Elevator Fault Early Warning and Intelligent Dispatch Method and System for Maintenance Manpower

By acquiring elevator operating status and building service demand data, fault risk events and time-segmented service impacts are identified, executable maintenance task candidates are constructed, and joint scheduling schemes are generated. This solves the problem of matching fault handling with service capabilities in the elevator management system, and improves the continuity of elevator services and maintenance efficiency.

CN123070601APending Publication Date: 2026-11-10SHENZHEN MINGZHE PROPERTY MANAGEMENT CO LTD +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611301829.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-26
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing elevator management systems struggle to simultaneously address fault response timelines, alternative elevator service capabilities, and the actual working conditions of maintenance personnel in public buildings, leading to unnecessary elevator shutdowns during peak hours, delays in response, or mismatches between personnel and fault types.

Method used

By acquiring elevator operation status data, building service demand data, and maintenance resource data, a sequence of operation segment characteristics is formed to identify fault risk events and time-segmented service impacts. Candidate executable maintenance tasks are constructed, and a joint scheduling scheme is generated. Taking into account the time boundaries of personnel, resources, and elevators, the maintenance task arrangement is optimized.

Benefits of technology

This approach achieves the goal of minimizing the impact of elevator downtime on building service continuity while meeting fault handling time limits, improving the utilization rate of maintenance manpower and the adaptability of dispatching, and ensuring the continuity and efficiency of elevator services.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123070601A_ABST
    Figure CN123070601A_ABST
Patent Text Reader

Abstract

This application discloses a method and system for elevator fault early warning and intelligent scheduling of maintenance personnel. The method includes: acquiring scheduling input data; forming a sequence of operating segments based on elevator operating status data; comparing the sequence of operating segments with the normal operating baseline to identify fault risk events that include suspected faulty components, risk evolution status, and the latest handling time; determining a time-segmented service impact sequence based on building service demand data, elevator service relationships, elevator operating status data, and fault risk events; constructing executable maintenance task candidates based on fault risk events, time-segmented service impact sequences, and maintenance resource data; generating a joint scheduling scheme based on executable maintenance task candidates, fault risk events, time-segmented service impact sequences, and currently scheduled tasks; and outputting the joint scheduling scheme to the maintenance terminal. This application helps users manage files efficiently, meets user needs, and improves user experience.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the fields of elevator status monitoring, fault prediction and operation and maintenance resource scheduling, and in particular to a method and system for elevator fault early warning and intelligent scheduling of maintenance manpower. Background Technology

[0002] Hospitals, schools, and other public buildings are typically equipped with multiple elevators serving different passage purposes. To ensure the safe operation of elevators, existing elevator management systems can collect data such as vibration, current, door operator status, temperature, and load, and generate alarms when equipment malfunctions; maintenance management systems then dispatch work orders based on alarm levels, staff shifts, or staff locations.

[0003] However, elevator usage demand in public buildings fluctuates significantly with changes in medical appointments, emergency calls, school hours, and passenger flow. The service impact of the same fault varies at different times. Currently, alarm handling and maintenance dispatch are usually independent, making it difficult to simultaneously consider fault response time limits, alternative elevator service capabilities, and the actual working conditions of maintenance personnel. This can easily lead to unnecessary elevator shutdowns during peak hours, delays in response, or mismatches between personnel and fault types. Summary of the Invention

[0004] This application proposes a method and system for early warning of elevator malfunctions and intelligent scheduling of maintenance personnel, aiming to transform elevator malfunction risks, building service demands, and maintenance resource constraints into executable scheduling results that can be updated as operations change.

[0005] In a first aspect, embodiments of this application provide an elevator fault early warning and intelligent scheduling method for maintenance manpower, applied to a maintenance scheduling server. The method includes: acquiring scheduling input data, which includes elevator operating status data, building service demand data, elevator service relationships, maintenance resource data, currently scheduled tasks, and the normal operating baseline of the corresponding elevator; forming an operating segment feature sequence based on the elevator operating status data; comparing the operating segment feature sequence with the normal operating baseline to determine a fault risk event containing a suspected faulty component, a risk evolution state, and a latest handling time; determining a time-segmented service impact sequence based on the building service demand data, the elevator service relationships, the elevator operating status data, and the fault risk event; constructing executable maintenance task candidates based on the fault risk event, the time-segmented service impact sequence, and the maintenance resource data; generating a joint scheduling scheme based on the executable maintenance task candidates, the fault risk event, the time-segmented service impact sequence, and the currently scheduled tasks; and outputting the joint scheduling scheme to a maintenance terminal.

[0006] As can be seen, this embodiment uses the time boundary of fault risk, the time-segmented service impact of the target elevator being taken out of service, and the constraints of personnel and operational resources in the task construction and scheduling. This can convert the equipment warning results into maintenance tasks with clear personnel, time periods, and resources, thereby reducing the impact of elevator shutdown on the continuity of building services while meeting the fault handling time limit.

[0007] In one possible example, generating a joint scheduling scheme based on the executable maintenance task candidates, the fault risk events, the time-segmented service impact sequence, and the currently scheduled tasks includes: forming candidate occupancy relationships based on the maintenance personnel associated with each executable maintenance task candidate, the expected arrival time, the candidate work period, the expected work duration, the work resources, and the target elevator; excluding candidates with personnel conflicts, resource conflicts, or elevator conflicts based on the candidate occupancy relationships and the currently scheduled tasks, thus obtaining candidate tasks to be evaluated; sorting the candidate tasks to be evaluated and determining the target maintenance tasks based on the time-segmented service impact sequence and the latest handling time of the fault risk events; and generating the joint scheduling scheme based on the priority relationship of the target maintenance tasks.

[0008] As can be seen, this example uses candidate occupancy relationships to eliminate execution conflicts between personnel, resources, and target elevators, and then uses the latest handling time and service impact to determine the target maintenance task, so that the generated joint scheduling scheme has executable personnel and time segment boundaries.

[0009] In one possible example, the step of sorting the candidate tasks to be evaluated and determining the target maintenance task based on the time-segmented service impact sequence and the latest handling time of the fault risk event includes: extracting candidate time attributes from the candidate tasks to be evaluated, the candidate time attributes including the expected arrival time, candidate operation period, expected operation duration, and corresponding latest handling time; determining the expected completion time of each candidate task to be evaluated based on the candidate time attributes and the currently scheduled tasks; determining the risk waiting status based on the expected completion time and the corresponding latest handling time; reading the service impact corresponding to the candidate operation period from the time-segmented service impact sequence to form a candidate evaluation record containing the risk waiting status and the service impact; sorting the candidate tasks to be evaluated according to the candidate evaluation record and determining the target maintenance task from them.

[0010] As can be seen, this example measures the risk waiting status by the expected completion time rather than just the dispatch time, and writes the risk waiting status and the service impact of the corresponding time period into the candidate evaluation record together, which can avoid sorting by a single dimension, such as the order of alarms or the low passenger flow period.

[0011] In one possible example, determining the estimated completion time of each candidate task to be evaluated based on the candidate time attribute and the currently scheduled tasks includes: extracting the estimated arrival time, the candidate job time period, and the estimated job duration from the candidate time attribute; determining an available job interval based on the candidate job time period and the end time of tasks in the currently scheduled tasks that have a candidate occupancy relationship with the current candidate task to be evaluated; determining the candidate start time based on the available job interval and the estimated arrival time; and determining the estimated completion time as the end time after the estimated job duration is occupied from the candidate start time.

[0012] As can be seen, this example combines personnel presence, candidate work periods, and the occupancy status of scheduled tasks with the expected completion time, enabling the identification of candidate tasks that appear to be within the candidate time period but are actually unable to complete the work.

[0013] In one possible example, determining the candidate start time based on the available work intervals and the expected arrival time includes: comparing the expected arrival time with the start and end times of each available work interval, and determining the available work intervals whose end time is later than the expected arrival time as arrival adaptation intervals; in the earliest arrival adaptation interval, determining the earliest time that is no earlier than the expected arrival time and no earlier than the start time of the arrival adaptation interval as a candidate start time; verifying whether the candidate start time matches the expected work duration in the arrival adaptation interval, and obtaining a verification result; when the verification result is a match, determining the candidate start time as the candidate start time; when the verification result is not a match, moving to the next arrival adaptation interval to re-determine the candidate start time, until the candidate start time is obtained.

[0014] As can be seen, this example verifies the candidate start time interval by interval in chronological order. It can automatically switch to the next interval when the personnel arrive later than the start time of the interval or when the earliest interval has insufficient remaining time, thus reducing the possibility of an uncompleted time period being selected incorrectly.

