the current system timestamp T_sys. Assuming Wang opened the task at 15:05:00 on March 13, 2024, the system converts it to a timestamp format (e.g., a Unix timestamp). The system iterates through the returned list of permission records, extracts the effective and expiration timestamps for each record, compares them one by one with the current system timestamp, and filters out valid permission records that satisfy the condition: effective timestamp ≤ current system timestamp ≤ expiration timestamp.

[0301] Regarding the three records above:

[0302] Record 1: Effective 2024-03-01 00:00:00, Expired 2024-03-10 23:59:59, Current time 2024-03-13 15:05:00 has exceeded the expiration time, therefore the condition is not met;

[0303] Record 2: Effective 2024-03-12 08:00:00, Expired 2024-03-15 18:00:00. The current time is within this interval, thus meeting the conditions.

[0304] Record 3: Effective 2024-03-01 00:00:00, Expired 2024-12-31 23:59:59. The current time is within this range, which meets the conditions, but its permission object is 03 (final review node), which is unrelated to the current review node.

[0305] Therefore, the set of valid permission records includes record 2 and record 3, but the permission object of record 3 is not the current node.

[0306] Step S604: Match and verify the node identifier with the permission object field in the set of valid permission records.

[0307] The system compares the node identifier 02 of the current approval node with the permission object field of each record in the set of valid permission records:

[0308] For record 2: the permission object is 02, which matches the node identifier 02;

[0309] For record 3: the permission object is 03, which does not match the node identifier 02.

[0310] Step S605: Determine the permission verification result

[0311] Since at least one permission record (record 2) in the valid permission record set has a permission object field that matches the node identifier, the system determines the approval permission of the current approver, Wang Gongshu, for the review node, and the permission verification result is that the approver has the approval permission.

[0312] Step S606: Handling insufficient permissions (not triggered in this scenario)

[0313] If the set of valid permission records is empty, or if the permission objects of all valid records do not match the node identifier, the permission verification result is that the user does not have approval permission. In this case, the system will generate a permission insufficient message, such as: "You currently do not have approval permission for the review node. Please contact the administrator." This message will be simultaneously pushed to the current approver's end (Wang's APP / PC) and the job initiator's end (such as the on-site safety officer) so that relevant personnel can handle the situation promptly.

[0314] Example continuation: In this scenario, if permission verification passes, the system will continue with the following processes: collecting the progress of the approval task and the submission status of approval comments, and proceeding to the next step (such as pushing to the next node or generating a timeout reminder) based on whether the conditions for progress are met. If Engineer Wang opens the task on March 11 (the permission gap period), the set of valid permission records will be empty, and permission verification will fail. The system will immediately push a permission insufficient prompt to avoid erroneous approvals due to invalid permissions.

[0315] Through the aforementioned dynamic permission verification mechanism, permission verification in this embodiment no longer relies on a static list. Instead, it obtains real-time permission data, including effective and expiration times, from the permission center, enabling precise adaptation to dynamic scenarios such as temporary authorization, job adjustments, and proxy authorization. Simultaneously, by comparing timestamps, it ensures that only permissions valid at the current moment are recognized, effectively avoiding security risks caused by expired or prematurely invalidated permissions. When permissions are insufficient, the system immediately pushes a prompt, ensuring that the cause of process blockage is transparent and traceable.

[0316] Building upon the solutions for dynamic form generation, intelligent prediction of push timing, and dynamic permission verification, the approval timeliness detection process still faces the challenge of a disconnect between static timing and the approver's actual working time. Traditional fixed thresholds (such as 24 hours) do not take into account holidays, weekends, and non-working hours, resulting in the timer continuing to run even when the approver is not actually on duty. This can easily lead to misjudgments of timeouts, triggering invalid reminders or even process interruptions. For example, an approval task initiated before the end of the workday on Friday, if timed according to 24 hours, will be judged as timeout by the end of the workday on Saturday, even though the approver only processes it on weekdays. This static timing method cannot accurately reflect the effective working time window. Therefore, it is necessary to extend the aforementioned online pattern prediction idea based on historical behavior analysis to the timeliness detection process, establishing a dynamic timeliness calculation mechanism that automatically deducts holidays and non-working hours based on the approver's historical working hours, so that the timeliness threshold matches the actual available approval time, thereby improving the accuracy of timeliness detection and the rationality of process control. To this end, this embodiment obtains the following timeliness detection results:

[0317] S701. Obtain the task push time T_start, the current time T_end, and the approval time limit T_limit for the current approval node; for example, assume that the dredging operation of Tank No. 3 entered the review node at 14:30 on March 13, 2024 (Thursday). Based on the dynamic push time detection mechanism, the system determines the task push time to be 15:00 and pushes the dedicated approval form for the review node to the approver, Engineer Wang, at 15:00. Engineer Wang opens the APP at 15:05 to view the tasks awaiting approval and completes the approval submission at 15:07:45. At this time, the system enters the time limit detection stage and obtains the following data:

[0318] Task push time T_start: 2024-03-13 15:00:00 (i.e., the time when the task enters the review node and is pushed to the approver)

[0319] Current time T_end: 2024-03-13 15:07:45 (i.e., the time when Engineer Wang submitted the approval).

[0320] Approval time limit T_limit: 60 minutes (the preset approval time limit for the review node)

[0321] S702. Obtain preset work calendar data. The work calendar data includes the date type of each date and the start time S_work and end time E_work of the effective working hours for each workday. The date type is used to distinguish between workdays, rest days, and statutory holidays. For example, the work calendar set in this embodiment is as follows: {

[0322] "calendar":[

[0323] {"date":"2024-03-13","dateType":"Workday","workStart":"09:00:00","workEnd":"12:00:00"},

[0324] {"date":"2024-03-13","dateType":"Workday","workStart":"13:00:00","workEnd":"18:00:00"},

[0325] {"date":"2024-03-14","dateType":"Workday","workStart":"09:00:00","workEnd":"12:00:00"},

[0326] {"date":"2024-03-14","dateType":"Workday","workStart":"13:00:00","workEnd":"18:00:00"},

[0327] {"date":"2024-03-15","dateType":"Workday","workStart":"09:00:00","workEnd":"12:00:00"},

[0328] {"date":"2024-03-15","dateType":"Workday","workStart":"13:00:00","workEnd":"18:00:00"},

[0329] {"date":"2024-03-16","dateType":"Rest Day"},

[0330] {"date":"2024-03-17","dateType":"Rest Day"},

[0331] / / Other date configurations... ]

[0333] };

[0334] The date types include weekdays, rest days, and statutory holidays; each weekday can contain the start time S_work and end time E_work of multiple valid working periods (e.g., 09:00-12:00 AM, 13:00-18:00 PM).

[0335] S703. Based on the work calendar data, identify all time sub-intervals from T_start to T_end that satisfy the date type as a workday and the time within the range [S_work, E_work]. Calculate the length of each time sub-interval and sum them to obtain the valid approval time T_valid. For example, the system divides the time interval from T_start to T_end into consecutive time periods and determines whether each time period belongs to valid working time.

[0336] T_start=15:00:00 and T_end=15:07:45, both of which fall within the valid working hours of 13:00-18:00 on March 13, 2024 (working day).

[0337] Therefore, the entire time interval from 15:00:00 to 15:07:45 is considered valid working time.

[0338] The system calculates the length of this time interval: 7 minutes and 45 seconds, which is 0.129 hours (for easy comparison, the unit can be the same as the threshold, which is minutes).

[0339] T_valid = 7 minutes 45 seconds ≈ 7.75 minutes.

[0340] S704. Compare the valid approval duration T_valid with the approval time limit T_limit:

[0341] S705. If T_valid ≥ T_limit, it is determined that the aging detection result is a timeout.

[0342] S706. If T_valid < T_limit, it is determined that the aging detection result is not a timeout.

[0343] To more clearly show the processing effect of this mechanism on non - working periods, assume another scenario: The preliminary review node completed the approval at 16:30 on Friday, March 15, 2024. The review node entered at 16:30, and the task push time T_start = 2024 - 03 - 15 16:30:00. Since 16:30 on Friday is within the valid working period of 13:00 - 18:00 on a working day, the task was pushed normally. However, due to reasons, Engineer Wang failed to process it in time and didn't open the approval task until 09:05 on Monday (March 18). At this time, the current time T_end = 2024 - 03 - 18 09:05:00.

[0344] The process of the system calculating the effective approval duration T_valid is as follows:

[0345] The time interval from T_start to T_end is from 16:30 on March 15 to 09:05 on March 18, during which it includes:

[0346] Valid working period on March 15 (Friday): 16:30 - 18:00, with a duration of 1.5 hours.

[0347] March 16 (Saturday): A rest day, not counted at all.

[0348] March 17 (Sunday): A rest day, not counted at all.

[0349] Valid working period on March 18 (Monday): 09:00 - 09:05, with a duration of 5 minutes (about 0.083 hours).

[0350] Accumulate the valid working duration: T_valid = 1.5 + 0.083 = 1.583 hours ≈ 95 minutes.

[0351] If the aging threshold T_limit of the review node is 2 hours (120 minutes), then T_valid < T_limit, and it is determined as not a timeout; if the threshold is 1 hour (60 minutes), then T_valid > T_limit, and it is determined as a timeout.

[0352] This embodiment introduces a dynamic timeliness detection mechanism based on a work calendar, optimizing the traditional continuous timing based on fixed threshold natural time into a cumulative timing based on effective working time. It automatically identifies and removes rest days, statutory holidays, and non-working hours from the time interval T_start to T_end, only accumulating the actual usable effective approval time T_valid and comparing it with the threshold T_limit to accurately determine whether a timeout has occurred. This mechanism ensures that the timeliness detection result strictly matches the approver's actual working time, effectively avoiding misjudgments of timeouts caused by continuous timing during weekends or holidays when the approver is not on duty. It eliminates the risk of invalid reminders and process interruptions, significantly improving the accuracy and rationality of timeliness control in confined space operation approval processes, and further strengthening the scientific nature and feasibility of approval time limits in high-risk operation scenarios.