[0015] In one possible example, verifying whether the candidate start time and the expected job duration are compatible with the arrival adaptation interval to obtain a verification result includes: determining the candidate end time from the candidate start time according to the expected job duration; comparing the candidate end time with the end time of the current arrival adaptation interval to obtain the interval occupancy relationship; when the interval occupancy relationship indicates that the candidate end time is not later than the end time of the current arrival adaptation interval, the verification result is determined to be compatible; otherwise, the verification result is determined to be incompatible.

[0016] As can be seen, this example completes the time executability verification by the interval occupancy relationship between the candidate end time and the interval end time, so that the verification result can be directly used in the candidate start time determination process.

[0017] In one possible example, determining the time-segmented service impact sequence based on the building service demand data, the elevator service relationships, the elevator operating status data, and the fault risk event includes: dividing the time range from the current time to the latest handling time of the fault risk event into multiple candidate operation periods based on the latest handling time of the fault risk event; determining the target elevator service demand within each candidate operation period based on the building service demand data; determining the alternative service capacity for each candidate operation period based on the elevator service relationships and the elevator operating status data; determining the service gap and service impact level for each candidate operation period based on the target elevator service demand and the alternative service capacity; and combining the service impact levels in chronological order to obtain the time-segmented service impact sequence.

[0018] As can be seen, this example compares the target elevator service demand and alternative service capacity on a time-by-time basis within the time limit defined by the latest handling time, so that the impact of elevator stoppage can change with changes in building service demand and other elevator status.

[0019] In one possible example, determining the alternative service capabilities for each of the candidate work periods based on the elevator service relationship and the elevator operation status data includes: determining candidate alternative elevators from other elevators that overlap with the service floors or service directions of the target elevator based on the elevator service relationship; excluding elevators that are out of service, under maintenance, or at risk of failure from the candidate alternative elevators based on the elevator operation status data to obtain available alternative elevators; forming an alternative capability record for each of the candidate work periods based on the load status, operating frequency, and assigned service tasks of the available alternative elevators; and aggregating the alternative capability records to obtain the alternative service capabilities for each of the candidate work periods.

[0020] As can be seen, this example does not simply identify alternative elevators based on overlapping floors, but further verifies the actual operating status and assigned tasks of candidate alternative elevators, which can improve the authenticity of alternative service capabilities.

[0021] In one possible example, after outputting the joint scheduling scheme to the maintenance terminal, the method further includes: acquiring execution change data, which includes the arrival and operation status, newly added fault risk events, changes in building service demand, and changes in elevator operation status reported back by the maintenance terminal; updating the corresponding fault risk events, time-segmented service impact sequences, and maintenance resource data based on the execution change data to obtain updated scheduling constraints; comparing the updated scheduling constraints with the latest handling time, candidate operation time period, and task occupancy status associated with the joint scheduling scheme to determine the scheme's effective status; when the scheme's effective status is invalid, retaining the executed tasks, and reconstructing executable maintenance task candidates for the unexecuted tasks based on the updated scheduling constraints to generate an updated joint scheduling scheme; and outputting the updated joint scheduling scheme and change notification to the maintenance terminal.

[0022] As can be seen, this example updates only the affected unexecuted tasks when passenger flow, equipment risks, or personnel execution status change, while retaining the executed tasks. This reduces the duplication of tasks and keeps the scheduling results consistent with the on-site status.

[0023] Secondly, embodiments of this application provide an elevator fault early warning and maintenance manpower intelligent scheduling system, comprising: a data acquisition module for acquiring scheduling input data, the scheduling input data including elevator operating status data, building service demand data, elevator service relationships, maintenance resource data, currently scheduled tasks, and the normal operating baseline of the corresponding elevator; an operating segment construction module for forming an operating segment feature sequence based on the elevator operating status data; a risk determination module for comparing the operating segment feature sequence with the normal operating baseline to determine a fault risk event containing a suspected faulty component, a risk evolution state, and a latest handling time; a service impact determination module for determining a time-segmented service impact sequence based on the building service demand data, the elevator service relationships, the elevator operating status data, and the fault risk event; a task construction module for constructing executable maintenance task candidates based on the fault risk event, the time-segmented service impact sequence, and the maintenance resource data; a joint scheduling module for generating a joint scheduling scheme based on the executable maintenance task candidates, the fault risk event, the time-segmented service impact sequence, and the currently scheduled tasks; and a scheme output module for outputting the joint scheduling scheme to a maintenance terminal.

[0024] The above system achieves a data processing closed loop corresponding to the method described in the first aspect by sequentially transmitting scheduling input data, running segment feature sequences, fault risk events, time-segmented service impact sequences, executable maintenance task candidates, and joint scheduling schemes.

[0025] Thirdly, embodiments of this application provide an electronic device including a processor, a memory, a communication interface, and one or more programs, the one or more programs being stored in the memory and configured to be executed by the processor, the programs including instructions for performing the steps in the first aspect of embodiments of this application.

[0026] Fourthly, embodiments of this application provide a computer storage medium storing a computer program for electronic data interchange, wherein the computer program causes a computer to perform some or all of the steps described in the first aspect of this embodiment.

[0027] As can be seen, in this embodiment, compared to the existing method of processing elevator fault alarms and maintenance work orders separately, this application incorporates fault risks, building service impacts, and maintenance resource constraints into the same scheduling link. The maintenance scheduling server first obtains scheduling input data; secondly, it forms a sequence of operating segments based on elevator operating status data; then, it compares the sequence of operating segments with the normal operating baseline to determine fault risk events; next, it determines a time-segmented service impact sequence based on building service demand data, elevator service relationships, elevator operating status data, and fault risk events; subsequently, it constructs executable maintenance task candidates based on fault risk events, time-segmented service impact sequences, and maintenance resource data; furthermore, it generates a joint scheduling scheme based on executable maintenance task candidates, fault risk events, time-segmented service impact sequences, and currently scheduled tasks; finally, it outputs the joint scheduling scheme to the maintenance terminal. Because the time boundary of the fault risk, the time-segmented service impact of the target elevator being taken out of service, and the constraints of maintenance personnel and work resources are all used in the construction and scheduling of candidate tasks, the fault warning results can be directly converted into maintenance tasks with clear personnel, time periods and resources. This reduces unnecessary elevator shutdowns during periods of high service demand while meeting the fault handling time limit, which is conducive to improving elevator service continuity, maintenance manpower utilization and scheduling adaptability under abnormal changes. Attached Figure Description

[0028] Figure 1 A schematic diagram of the hardware architecture of the elevator fault early warning and maintenance manpower intelligent dispatching system provided in the embodiments of this application;

[0029] Figure 2 A schematic diagram of the software architecture of the maintenance scheduling server provided in the embodiments of this application;

[0030] Figure 3 This is a schematic diagram of a comprehensive hospital elevator fault early warning and maintenance scheduling business scenario provided in an embodiment of this application;

[0031] Figure 4 A flowchart illustrating the elevator fault early warning and intelligent scheduling method for maintenance personnel provided in this application embodiment;

[0032] Figure 5 A schematic diagram illustrating the time relationship between equipment risk, service impact, and maintenance resources provided for embodiments of this application;

[0033] Figure 6 A flowchart illustrating the process of generating a joint scheduling scheme as provided in the embodiments of this application;

[0034] Figure 7 This is a schematic diagram of candidate occupancy relationships and conflict filtering provided in an embodiment of this application;

[0035] Figure 8 A flowchart illustrating the process of determining the target maintenance task provided in this application embodiment;

[0036] Figure 9 This is a schematic diagram illustrating the formation of candidate evaluation records provided in an embodiment of this application.

[0037] Figure 10 A flowchart illustrating the process of determining the expected completion time provided in an embodiment of this application;

[0038] Figure 11 A schematic diagram illustrating the relationship between available work intervals and expected completion times provided in embodiments of this application;

[0039] Figure 12 A flowchart illustrating the process of determining the candidate start time provided in an embodiment of this application;

[0040] Figure 13 A flowchart illustrating the verification process for operation duration and on-site adaptation intervals provided in this application embodiment;

[0041] Figure 14 A flowchart illustrating the process of generating sequences affected by time-segmented services provided in this application embodiment;

[0042] Figure 15 A flowchart illustrating the process of determining alternative service capabilities provided in this application embodiment;

[0043] Figure 16 A flowchart illustrating the process of updating the joint scheduling scheme under execution changes, as provided in the embodiments of this application;

[0044] Figure 17 A schematic diagram of the functional modules of the elevator fault early warning and maintenance manpower intelligent dispatching system provided in the embodiments of this application. Detailed Implementation

[0045] The technical solution of this application will be clearly and completely described below with reference to the accompanying drawings and specific embodiments. It should be understood that the described embodiments are only used to explain this application and are not intended to limit the scope of protection of this application. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0046] In this application, "operational segment" refers to a data set corresponding to a complete operation, a door opening / closing action, or other elevator work cycle with comparable working boundaries; "normal operating baseline" refers to a normal state reference saved for elevator model, component, operating direction, load range, or floor range; "latest handling time" refers to the time boundary at which inspection or safety control should be completed under the current risk evolution state, and is not equivalent to the time corresponding to a certain fixed alarm threshold; "time-segmented service impact sequence" refers to the service impact generated by the target elevator's exit from operation within consecutive candidate time periods, recorded in chronological order; "candidate executable maintenance tasks" refers to a combination of tasks that has been associated with the target elevator, maintenance personnel, candidate work periods, and work resources, and has the possibility of completion.

[0047] See Figure 1 The elevator fault early warning and maintenance manpower intelligent dispatch system 100 may include a maintenance dispatch server 110, an elevator status acquisition component 120, a car load acquisition component 130, an edge monitoring gateway 140, a building service demand device 150, an elevator control interface 160, a maintenance terminal 170, and a communication network 180. The maintenance dispatch server 110, the elevator control interface 160, and the maintenance terminal 170 are connected through the communication network 180. The edge monitoring gateway 140 is connected to the elevator status acquisition component 120, the car load acquisition component 130, and the maintenance dispatch server 110, respectively. The building service demand device 150 is connected to the maintenance dispatch server 110.

[0048] The maintenance scheduling server 110 may include a processor 111, a memory 112, a communication interface 113, and a scheduling data memory 114. The processor 111 executes the scheduling program stored in the memory 112; the communication interface 113 receives edge monitoring data, building service demand data, and maintenance terminal feedback; the scheduling data memory 114 stores the normal operating baseline, elevator service relationships, maintenance resource data, currently scheduled tasks, fault risk events, and joint scheduling schemes. The elevator status acquisition component 120 may include a vibration sensor 121, a current acquisition device 122, a door operator status acquisition device 123, a temperature sensor 124, and a controller status interface 125. The car load acquisition component 130 collects the load status or passenger status during the operating cycle.

[0049] The edge monitoring gateway 140 adds a unified clock marker to the data output by the elevator status acquisition component 120 and the car load acquisition component 130, and can form operation segments according to start / stop, floor changes, and door opening / closing events. The building service requirement equipment 150 may include at least one of the following: a hospital business interface 151, a school schedule and access control interface 152, and a floor passenger flow acquisition device 153. The hospital business interface 151 can provide anonymized registration numbers, appointment times, or emergency room operation status; the school schedule and access control interface 152 can provide class times and floor passage numbers; and the floor passenger flow acquisition device 153 can provide waiting area counting results. The elevator control interface 160 provides elevator operation, shutdown, and work preparation status. The maintenance terminal 170 is used to receive tasks and transmit on-site, inspection, trial operation, and completion statuses.

[0050] The aforementioned hardware connection enables the device status evidence generated by the edge monitoring gateway 140, the service requirements generated by the building service demand device 150, and the execution feedback generated by the maintenance terminal 170 to be correlated along the same time axis at the maintenance scheduling server 110. Therefore, the maintenance scheduling server 110 can perform pre-execution verification of the scheduling results using the status provided by the elevator control interface 160 without altering the original safety control loop of the elevator.

[0051] See Figure 2 The maintenance and dispatch server software architecture 200 may include a device access layer 210, a time-series governance layer 220, a state awareness layer 230, a service impact modeling layer 240, a resource constraint layer 250, a joint decision-making layer 260, an execution feedback layer 270, and a data storage layer 280. Each software layer is divided according to its data processing responsibilities and does not form a one-to-one correspondence with the virtual functional modules described later.

[0052] The device access layer 210 adapts to the data protocols of the edge monitoring gateway 140, building service demand device 150, elevator control interface 160, and maintenance terminal 170. The time-series governance layer 220 performs time alignment, device identifier normalization, missing status marking, and trusted status verification on data from different sampling frequencies. The status cognition layer 230 determines suspected faulty components, risk evolution status, and the latest response time based on the characteristic sequence of operational segments and the normal operation baseline.

[0053] The service impact modeling layer 240 maintains the building's purpose, floor service relationships, service direction, and other elevator availability status, forming a time-segmented service impact sequence. The resource constraint layer 250 maintains the maintenance personnel's skills, shifts, locations, work durations, and tool and spare parts status. The joint decision-making layer 260 constructs executable maintenance task candidates, performs conflict screening, candidate evaluation, and task ranking. The execution feedback layer 270 verifies the validity of the trigger scheme based on the feedback from the maintenance terminal 170 and performs rolling updates. The data storage layer 280 stores the data objects and their versions generated by the above layers, enabling the same fault risk event to be associated with the baseline used during generation, service impact, personnel resources, and execution results.

[0054] See Figure 3 In the schematic diagram of the elevator fault early warning and maintenance scheduling business scenario in a general hospital, during the morning outpatient period, the traction machine vibration and current response of medical elevator A (310) continuously deviate from the normal operating baseline, but have not yet reached the condition for immediate shutdown. Outpatient service demand 340 indicates that medical elevator A (310) will undertake concentrated upward traffic tasks in the next forty minutes. Medical elevator B (320) is operating normally and has partially overlapping service floors with medical elevator A (310). Passenger elevator C (330) cannot be used as a usable alternative elevator due to planned maintenance. The maintenance scheduling server 110 thus determines that elevator shutdowns have a higher impact during the morning period, while the low-load midday period 350 has a lower service impact.

[0055] In this scenario, the first maintenance worker 360 possesses traction system repair skills but arrives later than the midday low-load period 350; the second maintenance worker 370 possesses the same skills and carries testing tools, allowing them to arrive before the start of that period. Based on this, the maintenance dispatch server 110 generates a dispatch task 380 to be executed by the second maintenance worker 370. If the risk of elevator A 310 increases rapidly, the maintenance dispatch server 110 can shorten the waiting time and reassign the task. This example illustrates that the same fault risk does not directly correspond to a fixed dispatch result, but rather the work arrangement is determined based on actual service needs and personnel execution conditions.

[0056] In one possible example, see Figure 4 The elevator fault early warning and maintenance manpower intelligent scheduling method can be implemented through the following steps S101 to S107.

[0057] S101. Obtain scheduling input data, which includes elevator operation status data, building service demand data, elevator service relationships, maintenance resource data, currently scheduled tasks, and the normal operation baseline of the corresponding elevators.

[0058] The scheduling input data is a set of higher-level data set set up to avoid transmitting and verifying multiple types of heterogeneous data separately in the first step. Elevator operating status data can include vibration, current, door operator status, temperature, load, floor, direction of travel, and controller status; building service demand data represents the building's demand for elevator access services at different times; elevator service relationships represent the correspondence between elevators and floors, directions, areas, or dedicated tasks; maintenance resource data represents the availability of personnel skills, shifts, locations, tools, and spare parts. The maintenance scheduling server 110 can perform integrity verification on the scheduling input data according to data source, equipment identification, and collection time.

[0059] S102. Form a feature sequence of operating segments based on the elevator operating status data.

[0060] In this embodiment, the maintenance scheduling server 110 or the edge monitoring gateway 140 can divide the operation into segments based on start / stop events, floor changes, running direction, and door opening / closing events, and align data with different sampling frequencies within each segment. Segment characteristics may include vibration changes, drive current changes, door operator operation duration, component temperature rise, load range, and control status, but it is not required that all types be provided for each elevator. Each operation segment is arranged according to the acquisition time to form an operation segment characteristic sequence, so that adjacent working cycles and historical cycles under the same working conditions have comparable data boundaries.

[0061] S103. Compare the operational segment feature sequence with the normal operation baseline to determine the fault risk event that includes suspected faulty components, risk evolution status, and the latest handling time.

[0062] In this embodiment, the normal operating baseline can be stored separately for elevator model, components, operating direction, load range, and floor range. The maintenance scheduling server 110 first selects a baseline that matches the current operating segment's condition, and then compares the changing trends of adjacent segments and the response relationships between multiple sensor channels. Suspected faulty components are used to indicate the traction system, door operator system, or electrical system objects that need to be inspected; the risk evolution status is used to describe whether the risk remains, grows slowly, or accelerates; the latest handling time converts the risk evolution result into a time boundary that can be directly used by subsequent scheduling.