[0353] Building upon dynamic push timing prediction and permission verification, the next node push process for approval tasks still faces the challenge of a singular priority system: existing methods determine the push order solely based on the approver's permission level and historical response time, ignoring the task's time urgency (e.g., the work schedule start time is approaching) and the approver's current workload. For example, when two tasks are pushed to the same approver simultaneously, urgent dredging operations may be delayed because they are scheduled after routine inspection tasks; or a heavily loaded approver may continuously receive new tasks, leading to a backlog, while idle approvers may not be able to take on tasks in a timely manner. This single-dimensional priority mechanism cannot achieve a balance between task urgency, approver's real-time workload, and historical performance. There is an urgent need to establish a dynamic priority calculation model that integrates multiple factors to ensure that urgent tasks are prioritized and approval resources are allocated rationally. Therefore, this embodiment generates a push priority based on the approver's permission level and historical approval response time for the next approval node, including:

[0354] Obtain the set of pending approval tasks to be pushed to the same approver at the current time. Each pending approval task in the set is associated with the next approval node identifier, the urgency level value of the current task, and the task application materials.

[0355] For each pending approval task, obtain the approver identifier corresponding to the next approval node identifier, and obtain the permission level value, historical average approval response time and current number of pending tasks corresponding to the approver identifier;

[0356] A basic priority coefficient is set based on the emergency level value, and the basic priority coefficient is positively correlated with the emergency level value.

[0357] The permission weight is set according to the permission level value, and the permission weight is positively correlated with the permission level value.

[0358] A response speed weight is set based on the historical average approval response time, and the response speed weight is negatively correlated with the historical average approval response time.

[0359] A load weight is set based on the current number of pending tasks, and the load weight is negatively correlated with the current number of pending tasks.

[0360] The basic priority coefficient, permission weight, response speed weight, and load weight are weighted and summed according to preset weight coefficients to obtain the comprehensive priority value of each task to be pushed for approval; the preset weight coefficients in this embodiment are set by those skilled in the art.

[0361] According to the overall priority value from high to low, each task to be pushed for approval is pushed to the approver's terminal in sequence.

[0362] For example, the dredging operation of Tank No. 3 in this embodiment was approved at the review node at 15:07:45 on March 13, 2024. The system determined that the conditions for process advancement were met and determined that the next approval node was the final review node (node ​​identifier 03). The approval task needs to be pushed to the final approver - Vice President Sun in charge (approver identifier SG001).

[0363] At the final review task push time (March 13, 2024, 15:07:45), the system scanned all approval tasks awaiting push to General Manager Sun (SG001) and constructed a set of tasks to be pushed for approval. At this time, in addition to the dredging operation of Tank No. 3, another routine inspection operation of Tank Area No. 2 had also completed its review and was awaiting General Manager Sun's final review. The specific information of the two tasks is shown in Table 1 below:

[0364] Table 1: Task Information Table

[0365]

[0366] The system encapsulates the above two tasks into a set of tasks to be pushed for approval: {TASK-001, TASK-002}.

[0367] The system retrieves the corresponding approver identifier SG001 (General Manager Sun) from the approval parameter set based on the final review node identifier 03. It then obtains relevant indicators for General Manager Sun from the user profile database.

[0368] Permission level value: 3 (The final review node is the highest permission level, with a preset value range of 1-3, and 3 being the highest).

[0369] Historical average approval response time: 2.5 hours (based on historical data from the past 3 months)

[0370] Current number of pending tasks: 2 (i.e., the two tasks mentioned above that are yet to be pushed to the system; at this time, Mr. Sun has no other unprocessed tasks).

[0371] The system calculates a comprehensive priority value for each task to be pushed in the set according to preset rules. The preset weight coefficients are: basic priority coefficient 0.4, permission weight 0.2, response speed weight 0.2, and load weight 0.2.

[0372] For TASK-001 (dredging operation of storage tank No. 3):

[0373] Emergency level value = 5 → Base priority coefficient = 5 × 0.4 = 2.0

[0374] Permission level value = 3 → Permission weight = 3 × 0.2 = 0.6

[0375] Historical average approval response time = 2.5 hours → Response speed weight = (1 / 2.5) × 0.2 = 0.08

[0376] Current number of pending tasks = 2 → Load weight = (1 / 2) × 0.2 = 0.1

[0377] Overall priority value = 2.0 + 0.6 + 0.08 + 0.1 = 2.78

[0378] For TASK-002 (Daily inspection of Tank Area No. 2):

[0379] Emergency level value = 2 → Base priority coefficient = 2 × 0.4 = 0.8

[0380] Permission level value = 3 → Permission weight = 3 × 0.2 = 0.6

[0381] Historical average approval response time = 2.5 hours → Response speed weight = (1 / 2.5) × 0.2 = 0.08

[0382] Current number of pending tasks = 2 → Load weight = (1 / 2) × 0.2 = 0.1

[0383] Overall priority value = 0.8 + 0.6 + 0.08 + 0.1 = 1.58

[0384] The system sorts TASK-001 (2.78) from highest to lowest overall priority value: TASK-001 (2.78) > TASK-002 (1.58). Therefore, the system first pushes the final approval form and application materials for the dredging operation of Tank No. 3 to Mr. Sun's APP and PC; after TASK-001 is pushed, TASK-002 is pushed immediately.