[0063] S104. Based on the building service demand data, the elevator service relationship, the elevator operation status data, and the fault risk event, determine the time-segmented service impact sequence.

[0064] In this embodiment, the maintenance scheduling server 110 sets multiple candidate time periods between the current time and the latest handling time, and determines the service demand of the target elevator in each candidate time period. The maintenance scheduling server 110 further identifies other elevators with overlapping service floors or service directions, and determines the alternative service capabilities after the target elevator is taken out of service based on their operating status. Service impact can characterize the passage gap, the degree of impact on critical tasks, or changes in waiting pressure that occur after the target elevator is out of service. The service impacts of each candidate time period are arranged chronologically to form a time-segmented service impact sequence.

[0065] S105. Construct executable maintenance task candidates based on the fault risk events, the time-segmented service impact sequence, and the maintenance resource data.

[0066] In this embodiment, the maintenance scheduling server 110 maps suspected faulty components to the required skills, tools, and spare parts. It excludes personnel combinations that cannot arrive on time or complete the task based on the latest possible handling time, and then associates the remaining personnel combinations with candidate work periods. Each executable maintenance task candidate can carry the target elevator, maintenance personnel, estimated arrival time, candidate work period, estimated work duration, and work resources. Here, "executable" indicates that the candidate has the potential to meet the basic skill, time, and resource requirements; the final selection is determined by the subsequent joint scheduling process.

[0067] S106. Generate a joint scheduling scheme based on the executable maintenance task candidates, the fault risk events, the time-segmented service impact sequence, and the currently scheduled tasks.

[0068] In this embodiment, the joint scheduling scheme may include one or more target maintenance tasks with a sequential execution relationship. The maintenance scheduling server 110 first eliminates occupancy conflicts between personnel, work resources, and target elevators, then compares the relationship between the expected completion time of the task and the latest handling time, and determines the target maintenance task by combining the service impact of candidate work periods. When multiple elevators pose a risk simultaneously, the joint scheduling scheme can allow the same maintenance personnel to execute multiple tasks sequentially according to the route and task time limit, instead of generating conflicting work orders for each alarm separately.

[0069] S107. Output the joint scheduling scheme to the maintenance terminal.

[0070] In this embodiment, the maintenance terminal 170 can display the target elevator, suspected faulty components, arrival time limit, candidate work periods, required tools and spare parts, and receive arrival, start work, trial operation, and completion operations. The status returned by the maintenance terminal 170 can be associated with the task identifier of the joint scheduling scheme for subsequent scheme validity judgment and scheduling history tracing.

[0071] Through steps S101 to S107, the scheduling input data sequentially forms a sequence of operational segment characteristics, fault risk events, time-segmented service impact sequences, candidate executable maintenance tasks, and a joint scheduling scheme. Intermediate outputs are used in subsequent steps. Thus, the early warning results are no longer limited to risk scores or alarm levels, but can be transformed into specific execution tasks that meet handling time limits and consider the continuity of building services.

[0072] In one possible example, see Figure 5 In the time-related diagram of equipment risk, service impact, and maintenance resources, fault risk event 520 is determined based on the operational segment characteristic sequence 510, and the candidate time range 530 is limited by the latest handling time. The time-segmented service impact sequence 540 is formed within the candidate time range 530. The maintenance resource status 550 and the time-segmented service impact sequence 540 jointly constrain the executable maintenance task candidates 560, and the joint scheduling scheme 570 is determined from the executable maintenance task candidates 560. This correlation method ensures that the latest handling time limits both the evaluation scope of service impact and the personnel arrival and task completion times, preventing a disconnect between the equipment risk timeline and the scheduling timeline.

[0073] In one possible example, see Figure 6 The process of generating a joint scheduling scheme in step S106 can be achieved through the following steps S201 to S204.

[0074] S201. Based on the maintenance personnel, expected arrival time, candidate operation period, expected operation duration, operation resources, and target elevator associated with each of the executable maintenance task candidates, a candidate occupancy relationship is formed.

[0075] The candidate occupancy relationship represents the occupancy of personnel, tools, spare parts, and the target elevator during the planned execution of an executable maintenance task. The candidate occupancy relationship can include the occupancy object identifier, the start and end times of the occupancy, and the occupancy type. The estimated arrival time represents the predicted time when maintenance personnel will arrive at the target elevator based on their current location, travel path, and current task status; the estimated operation duration can be determined based on the suspected faulty component, the planned handling type, and historical similar tasks.

[0076] S202. Based on the candidate occupancy relationship and the currently scheduled tasks, eliminate candidates with personnel conflicts, resource conflicts, or elevator conflicts to obtain candidate tasks to be evaluated.

[0077] In this embodiment, personnel conflict can refer to two tasks performed by the same maintenance personnel overlapping in time, or the personnel being unable to complete the subsequent task on time after the previous task is finished; resource conflict can refer to multiple tasks simultaneously occupying the same specialized tool or the same spare parts inventory; elevator conflict can refer to the presence of authorized or uninterrupted operating tasks on the target elevator during the candidate operation period. The candidate tasks to be evaluated are those selected through the above basic executability screening and are yet to be ranked based on risk and service impact.

[0078] See one example of candidate filtering. Figure 7 In the candidate occupancy relationship and conflict screening diagram, the first task candidate 710 and the currently scheduled task 730 use the same maintenance personnel and their times overlap, resulting in a personnel conflict 740; the second task candidate 720 no longer causes resource conflicts 750 or elevator conflicts 760 after adjusting the tool collection time, and therefore can enter the candidate task 770 for evaluation. This example illustrates that conflict judgment is based on the specific occupancy object and time period, rather than simply excluding tasks based on the number of tasks.

[0079] S203. Based on the time-segmented service impact sequence and the latest handling time of the fault risk event, sort the candidate tasks to be evaluated and determine the target maintenance task.

[0080] In this embodiment, the maintenance scheduling server 110 can determine the estimated completion time, risk waiting status, and service impact corresponding to the candidate operation period for each candidate task to be evaluated. When the estimated completion time is close to the latest handling time, the urgency of handling the task can be increased; when the service impact is high, other candidate operation periods can be selected while still meeting the latest handling time requirement. The target maintenance task is the task selected as the planned execution object after the above sorting.

[0081] In one possible example, to address the issue of frequent switching of target maintenance tasks due to small fluctuations in building service demand, scheduling stability conditions can be set for the current target maintenance task. When the difference in service impact between a new candidate and the current target maintenance task is within a preset stable range, and the current target maintenance task still meets the latest handling time requirement, the maintenance scheduling server 110 retains the current target maintenance task; switching is only performed when the difference exceeds the stable range or the original task becomes unfeasible. This extended concept can reduce scheduling oscillations caused by short-term passenger flow noise without relaxing fault handling time limits.

[0082] S204. Generate the joint scheduling scheme based on the sequence of the target maintenance tasks.

[0083] In this embodiment, the sequence of target maintenance tasks can be determined based on the latest task handling time, personnel movement order, work resource handover order, and target elevator service hours. The joint scheduling scheme records at least the task identifier, target elevator, maintenance personnel, execution time, and work resources; when multiple tasks exist, the completion status of the previous task can also be recorded as the starting condition for the next task.

[0084] Through steps S201 to S204, candidate occupancy relationships are used to screen out basic conflicts. The selected candidate tasks to be evaluated continue to participate in risk and service impact ranking, and the ranking results further form a joint scheduling scheme. This reduces the number of conflicting work orders entering the execution end and ensures that tasks corresponding to multiple elevators maintain a clear execution order under the same personnel and resource constraints.

[0085] In one possible example, see Figure 8 The process of determining the target maintenance task in step S203 can be achieved through the following steps S301 to S305.

[0086] S301. Extract candidate time attributes from the candidate tasks to be evaluated. The candidate time attributes include the expected arrival time, the candidate operation period, the expected operation duration, and the corresponding latest disposal time.

[0087] Among them, the candidate time attribute is a set of attributes used to describe the time feasibility of the candidate task to be evaluated. The expected arrival time can be determined by the current location of the maintenance personnel, the expected end time of the current task, and the travel time to the target elevator; the candidate operation period can be drawn from consecutive periods in the time-segmented service impact sequence; the expected operation duration can be determined by the historical operation records corresponding to the suspected faulty parts and the handling type; the latest handling time comes from the corresponding fault risk event.