[0385] At 15:08, Mr. Sun received two pending notifications simultaneously, but the app displayed them in priority order, with the dredging task pinned to the top. Upon opening the app, he first saw the urgent dredging task, which he could handle immediately; while the routine inspection task, though to be handled later, had a more flexible timeframe and wouldn't disrupt the overall workflow. If the traditional single-dimensional (permission level only) sorting method had been used, the two tasks, due to their identical permission levels, would have been pushed side-by-side, potentially leading Mr. Sun to randomly select one or process it in list order, posing a risk of delaying the urgent task.

[0386] This embodiment introduces a dynamic priority calculation model that integrates multiple factors. It weights and quantifies the urgency level of the task, the approver's authority level, the historical average approval response time, and the current number of pending tasks to generate a comprehensive priority value, which is then pushed sequentially. This effectively solves the problem of traditional single-dimensional sorting neglecting task urgency and the real-time load of approvers. For example, when the dredging operation of tank No. 3 and the daily inspection operation of tank area No. 2 are simultaneously pushed to the final approver, the dredging operation, due to its higher urgency level (scheduled to start at 16:00), receives a higher basic priority coefficient. Although the authority levels are the same, its comprehensive priority value is significantly higher than that of the inspection task, ensuring that urgent tasks are displayed at the top of the app and processed with priority, avoiding delays caused by unreasonable sorting. Simultaneously, by introducing a load weight (negatively correlated with the number of pending tasks), it prevents excessive backlog for heavily loaded approvers, achieving a dynamic and balanced allocation of approval resources. This mechanism significantly improves response efficiency and resource utilization rationality in multi-task concurrent scenarios, further strengthening the timeliness guarantee capability for approving high-risk operations in confined spaces.

[0387] In this embodiment, when an approval timeout occurs, the traditional method only sends a single reminder. If the approver does not address it promptly, it fails to provide effective oversight, allowing the timeout issue to escalate and ultimately delaying the start of the operation. Furthermore, the reminder only includes a simple reason for the timeout, lacking guidance on subsequent actions. The approver may not know how to handle the situation, and superiors cannot intervene in a timely manner. In addition, the timeout levels are not differentiated, making it impossible to distinguish between minor and severe timeouts and to implement differentiated response measures. Therefore, this embodiment generates an approval timeout reminder that includes the reason for the timeout or missing information, including:

[0388] S101. Obtain the node identifier, current approver identifier, job initiator identifier, and superior management identifier of the current approval node;

[0389] S102. Obtain the task push time T_start, the current time T_end, and the approval time limit T_limit of the current approval node;

[0390] S103. Calculate the current timeout duration T_over based on T_start, T_end, and T_limit: T_over = T_end - (T_start + T_limit);

[0391] S104. Calculate the timeout ratio R_over = T_over / T_limit;

[0392] S105. Obtain the preset first-level timeout threshold R1 and second-level timeout threshold R2, where R1 < R2; the first-level timeout threshold R1 and the second-level timeout threshold R2 are proportional values preset according to the tolerance of the business for the severity of approval timeout. For example, in the embodiment, R1 = 0.5 and R2 = 1.0. Here, R1 indicates that a first-level timeout is triggered when the timeout duration has reached 50% of the time limit threshold, and R2 indicates that a second-level timeout is triggered when the timeout duration has reached 100% of the time limit threshold; these two thresholds can be flexibly configured in the system by the system administrator according to the job risk level, job responsibility requirements, and management specifications. The higher the risk or the more critical the position, the lower the threshold can be accordingly to strengthen the time limit control.

[0393] S106. Compare R_over with R1 and R2 respectively:

[0394] If R1 ≤ R_over < R2, determine that the timeout level is a first-level timeout;

[0395] If R_over ≥ R2, determine that the timeout level is a second-level timeout;

[0396] S***107. Obtain the operation status feedback value of the current approval node, and extract the missing information items and timeout reasons according to the approval task processing progress and approval opinion submission situation in the operation status feedback value;

[0397] S108. Generate corresponding timeout reminder content according to the timeout level:

[0398] If it is a first-level timeout, generate a first-level timeout reminder instruction including the timeout reason, missing information items, and recommended handling method, and push it to the current approver's end;

[0399] If it is a second-level timeout, generate a second-level timeout reminder instruction including the timeout reason, missing information items, and recommended handling method, and push it to the current approver's end, the superior management end, and the job initiator's end at the same time;

[0400] S109. In the approval process progress record, mark the timeout level of the current node as a first-level timeout or a second-level timeout, and record the timeout reminder push timestamp.

[0401] For example, in the approval process for the dredging operation of Tank No. 3 in this embodiment, after the review node is completed, the system pushes the final review node's dedicated approval form to the approver, Mr. Sun (SG001), at 15:08 on March 13, 2024. The final review node approval time limit T_limit is 120 minutes, meaning the task should be completed within 120 minutes of being pushed at 15:08 (i.e., before 17:08 on the same day). However, Mr. Sun was out and unable to process it in time. As of 10:30 on March 14 the following day, the system detected that the task was still not completed, triggering the timeout reminder generation process. The specific steps are as follows:

[0402] The system obtains the current approval node identifier 03 (final review node), the current approver identifier SG001, the job initiator identifier LISI (safety officer Li Si), the superior management identifier ZONGJINGLI (general manager), the task push time T_start=2024-03-13 15:08:00, the current time T_end=2024-03-14 10:30:00, and the approval time limit threshold T_limit=120 minutes.

[0403] Theoretically, the latest completion time is T_start + T_limit = 2024-03-13 17:08:00.

[0404] Timeout duration T_over = T_end - (T_start + T_limit) = 2024-03-14 10:30:00 -2024-03-13 17:08:00 = 17 hours 22 minutes = 1042 minutes.

[0405] The timeout ratio R_over = T_over / T_limit = 1042 / 120 ≈ 8.68.

[0406] The system presets a Level 1 timeout threshold of R1=0.5 and a Level 2 timeout threshold of R2=1.0. Since R_over=8.68 ≥ R2, the timeout level is determined to be Level 2.

[0407] Query the task's running status feedback values ​​(up to the current time):

[0408] {

[0409] "approvalId": "APP-20240313-005",

[0410] "nodeId": "03",

[0411] "approverId": "SG001",

[0412] "onlineStatus": "Offline",

[0413] "taskProgress": "Not started",

[0414] "submissionStatus": "No required information was submitted",

[0415] "progressStartTime": null,

[0416] "progressCompleteTime": null,

[0417] "timestamp": "2024-03-14 10:30:00"

[0418] }

[0419] Based on this, the missing information items are that neither the overall risk confirmation nor the final approval opinion was filled in, and the reason for the timeout is that the approver did not process it.

[0420] Because the timeout level is Level 2, the system has generated a Level 2 timeout reminder, which includes: Reason for timeout: approver not processed; overall risk confirmation for missing information items; final approval opinion; suggested handling method. Please log in to the system as soon as possible to complete the approval. If you are unable to handle it, please authorize someone else. This reminder is also pushed to:

[0421] Current approver's platform: Mr. Sun's (SG001) APP and PC platform;

[0422] Upper-level management: The General Manager's (ZONGJINGLI) APP and PC;

[0423] Task initiation platform: Safety officer Li Si's (LISI) APP and PC.

[0424] In the approval process progress record, locate the final review node (identified as 03) and mark the timeout level as Level 2 timeout, and record the timeout reminder push timestamp 2024-03-14 10:30:00.

[0425] If the timeout ratio is only between 0.5 and 1.0 (Level 1 timeout), the reminder will only be sent to the current approver to avoid excessive interference. Through the tiered timeout reminder mechanism, the system achieves differentiated responses to the severity of timeouts, ensuring that urgent timeouts can be reported to relevant parties in a timely manner, effectively guaranteeing the timeliness and controllability of high-risk operation approvals.

[0426] This embodiment introduces a tiered timeout reminder mechanism, classifying timeout levels into Level 1 and Level 2 based on the timeout ratio, and dynamically adjusting the push notification scope and reminder content. This effectively solves the problems of insufficient supervision, ambiguous information, and limited response measures associated with traditional single-reminder reminders. Taking the timeout of the final review node for the dredging operation of Tank No. 3 as an example, when the timeout exceeds 8.68 times the threshold, the system automatically determines it as a Level 2 timeout. It simultaneously pushes a reminder containing the specific reason for the timeout (approver's failure to process), missing information items (overall risk confirmation, final approval opinion), and handling suggestions to the approver, the superior management, and the operation initiator. This achieves differentiated responses based on the severity of the timeout, ensuring that urgent timeouts can trigger multi-level collaborative intervention in a timely manner. For minor timeouts (Level 1), only the approver is reminded, avoiding excessive interference. This mechanism not only strengthens the effective supervision of timeout issues but also balances process control and resource consumption through tiered push notifications, significantly improving the accuracy, coordination, and controllability of timeout handling in the approval of high-risk operations in confined spaces.