[0088] S302. Determine the estimated completion time of each of the candidate tasks to be evaluated based on the candidate time attributes and the currently scheduled tasks.

[0089] In this embodiment, the maintenance scheduling server 110 first identifies the intervals in the candidate work time slots that are not currently occupied by scheduled tasks. Then, it determines the earliest time when the work can begin based on the expected arrival time, and combines this with the expected work duration to determine the expected completion time. The expected completion time represents the time when the candidate task to be evaluated is expected to complete inspection or safety handling under the combined constraints of personnel, time slot, and scheduled tasks.

[0090] S303. Determine the risk waiting status based on the expected completion time and the corresponding latest disposal time.

[0091] In this embodiment, the risk waiting state is used to characterize the remaining margin of the task completion time relative to the risk time boundary, and may include states such as sufficient margin, tight margin, or exceeding the boundary. The risk waiting state is not another name for the equipment risk level, but rather a scheduling state formed by associating the latest handling time corresponding to the equipment risk with the expected completion time of the specific task.

[0092] S304. Read the service impact corresponding to the candidate operation time period from the time-segmented service impact sequence to form a candidate evaluation record containing the risk waiting status and the service impact.

[0093] In this embodiment, the candidate evaluation record may include a task identifier, risk waiting status, service impact level, personnel movement duration, and candidate operation period. The risk waiting status is used to constrain the task from being indefinitely postponed in order to reduce service impact, while the service impact is used to select the candidate with the least disturbance to building operation among multiple candidates that all meet the risk time boundary.

[0094] In a time evaluation example, see Figure 9 The candidate evaluation record formation diagram includes the current time 910, estimated arrival time 920, candidate operation period 930, estimated completion time 940, latest handling time 950, risk waiting status 960, service impact 970, and candidate evaluation record 980. After the estimated arrival time 920 falls within the candidate operation period 930, the maintenance scheduling server 110 combines the estimated operation duration to obtain the estimated completion time 940, and compares it with the latest handling time 950 to obtain the risk waiting status 960. The service impact 970 corresponding to the candidate operation period 930 and the risk waiting status 960 are jointly written into the candidate evaluation record 980.

[0095] In one possible example, to address the issue of service impact levels fluctuating near boundaries between adjacent candidate operation periods, a service impact credibility status can be added to the candidate evaluation records. The service impact credibility status can be determined based on data integrity, consistency of changes over consecutive time periods, and the stability of alternative elevator statuses. When credibility status is insufficient, the maintenance scheduling server 110 can retain candidates with better risk waiting statuses or request new passenger flow and elevator status data before re-sorting. This extended concept avoids low-credibility service impacts covering clearly defined fault handling timelines.

[0096] S305. Sort the candidate tasks to be evaluated according to the candidate evaluation records, and determine the target maintenance task from them.

[0097] In this embodiment, the sorting process first excludes candidates whose risk waiting status exceeds the boundary, and then compares the service impact, personnel movement time, and resource consumption among the remaining candidates. The sorting rules can be configured according to hospitals, schools, or other building types, but any configuration must not exceed the latest handling time. The candidate ranked first or meeting the preset selection criteria is determined as the target maintenance task.

[0098] Through steps S301 to S305, the candidate time attribute first forms the expected completion time, which then forms the risk waiting status. The risk waiting status and service impact together form the candidate evaluation record, which is then used to determine the target maintenance task. This continuous processing relationship enables the timely assessment of risk and the impact of building services to be traceably evaluated on the same candidate object.

[0099] In one possible example, see Figure 10 The process of determining the expected completion time in step S302 can be achieved through the following steps S401 to S404.

[0100] S401. Extract the expected arrival time, the candidate operation period, and the expected operation duration from the candidate time attributes.

[0101] The estimated arrival time, candidate work period, and estimated work duration describe the personnel accessibility boundary, the building's permitted work boundary, and the task's continuous occupancy boundary, respectively. The maintenance scheduling server 110 can extract these three attributes based on the task identifier of the same candidate task to be evaluated, avoiding the cross-use of time attributes of different candidate tasks.

[0102] S402. Determine the available job interval based on the candidate job time period and the end time of the currently scheduled tasks that have a candidate occupancy relationship with the current candidate task to be evaluated.

[0103] In this embodiment, the available work interval is a continuous time interval formed by deducting scheduled tasks that have personnel, resource, or target elevator occupancy relationships with the current candidate task from the candidate work time period. If a candidate work time period is divided by scheduled tasks, multiple available work intervals can be formed. Scheduled tasks that do not have candidate occupancy relationships with the current candidate are not included in the interval deduction to avoid irrelevant tasks excessively compressing the available time.

[0104] In one possible example, an execution buffer can be set for available work intervals based on historical arrival deviations of maintenance personnel and historical work duration deviations for similar tasks. The execution buffer is reserved from the end of the interval and is used only to absorb uncertainties in arrival or work duration; it is not treated as a new available time slot. The specific problem it addresses is that while the statically estimated work duration may fall within the interval, slight on-site delays may cause the work to span periods of high service demand. By setting an execution buffer, the consistency between the estimated completion time and the actual completion time can be improved.

[0105] S403. Determine the candidate start time based on the available work interval and the expected arrival time.

[0106] In this embodiment, the maintenance scheduling server 110 checks the available work intervals in chronological order. When the expected arrival time is earlier than the start time of the interval, the start time of the interval is used as the candidate start time; when the expected arrival time is within the interval, the expected arrival time is used as the candidate start time; when the expected arrival time is later than the end time of the interval, the system switches to the next available work interval.

[0107] In one time interval example, see Figure 11 The diagram illustrating the relationship between available work intervals and estimated completion times includes candidate work time slots 930, currently scheduled tasks 730, available work intervals 1130, estimated arrival time 920, candidate start time 1150, estimated work duration 1160, and estimated completion time 940. Currently scheduled tasks 730 divide candidate work time slots 930 into two available work intervals 1130; the estimated arrival time 920 falls into the latter available work interval 1130, and the candidate start time 1150 is determined accordingly. The estimated work duration 1160 continuously occupies the time slot from the candidate start time 1150 and forms the estimated completion time 940.

[0108] S404. The estimated completion time is determined as the end time after the estimated operation duration has been occupied from the candidate start time.

[0109] In this embodiment, the estimated operation duration may include the continuous time required for safety preparation, fault inspection, component handling, and trial operation. If the operation rules require that a certain step cannot cross the candidate operation time period boundary, the maintenance scheduling server 110 can treat the duration corresponding to that step as an indivisible occupied unit. The estimated completion time obtained in this way can be directly compared with the latest handling time.

[0110] Through steps S401 to S404, each time feature in the candidate time attribute is used to determine the available work interval, candidate start time, and expected completion time, avoiding the direct use of the expected arrival time as the basis for task completion. This improves the accuracy of judging the time feasibility of the candidate task to be evaluated.

[0111] In one possible example, see Figure 12 The process of determining the candidate start time in step S403 can be implemented through the following steps S501 to S504.

[0112] S501. Compare the estimated arrival time with the start and end times of each available work interval, and determine the available work interval whose end time is later than the estimated arrival time as the arrival adaptation interval.

[0113] The arrival adaptation interval refers to the available work interval that retains usable time after the maintenance personnel are expected to arrive. This definition only completes the initial adaptation between the arrival boundary and the interval boundary, and does not mean that the expected work duration will necessarily fall entirely within this interval. Available work intervals with an end time no later than the expected arrival time are excluded to avoid using past periods before personnel have arrived for task scheduling.

[0114] S502. In the earliest arrival adaptation interval, the earliest time that is no earlier than the expected arrival time and no earlier than the start time of the arrival adaptation interval is determined as the candidate start time.

[0115] In this embodiment, the candidate start time is constrained by both the expected arrival time and the start time of the arrival adaptation interval. When maintenance personnel arrive early, the start time of the interval is used; when maintenance personnel arrive after the start of the interval, the expected arrival time is used. Selecting the earliest arrival adaptation interval helps to retain more leeway for subsequent processing without violating service time constraints.

[0116] S503. Verify whether the candidate start time and the expected operation duration are compatible with the arrival adaptation interval, and obtain the verification result.

[0117] In this embodiment, the maintenance scheduling server 110 continuously places the estimated job duration starting from the candidate start time and checks whether the resulting candidate end time exceeds the end time of the current arrival adaptation interval. The verification result can include adaptation and non-adaptation. This result is directly used in subsequent steps to confirm the current candidate or move to the next time period.