[0427] Example 2

[0428] Please see Figure 2 Another embodiment of the present invention provides: an intelligent push system for confined space operation approval process, comprising:

[0429] The process configuration module is configured to: obtain a set of pending approval node identifiers corresponding to the target job application number, wherein the set of pending approval node identifiers contains unique identifiers of the initial review node, secondary review node and final review node that the target job application must pass through in the approval process; obtain a set of approval parameters associated with the pending approval node identifiers, wherein the set of approval parameters includes a list of approver permissions for each approval node, mandatory approval field requirements and approval time threshold;

[0430] The form generation module is configured to: map and convert the target job application information according to the required fields in the approval parameter set, and generate a unique approval form corresponding to each approval node. The unique approval form contains the approval opinion field and the required approval information to be filled in by the approval node.

[0431] The push scheduling module is configured to: when the task push time of the current approval node is detected to arrive, obtain the real-time approval task cache maintained for the current approval node, and refresh the approval task content in the real-time approval task cache using the exclusive approval form of the current approval node;

[0432] The status acquisition module is configured to: collect the online status of the approver, the progress of the approval task processing, and the status of the submission of approval opinions for the current approval node.

[0433] The permission verification module is configured to match and verify the approver's identity information in the running status feedback value with the current approver's real-time permission data to obtain the permission verification result.

[0434] The timeliness detection module is configured to: compare the approval task processing progress in the running status feedback value with the approval timeliness threshold to obtain the timeliness detection result;

[0435] The process advancement decision module is configured to: when the permission verification result indicates that the user has approval authority and the timeliness detection result indicates that the timeout has not expired, obtain the processing progress of the approval task and the submission status of the approval opinion;

[0436] If the processing progress of the approval task is completed and the submission status of the approval opinion is that all required information has been submitted, then the conditions for process advancement are met, the next approval node identifier is determined, and the approval task containing the exclusive approval form and work application materials is pushed to the approver's APP or PC terminal of the next approval node according to the approval authority level and historical approval response time. At the same time, the current node status and the timestamp of entering the next node in the approval process progress record are updated.

[0437] Otherwise, if the conditions for process advancement are not met, an approval timeout reminder containing the reason for the timeout or missing information is generated based on the timeliness detection result and the approval task processing progress. The approval timeout reminder is pushed to the approver's end and the job initiator's end of the current approval node, and the abnormal status of the current node is marked in the approval process progress record.

[0438] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments under the guidance of the present invention without departing from the spirit and scope of the present invention. All of these variations are within the protection scope of the present invention.

[0439] If the technical solution disclosed herein involves personal information, the product using this technical solution has clearly informed the user of the personal information processing rules and obtained the user's voluntary consent before processing the personal information. If the technical solution disclosed herein involves sensitive personal information, the product using this technical solution has obtained the user's separate consent before processing the sensitive personal information, and also meets the requirement of explicit consent. For example, at personal information collection devices such as cameras, clear and prominent signs are set up to inform users that they have entered the scope of personal information collection and that personal information will be collected. If an individual voluntarily enters the collection scope, it is deemed that they have agreed to the collection of their personal information; or on the personal information processing device, with clear signs / information informing users of the personal information processing rules, authorization is obtained from the individual through pop-up information or by asking the individual to upload their personal information; wherein, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the types of personal information processed.

Claims

1. A method for intelligently pushing approval processes for confined space operations, characterized in that, include: Obtain the set of pending approval node identifiers corresponding to the target job application number. The set of pending approval node identifiers contains the unique identifiers of the initial review node, secondary review node and final review node that the target job application needs to go through in the approval process. Obtain the set of approval parameters associated with the identifier of the node to be approved, wherein the set of approval parameters includes the list of approver permissions for each approval node, the required fields for approval, and the approval time limit threshold; The target job application information is mapped and formatted according to the required fields in the approval parameter set to generate a unique approval form corresponding to each approval node. The unique approval form contains the approval opinion field and the required approval information to be filled in by the approval node. When the task push time of the current approval node is detected, the real-time approval task cache maintained for the current approval node is obtained, and the approval task content in the real-time approval task cache is refreshed using the exclusive approval form of the current approval node. Collect operational status feedback values ​​for the current approval node, including the online status of the approver, the progress of the approval task processing, and the submission status of the approval opinions; The approver's identity information in the operation status feedback value is matched and verified with the current approver's real-time permission data to obtain the permission verification result; The timeliness is compared with the approval task processing progress in the operation status feedback value and the approval timeliness threshold to obtain the timeliness detection result.

2. The intelligent push method for confined space operation approval process as described in claim 1, characterized in that, The intelligent push method for the approval process also includes: When the permission verification result indicates that the user has approval authority and the timeliness detection result indicates that the timeout has not expired, the processing progress of the approval task and the submission status of the approval comments are obtained. If the processing progress of the approval task is completed and the submission status of the approval opinion is that all required information has been submitted, then the conditions for process advancement are met, the next approval node identifier is determined, and the approval task containing the exclusive approval form and work application materials is pushed to the approver's APP or PC terminal of the next approval node according to the approval authority level and historical approval response time. At the same time, the current node status and the timestamp of entering the next node in the approval process progress record are updated. Otherwise, if the conditions for process advancement are not met, an approval timeout reminder containing the reason for the timeout or missing information is generated based on the timeliness detection result and the approval task processing progress. The approval timeout reminder is pushed to the approver's end and the job initiator's end of the current approval node, and the abnormal status of the current node is marked in the approval process progress record.

3. The intelligent push method for confined space operation approval process as described in claim 2, characterized in that, The step of mapping and formatting the target job application information according to the mandatory field requirements in the approval parameter set includes: Obtain the node type identifier of the current approval node and the job type identifier of the current operation; the node type identifier is a preset type code used to uniquely distinguish between the initial review node, the secondary review node and the final review node; the job type identifier is a preset category code used to uniquely distinguish between daily inspection operations, maintenance operations and cleaning operations. Using the combination of the node type identifier and the job type identifier as a query index, a target form template uniquely associated with the query index is retrieved from a pre-built form template library. The form template library stores multiple form templates, each of which is pre-configured with a corresponding combination of node type identifier and job type identifier. Each form template also defines the set of required fields for that node type under that job type, as well as the field mapping relationship between each required field and the data fields in the target job application information. Parse the target form template to extract the set of required fields and the field mapping relationship corresponding to each required field in the set of required fields; Based on the field mapping relationship, the field data corresponding to each required field is read from the target job application information, and the read field data is written into the required field position of the target form template to generate the exclusive approval form for the current approval node; the exclusive approval form contains the set of required fields and the filled field data.

4. The intelligent push method for confined space operation approval process as described in claim 3, characterized in that, The step of detecting when the task push time for the current approval node arrives includes: Obtain the node identifier of the current approval node, the approver identifier of the current approver, and the approval time limit threshold of the current approval node; Obtain historical online behavior data and real-time online status value associated with the approver's identifier. The historical online behavior data includes the approver's daily login time sequence and logout time sequence within a preset historical period. The real-time online status value is used to characterize whether the approver is currently online or offline. Statistical analysis is performed on the historical online behavior data to extract the online probability distribution characteristics of the approver in each time period; Based on the online probability distribution characteristics, starting from the current time and using the approval timeliness threshold as the prediction window length, predict a expected online time periods for the approver within the prediction window. Each expected online time period includes an estimated start time and an estimated duration. Obtain the current system time and match it with the real-time online status value: If the real-time online status value is online, then the current moment is determined to be the time when the task is pushed. If the real-time online status value is offline, then the estimated start time of the estimated online period with the smallest difference from the current time is selected from the at least one estimated online period, and the estimated start time of the estimated online period is determined as the task push time. When the system clock reaches the task push time, the approval task push instruction is triggered, and the exclusive approval form of the current approval node is pushed to the approver's end.

5. The intelligent push method for confined space operation approval process as described in claim 4, characterized in that, The obtained permission verification result includes: Send a permission verification request containing the approver's identifier to the permission center to obtain real-time permission data corresponding to the approver's identifier; Receive real-time permission data returned by the permission center. The real-time permission data contains m permission records. Each permission record contains a permission object field, an effective timestamp, and an expiration timestamp. Extract the effective timestamp and expiration timestamp of each permission record in the real-time permission data, and compare them one by one with the current system timestamp to filter out a set of valid permission records that satisfy the condition: effective timestamp ≤ current system timestamp ≤ expiration timestamp; Match and verify the node identifier with the permission object field in the set of valid permission records: If there is at least one permission record in the set of valid permission records whose value of the permission object field matches the node identifier, determine that the permission verification result is having approval permission; Otherwise, determine that the permission verification result is not having approval permission. When the permission verification result is not having approval permission, generate a permission deficiency prompt message and push the permission deficiency prompt message to the approver side and the job initiation side of the current approval node.

6. The intelligent push method for confined space operation approval process as described in claim 5, characterized in that, The obtaining of the aging detection result includes: Obtain preset work calendar data, where the work calendar data includes the date type of each date and the start time S_work and end time E_work of the effective working period of each working day, and the date type is used to distinguish working days, rest days, and legal holidays; Based on the work calendar data, identify all time sub-intervals that satisfy the date type being a working day and the time being within the range of [S_work, E_work] in the time interval from T_start to T_end, calculate the length of each time sub-interval and accumulate them to obtain the effective approval duration T_valid; Compare the effective approval duration T_valid with the approval aging threshold T_limit: If T_valid ≥ T_limit, determine that the aging detection result is overtime; If T_valid < T_limit, determine that the aging detection result is not overtime.