[0118] In one possible example, when the anticipated work process includes a preparation phase that allows for interruptions and a maintenance phase that does not allow for interruptions, interval adaptation conditions can be set separately for each type of phase. The preparation phase can be scheduled in the earlier arrival adaptation interval, while the maintenance and trial operation phases need to fall continuously within the same arrival adaptation interval. This extended concept addresses the problem of insufficient available intervals when preparation work is bundled with core maintenance, while avoiding safety risks caused by interruptions across time periods by maintaining the continuity of core maintenance.

[0119] S504. When the verification result is a match, the candidate start time is determined as the candidate start time; when the verification result is not a match, the process moves to the next arrival match interval to re-determine the candidate start time until the candidate start time is obtained.

[0120] In this embodiment, when switching to the next on-site adaptation interval, the same estimated arrival time and estimated operation duration are used, and the candidate start time determination and interval adaptation verification are re-executed. Since the executable maintenance task candidates have already passed the basic time executability screening, the candidate start time can usually be obtained from the candidate interval sequence; if changes in the field cause all intervals to be unsuitable, the current candidate can be marked as invalid and the process can be returned to the upper-level scheduling process to reconstruct the candidate.

[0121] Through steps S501 to S504, the estimated arrival time is first used to filter the arrival matching intervals. The earliest arrival matching interval is then used to form candidate start times. The verification results then control the confirmation of candidates or interval transfer. This avoids overestimating the remaining operation time due to personnel arriving in the middle of the candidate time period.

[0122] In one possible example, see Figure 13 The interval adaptation verification in step S503 can be achieved through the following steps S601 to S603.

[0123] S601. Starting from the candidate start time, determine the candidate end time according to the expected operation duration.

[0124] The candidate end time is the termination boundary of the estimated operation duration on the timeline after continuous occupancy from the candidate start time. The estimated operation duration can be derived from historical completion records of tasks involving the same model of component, similar fault, and the same level of handling, or it can be confirmed by maintenance management personnel before the task is generated. For tasks requiring trial operation, the estimated operation duration includes the time required for the trial operation.

[0125] S602. Compare the candidate end time with the end time of the current arrival adaptation interval to obtain the interval occupancy relationship.

[0126] In this embodiment, the interval occupancy relationship is used to describe whether the candidate job process is completely contained within the current arrival adaptation interval. A unified clock and the same time granularity are used for comparison, and a preset execution buffer can be incorporated into the candidate's end time to avoid boundary misjudgments caused by different time formats or rounding.

[0127] S603. When the interval occupancy relationship indicates that the candidate end time is not later than the end time of the current arrival adaptation interval, the verification result is determined to be an adaptation; otherwise, the verification result is determined to be an incompatibility.

[0128] In this embodiment, a fit state indicates that the current interval can accommodate the expected operation process, while an unfit state indicates that the expected operation process will cross the end boundary of the current interval. The verification result only evaluates the time interval fit and does not replace the verification of personnel skills, tools, spare parts, or elevator safety status, thereby maintaining clear processing boundaries for different constraints.

[0129] Through steps S601 to S603, the candidate start time and the expected operation duration are first used to form the candidate end time. The candidate end time is then used to form the interval occupancy relationship, and the interval occupancy relationship finally forms the verification result. This process can provide a clear state for the higher-level interval transfer without recalculating candidate attributes that are unrelated to the current interval.

[0130] In one possible example, see Figure 14 The process of determining the time-segmented service impact sequence in step S104 can be achieved through the following steps S701 to S705.

[0131] S701. Based on the latest handling time of the fault risk event, divide the time range from the current time to the latest handling time into multiple candidate operation periods.

[0132] The candidate operation period is a continuous time interval used to evaluate the impact of elevator outages and to undertake maintenance tasks. Candidate operation periods can be divided according to a fixed time granularity or according to business event boundaries. For example, in a hospital scenario, they can be divided according to events such as the start of outpatient services, concentrated examinations, low load at lunchtime, and emergency room activation; in a school scenario, they can be divided according to events such as classes, breaks, and school dismissal. The latest handling time is used to define the latest boundary of the candidate operation period, preventing service impact assessments from postponing tasks beyond the acceptable risk range.

[0133] S702. Determine the target elevator service demand for each of the candidate operation periods based on the building service demand data.

[0134] In this embodiment, the target elevator service demand represents the expected traffic volume of the target elevator during the candidate operation period. The maintenance scheduling server 110 can determine the target elevator service demand based on historical passenger flow, reservation or timetable data, real-time elevator count, floor destination distribution, and dedicated task status. For tasks such as hospital emergency transport and surgical transport, service task identifiers different from those for ordinary passenger flow can be set to avoid measuring service demand solely based on the number of people.

[0135] S703. Determine the alternative service capabilities for each of the candidate operation periods based on the elevator service relationship and the elevator operation status data.

[0136] In this embodiment, the alternative service capacity refers to the amount and type of service that other available elevators can undertake on the same floor, in the same direction, or within a dedicated task range after the target elevator is taken out of service. The maintenance scheduling server 110 first determines candidate alternative elevators based on the elevator service relationship, and then determines the actual available capacity by combining the operation, load, and task status of the candidate alternative elevators during the corresponding candidate operation period.

[0137] S704. Determine the service gap and service impact level for each of the candidate operation periods based on the target elevator service demand and the alternative service capabilities.

[0138] In this embodiment, a service gap represents service demand that has not yet been met by other elevators after the target elevator has ceased operation. This gap can be characterized by a shortage in the number of passageways, a shortage in floor coverage, a shortage in service direction, or a shortage in dedicated tasks. The service impact level can be determined based on the scope, duration, and type of affected tasks of the service gap. For the same number of gaps, candidate operation periods involving emergency access or accessibility services can correspond to a higher service impact level.

[0139] In one possible example, to address the issue of building service demands originating from multiple devices and updating at varying frequencies, a demand evidence status can be maintained for each candidate work period. This demand evidence status records the data source, update time, coverage, and reliability. When real-time passenger flow is missing, it is supplemented using business schedules and historical data from similar periods, and the supplementary status is written into the service impact level evidence record. This extended concept allows service impact results to be traced back to the actual data source and avoids distortion of service impact caused by a single data collection device being offline.

[0140] S705. Combine the service impact levels according to the time order to obtain the time-segmented service impact sequence.

[0141] In this embodiment, each sequence item in the time-segmented service impact sequence is associated with at least the candidate operation time period, the target elevator service demand, the alternative service capacity, the service gap, and the service impact level. Adjacent candidate operation time periods with the same service impact level can be merged and displayed, but the original candidate operation time period boundaries are still retained during data storage for partial replacement during scheduling updates.

[0142] Through steps S701 to S705, the latest handling time limits the evaluation scope. Candidate operation periods respectively form the target elevator service demand and alternative service capabilities. These two factors further form the service gap and service impact level, ultimately resulting in a time-segmented service impact sequence. Therefore, operation periods with minimal disruption to building services can be identified without exceeding the fault handling time limit.

[0143] In one possible example, see Figure 15The process of determining alternative service capabilities in step S703 can be implemented through the following steps S801 to S804.

[0144] S801. Based on the elevator service relationship, determine candidate alternative elevators from other elevators that overlap with the service floor or service direction of the target elevator.

[0145] The elevator service relationship can include the elevator's permitted floors, operating zones, service direction, passenger or freight transport attributes, medical elevator attributes, fire or emergency use, and time-based control rules. Candidate alternative elevators are other elevators that are spatially or functionally capable of fulfilling the assigned task. Elevators located within the same building but without overlapping service floors or not permitted to perform the corresponding task are not included in the candidate alternative elevator set.

[0146] S802. Based on the elevator operating status data, elevators that are out of service, under maintenance, or at risk of failure are excluded from the candidate alternative elevators to obtain available alternative elevators.

[0147] In this embodiment, the shutdown status can include safe shutdown, planned shutdown, and control disablement; the maintenance status can include work preparation, inspection, and trial operation; the fault risk status can include a state where the risk has reached a point where it is not advisable to increase the load. Available alternative elevators refer to candidate alternative elevators that have basic operational qualifications during the current candidate work period; whether they have sufficient capacity to handle the load is still determined by subsequent steps.

[0148] S803. Based on the load status, operating frequency, and assigned service tasks of the available alternative elevators, a record of alternative capabilities for each candidate work period is generated.

[0149] In this embodiment, the alternative capacity record describes the alternative services that a single available alternative elevator can provide within a candidate operating period. Load status reflects the remaining carrying capacity, operating frequency reflects the number of services that can be completed per unit time period, and assigned service tasks are used to deduct existing services that cannot be transferred. The alternative capacity record can also carry the service floor, service direction, and available task type to prevent issues where local floors are not covered by aggregation based solely on total load.

[0150] S804. Aggregate the alternative capability records to obtain the alternative service capabilities for each candidate job time period.

[0151] In this embodiment, during aggregation, alternative capability records can be summarized separately according to service floor, service direction, or task type, and then compared with the target elevator service requirements using the same service dimension. If multiple available alternative elevators share the same elevator lobby or control group, the maintenance scheduling server 110 can also eliminate duplicate service capabilities based on the group control status.

[0152] Through steps S801 to S804, candidate alternative elevators are first formed based on spatial or task overlap. The actual operating status then forms available alternative elevators. Load, frequency, and assigned tasks further generate alternative capacity records, ultimately resulting in time-segmented alternative service capacity. This scheme avoids including elevators that nominally reach the same floors but are actually out of service, fully loaded, or performing other tasks in the alternative capacity count.

[0153] In one possible example, see Figure 16 The joint scheduling scheme update process after step S107 can be implemented through the following steps S901 to S905.

[0154] S901. Obtain execution change data, which includes the on-site and operation status, new fault risk events, changes in building service requirements, and changes in elevator operation status transmitted back by the maintenance terminal.

[0155] Among them, execution change data is a set of data that may change the executability or order of tasks after the joint scheduling scheme is output. Arrival and operation status can include early arrival, late arrival, start of maintenance, extended operation, trial operation, and completion; new fault risk events can come from the target elevator or other elevators; changes in building service demand can come from temporary emergency care, event arrangements, or sudden surges in passenger flow; elevator operation status changes can include shutdown, resumption of operation, or changes in the load of replacement elevators.

[0156] S902. Update the corresponding fault risk events, time-segmented service impact sequences, and maintenance resource data based on the execution change data to obtain the updated scheduling constraints.

[0157] In this embodiment, the updated scheduling constraints represent the risk time boundaries, service period boundaries, and personnel and operational resource boundaries that still need to be satisfied after the execution change occurs. The maintenance scheduling server 110 locates the affected data objects according to the elevator identifier, task identifier, and occurrence time associated with the execution change data, and only updates the fault risk events, sequence items, or resource statuses related to the change.

[0158] S903. Compare the updated scheduling constraints with the latest processing time, candidate job time period and task occupancy status associated with the joint scheduling scheme to determine the effective status of the scheme.

[0159] In this embodiment, the task occupancy status represents the current occupancy of personnel, resources, and target elevators by tasks in the joint scheduling scheme. The scheme's validity status can include valid and invalid; it is determined to be valid when all unexecuted tasks still meet the updated risk time boundaries, service period boundaries, and resource boundaries, and it is determined to be invalid when at least one unexecuted task fails to meet the above boundaries. Completed tasks are not re-determined to be invalid due to subsequent changes.

[0160] S904. When the effective state of the scheme is invalid, the executed tasks are retained, and the unexecuted tasks are reconstructed into executable maintenance task candidates based on the updated scheduling constraints to generate an updated joint scheduling scheme.

[0161] In this embodiment, executed tasks can include completed tasks or tasks that are not allowed to be reassigned midway according to safety rules. The maintenance scheduling server 110 fixes the actual occupancy and completion results of executed tasks, and only reconstructs and sorts the candidates for affected unexecuted tasks. If the latest handling time of a new fault risk event is earlier than the original task, the new task can be inserted before the original task; if there is a sudden increase in the demand for building services, tasks with remaining risk margin can be moved to a later low-impact period.

[0162] S905. Output the updated joint scheduling scheme and change notification to the maintenance terminal.

[0163] In this embodiment, the change notification may include the adjusted task, the reason for the adjustment, the new on-site requirements, and the original version of the task that has been cancelled. After the maintenance terminal 170 confirms receipt, the maintenance scheduling server 110 associates the confirmation result with the updated joint scheduling scheme to prevent the same person from executing both the old and new versions of the task simultaneously.

[0164] Through steps S901 to S905, the changed data is first transformed into updated scheduling constraints. The updated scheduling constraints then form the effective state of the plan, while the ineffective state triggers the refactoring of unexecuted tasks and forms an updated joint scheduling plan. This allows for responses to changes in risks, passenger flow, and resources while preserving the results of on-site execution, improving the continuity and traceability of the scheduling plan.

[0165] In one possible example, see Figure 17 The elevator fault early warning and maintenance manpower intelligent dispatch system 100 may include a data acquisition module 1710, an operation segment construction module 1720, a risk determination module 1730, a service impact determination module 1740, a task construction module 1750, a joint dispatch module 1760, and a solution output module 1770.

[0166] The data acquisition module 1710 is used to acquire scheduling input data, which includes elevator operation status data, building service demand data, elevator service relationships, maintenance resource data, currently scheduled tasks, and the normal operation baseline of the corresponding elevators.

[0167] The data acquisition module 1710 can call the device access layer 210 and the data storage layer 280 to receive data from field devices and read data stored on the server, respectively. The data acquisition module 1710 can attach data source, device identifier, task identifier, acquisition time and trust status to various types of data, and provide the verified scheduling input data to the running segment construction module 1720, service impact determination module 1740, task construction module 1750 and joint scheduling module 1760.

[0168] The running segment construction module 1720 is used to form a running segment feature sequence based on the elevator running status data.

[0169] In this embodiment, the operation segment construction module 1720 can determine the boundary of the operation segment based on start / stop, operation direction, floor changes, and door opening / closing events, and perform time alignment and feature extraction on the vibration, current, door operator, temperature, and load data within the segment. The operation segment construction module 1720 provides the formed operation segment feature sequence to the risk determination module 1730.

[0170] The risk determination module 1730 is used to compare the feature sequence of the running segment with the normal operating baseline to determine the fault risk event, which includes the suspected faulty component, the risk evolution state, and the latest handling time.

[0171] In this embodiment, the risk determination module 1730 first selects a normal operating baseline based on the operating conditions, and then determines fault risk events based on changes in adjacent segments and the response relationship of multiple sensor channels. The risk determination module 1730 provides the fault risk events to the service impact determination module 1740, the task construction module 1750, and the joint scheduling module 1760.

[0172] The service impact determination module 1740 is used to determine the time-segmented service impact sequence based on the building service demand data, the elevator service relationship, the elevator operation status data, and the fault risk event.

[0173] In this embodiment, the service impact determination module 1740 determines candidate operation periods within the time frame defined by the fault risk event, calculates the target elevator service demand and alternative service capabilities, and forms a time-segmented service impact sequence. The service impact determination module 1740 provides the time-segmented service impact sequence to the task construction module 1750 and the joint scheduling module 1760.

[0174] The task construction module 1750 is used to construct executable maintenance task candidates based on the fault risk event, the time-segmented service impact sequence, and the maintenance resource data.

[0175] In this embodiment, the task construction module 1750 maps suspected faulty components to personnel skills, tools, and spare parts requirements, eliminates combinations that cannot be completed before the latest handling time, and associates the remaining combinations with candidate task time periods. The task construction module 1750 provides executable maintenance task candidates to the joint scheduling module 1760.

[0176] The joint scheduling module 1760 is used to generate a joint scheduling scheme based on the executable maintenance task candidates, the fault risk events, the time-segmented service impact sequence, and the currently scheduled tasks.

[0177] In this embodiment, the joint scheduling module 1760 can perform candidate occupancy relationship formation, conflict screening, estimated completion time determination, risk waiting status determination, candidate evaluation, and task sorting, and provide the generated joint scheduling scheme to the scheme output module 1770. The joint scheduling module 1760 can also receive execution change data after the scheme is output to realize the rolling update of unexecuted tasks.

[0178] The scheme output module 1770 is used to output the joint scheduling scheme to the maintenance terminal.

[0179] In this embodiment, the scheme output module 1770 can generate a terminal task view according to the maintenance personnel's identity and task permissions, and associate the receipt confirmation returned by the maintenance terminal 170 with the joint scheduling scheme version. The scheme output module 1770 can also send work preparation information to the elevator control interface 160, but the elevator start-up, shutdown and safety control are still executed by the original elevator safety control system according to its own rules.

[0180] The data acquisition module 1710, runtime fragment construction module 1720, risk determination module 1730, service impact determination module 1740, task construction module 1750, joint scheduling module 1760, and solution output module 1770 can be implemented by software, hardware circuits, or a combination of software and hardware. Each module can be deployed in the same maintenance scheduling server 110, or part of the processing of the runtime fragment construction module 1720 can be deployed on the edge monitoring gateway 140, as long as the aforementioned data input, processing, and output relationships are maintained between the modules.