7. The intelligent push method for confined space operation approval process as described in claim 6, characterized in that, The generating of the push priority according to the approval authority level of the next approval node and the historical approval response duration includes: Obtain the set of pending approval tasks to be pushed to the same approver side at the current moment. Each pending approval task in the set of pending approval tasks is associated with a next approval node identifier, the emergency level value of the current job task, and the job application materials; For each pending approval task, obtain the approver identifier corresponding to the next approval node identifier, and obtain the authority level value, historical average approval response duration, and the current number of pending tasks corresponding to the approver identifier; Set the basic priority coefficient according to the emergency level value, and the basic priority coefficient has a positive correlation with the emergency level value; Set the authority weight according to the authority level value, and the authority weight has a positive correlation with the authority level value; Set the response speed weight according to the historical average approval response duration, and the response speed weight has a negative correlation with the historical average approval response duration; Set the load weight according to the current number of pending tasks, and the load weight has a negative correlation with the current number of pending tasks; Perform a weighted sum of the basic priority coefficient, authority weight, response speed weight, and load weight according to the preset weight coefficient to obtain the comprehensive priority value of each pending approval task; Push each approval task to be pushed to the approver terminal in order from high to low according to the comprehensive priority value.

8. The intelligent push method for confined space operation approval process as described in claim 6, characterized in that, Generating an approval timeout reminder including the timeout reason or missing information item includes: Obtain the node identifier of the current approval node, the current approver identifier, the job initiation terminal identifier, and the superior management terminal identifier; Obtain the task push time T_start of the current approval node, the current time T_end, and the approval time limit threshold T_limit; Calculate the current timeout duration T_over = T_end - (T_start + T_limit) according to T_start, T_end, and T_limit; Calculate the timeout ratio R_over = T_over / T_limit; Obtain the preset first-level timeout threshold R1 and second-level timeout threshold R2, where R1 < R2; Compare R_over with R1 and R2 respectively: If R1 ≤ R_over < R2, determine that the timeout level is a first-level timeout; If R_over ≥ R2, determine that the timeout level is a second-level timeout.

9. The intelligent push method for confined space operation approval process as described in claim 6, characterized in that, Generating an approval timeout reminder including the timeout reason or missing information item further includes: Obtain the operation status feedback value of the current approval node, and extract the missing information item and timeout reason according to the approval task processing progress and approval opinion submission situation in the operation status feedback value; Generate corresponding timeout reminder content according to the timeout level: If it is a first-level timeout, generate a first-level timeout reminder instruction including the timeout reason, missing information item, and recommended handling method, and push it to the current approver terminal; If it is a second-level timeout, generate a second-level timeout reminder instruction including the timeout reason, missing information item, and recommended handling method, and push it to the current approver terminal, the superior management terminal, and the job initiation terminal at the same time; In the approval process progress record, mark the timeout level of the current node as a first-level timeout or a second-level timeout, and record the timeout reminder push timestamp.

10. A confined space operation approval process intelligent push system, implemented based on any one of claims 1-9, characterized in that, Including: A process configuration module, configured to: obtain a set of approval node identifiers corresponding to the target job application number, and the set of approval node identifiers includes the unique identifiers of the preliminary review node, the review node, and the final review node that the target job application needs to pass through in sequence in the approval process; obtain a set of approval parameters associated with the set of approval node identifiers, and the set of approval parameters includes the approver authority list, approval mandatory item requirements, and approval time limit threshold for each approval node; A form generation module, configured to: perform field mapping and format conversion on the target job application information according to the approval mandatory item requirements in the set of approval parameters, and generate a dedicated approval form corresponding to each approval node, and the dedicated approval form includes the approval opinion field and approval mandatory information required for the approval node; A push scheduling module, configured to: when it is detected that the task push time of the current approval node arrives, obtain the real-time approval task cache maintained for the current approval node, and refresh the approval task content in the real-time approval task cache with the dedicated approval form of the current approval node; The status acquisition module is configured to: collect the online status of the approver, the progress of the approval task processing, and the status of the submission of approval opinions for the current approval node. The permission verification module is configured to match and verify the approver's identity information in the running status feedback value with the current approver's real-time permission data to obtain the permission verification result. The timeliness detection module is configured to: compare the approval task processing progress in the running status feedback value with the approval timeliness threshold to obtain the timeliness detection result; The process advancement decision module is configured to: when the permission verification result indicates that the user has approval authority and the timeliness detection result indicates that the timeout has not expired, obtain the processing progress of the approval task and the submission status of the approval opinion; If the processing progress of the approval task is completed and the submission status of the approval opinion is that all required information has been submitted, then the conditions for process advancement are met, the next approval node identifier is determined, and the approval task containing the exclusive approval form and work application materials is pushed to the approver's APP or PC terminal of the next approval node according to the approval authority level and historical approval response time. At the same time, the current node status and the timestamp of entering the next node in the approval process progress record are updated. Otherwise, if the conditions for process advancement are not met, an approval timeout reminder containing the reason for the timeout or missing information is generated based on the timeliness detection result and the approval task processing progress. The approval timeout reminder is pushed to the approver's end and the job initiator's end of the current approval node, and the abnormal status of the current node is marked in the approval process progress record.

Images