[0181] Through the aforementioned system, the scheduling input data sequentially forms a sequence of operational segment characteristics, fault risk events, time-segmented service impact sequences, executable maintenance task candidates, and joint scheduling schemes. Each intermediate output is then reused by subsequent modules. Thus, the system can realize the complete processing of the elevator fault early warning and intelligent maintenance manpower scheduling method, and retain records of the correlation between equipment risks, service impacts, task decisions, and execution feedback.

[0182] The technical features described above can be used in combination without contradicting each other. Those skilled in the art can make adaptive adjustments to the data source, time granularity, device deployment method, or task display method without departing from the technical concept of this application, and such adjustments should still fall within the protection scope defined by this application.

Claims

1. A method for early warning of elevator malfunctions and intelligent scheduling of maintenance personnel, characterized in that, Applied to a maintenance scheduling server, the method includes: Acquire scheduling input data, which includes elevator operation status data, building service demand data, elevator service relationships, maintenance resource data, currently scheduled tasks, and the normal operation baseline of the corresponding elevators; A sequence of operating segments is formed based on the elevator operating status data; The operational segment feature sequence is compared with the normal operation baseline to identify fault risk events that include suspected faulty components, risk evolution status, and the latest handling time. Based on the building service demand data, the elevator service relationships, the elevator operation status data, and the fault risk events, a time-segmented service impact sequence is determined; Based on the fault risk events, the time-segmented service impact sequence, and the maintenance resource data, an executable maintenance task candidate is constructed. A joint scheduling scheme is generated based on the executable maintenance task candidates, the fault risk events, the time-segmented service impact sequence, and the currently scheduled tasks. The joint scheduling scheme is output to the maintenance terminal.

2. The method according to claim 1, characterized in that, The step of generating a joint scheduling scheme based on the executable maintenance task candidates, the fault risk events, the time-segmented service impact sequence, and the currently scheduled tasks includes: Based on the maintenance personnel, expected arrival time, candidate operation period, expected operation duration, operation resources, and target elevator associated with each of the candidate executable maintenance tasks, a candidate occupancy relationship is formed; Based on the candidate occupancy relationship and the currently scheduled tasks, candidates with personnel conflicts, resource conflicts, or elevator conflicts are excluded, and candidate tasks to be evaluated are obtained. Based on the time-segmented service impact sequence and the latest handling time of the fault risk event, the candidate tasks to be evaluated are sorted and the target maintenance task is determined; The joint scheduling scheme is generated based on the sequence of the target maintenance tasks.

3. The method according to claim 2, characterized in that, The process of ranking the candidate tasks to be evaluated and determining the target maintenance task based on the time-segmented service impact sequence and the latest handling time of the fault risk event includes: Extract candidate time attributes from the candidate tasks to be evaluated. The candidate time attributes include the expected arrival time, the candidate operation period, the expected operation duration, and the corresponding latest disposal time. The estimated completion time of each candidate task to be evaluated is determined based on the candidate time attribute and the currently scheduled tasks. The risk waiting status is determined based on the expected completion time and the corresponding latest disposal time. Read the service impact corresponding to the candidate operation time period from the time-segmented service impact sequence to form a candidate evaluation record containing the risk waiting status and the service impact; The candidate tasks to be evaluated are sorted according to the candidate evaluation records, and the target maintenance task is determined from them.

4. The method according to claim 3, characterized in that, The step of determining the estimated completion time of each candidate task to be evaluated based on the candidate time attribute and the currently scheduled tasks includes: Extract the expected arrival time, the candidate operation period, and the expected operation duration from the candidate time attributes; The available job interval is determined based on the candidate job time period and the end time of the currently scheduled tasks that have a candidate occupancy relationship with the current candidate task to be evaluated. The candidate start time is determined based on the available work interval and the expected arrival time; The estimated completion time is determined by the end time after the estimated operation duration has been occupied from the candidate start time.

5. The method according to claim 4, characterized in that, The step of determining the candidate start time based on the available work interval and the expected arrival time includes: The estimated arrival time is compared with the start and end times of each available work interval, and the available work interval with the end time later than the estimated arrival time is determined as the arrival adaptation interval. Within the earliest arrival adaptation interval, the earliest time that is no earlier than the expected arrival time and no earlier than the start time of the arrival adaptation interval is determined as the candidate start time. Verify whether the candidate start time matches the expected operation duration within the arrival adaptation interval to obtain the verification result; When the verification result is a match, the candidate start time is determined as the candidate start time; when the verification result is an unmatch, the process moves to the next matching interval to re-determine the candidate start time until the candidate start time is obtained.

6. The method according to claim 5, characterized in that, The verification process, which checks whether the candidate start time matches the expected operation duration within the arrival adaptation interval, yields a verification result, including: Starting from the candidate start time, the candidate end time is determined according to the expected operation duration; By comparing the candidate end time with the end time of the current arrival adaptation interval, the interval occupancy relationship is obtained; When the interval occupancy relationship indicates that the candidate end time is not later than the end time of the current arrival adaptation interval, the verification result is determined to be an adaptation; otherwise, the verification result is determined to be an incompatibility.

7. The method according to any one of claims 1 to 6, characterized in that, The step of determining the time-segmented service impact sequence based on the building service demand data, the elevator service relationships, the elevator operating status data, and the fault risk events includes: Based on the latest time of handling the aforementioned fault risk event, the time range from the current time to the latest time of handling is divided into multiple candidate operation periods; The target elevator service demand for each of the candidate operation periods is determined based on the building service demand data. The alternative service capabilities for each of the candidate operation periods are determined based on the elevator service relationships and the elevator operation status data. The service gap and service impact level for each candidate operation period are determined based on the target elevator service demand and the alternative service capabilities. The service impact levels are combined according to the time sequence to obtain the time-segmented service impact sequence.

8. The method according to claim 7, characterized in that, The step of determining the alternative service capabilities for each candidate work period based on the elevator service relationship and the elevator operation status data includes: Based on the elevator service relationship, candidate alternative elevators that overlap with the service floors or service directions of the target elevator are identified from other elevators. Based on the elevator operating status data, elevators that are out of service, under maintenance, or at risk of failure are excluded from the candidate alternative elevators to obtain available alternative elevators; Based on the load status, operating frequency, and assigned service tasks of the available alternative elevators, a record of alternative capabilities for each candidate work period is generated. By aggregating the alternative capability records, alternative service capabilities for each candidate job time period are obtained.

9. The method according to any one of claims 1 to 6, characterized in that, After outputting the joint scheduling scheme to the maintenance terminal, the method further includes: Acquire execution change data, which includes the on-site and operation status returned by the maintenance terminal, new fault risk events, changes in building service requirements, and changes in elevator operation status; Based on the execution change data, update the corresponding fault risk events, time-segmented service impact sequences, and maintenance resource data to obtain the updated scheduling constraints; The updated scheduling constraints are compared with the latest processing time, candidate job time period, and task occupancy status associated with the joint scheduling scheme to determine the effective status of the scheme. When the effective state of the above scheme is invalid, the executed tasks are retained, and the candidates for executable maintenance tasks are reconstructed based on the updated scheduling constraints to generate an updated joint scheduling scheme. The updated joint scheduling scheme and change notification are output to the maintenance terminal.

10. An intelligent dispatching system for elevator fault early warning and maintenance manpower, characterized in that, include: The data acquisition module is used to acquire scheduling input data, which includes elevator operation status data, building service demand data, elevator service relationships, maintenance resource data, currently scheduled tasks, and the normal operation baseline of the corresponding elevators. A segment construction module is used to form a segment feature sequence based on the elevator operation status data; The risk determination module is used to compare the feature sequence of the running segment with the normal operating baseline to determine the fault risk event, which includes the suspected faulty component, the risk evolution state and the latest handling time. The service impact determination module is used to determine the time-segmented service impact sequence based on the building service demand data, the elevator service relationship, the elevator operation status data, and the fault risk event; The task construction module is used to construct executable maintenance task candidates based on the fault risk events, the time-segmented service impact sequences, and the maintenance resource data; The joint scheduling module is used to generate a joint scheduling scheme based on the executable maintenance task candidates, the fault risk events, the time-segmented service impact sequence, and the currently scheduled tasks; The scheme output module is used to output the joint scheduling scheme to the maintenance terminal.