Property service management method and system based on artificial intelligence
By deploying monitoring nodes inside elevators to conduct real-time environmental assessments and cleaning strategy analyses, a cleaning execution ranking dataset is generated. This solves the problem of insufficient manual inspections in traditional elevator hygiene management, enabling timely handling of elevator hygiene conditions and efficient utilization of cleaning resources.
Patent Information
- Application Number
- CN202511060534.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-30
- Publication Date
- 2025-11-11
AI Technical Summary
Traditional elevator hygiene management relies on manual inspections, which cannot promptly detect and address hygiene issues within the elevator, affecting the user experience of residents. Furthermore, it lacks intelligent cleaning strategies and resource optimization.
By deploying monitoring nodes to acquire monitoring data, conducting real-time environmental assessments and cleaning strategy level analyses, generating a cleaning execution ranking dataset, prioritizing the treatment of heavily polluted elevators, and improving the efficiency of cleaning resource utilization by combining service personnel scheduling.
It enables real-time monitoring of elevator hygiene and efficient cleaning response, optimizes the utilization of cleaning resources and the work path of service personnel, and improves the response speed and efficiency of property services.
Smart Images
Figure CN120931437A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of property management technology, and in particular to a property service management method and system based on artificial intelligence. Background Technology
[0002] In recent years, artificial intelligence technology has developed rapidly and achieved remarkable application results in many fields. As a key tool for vertical transportation, the safety and comfort of elevators have become an increasingly important focus. Property management services cover a wide range, and elevator management is a crucial part of it, directly affecting the living experience of residents and the overall service quality of the property. In the property management service system, by reasonably installing cameras inside elevators, real-time video image information inside the elevator can be obtained. Among them, computer vision technology and natural language processing technology provide new ideas and methods for solving elevator hygiene management problems. Using AI-based image recognition algorithms, images in the video can be analyzed to accurately identify the conditions inside the elevator and monitor elevator hygiene. However, for the specific scenario of elevators, there is still a lack of a complete, efficient, and intelligent property service management method in the property management service field.
[0003] In traditional property management, elevator hygiene is typically maintained through regular manual inspections, a method with several shortcomings:
[0004] Manual inspections require professional staff to conduct detailed checks on each elevator at fixed time intervals. During these intervals, hygiene issues may arise inside the elevators that cannot be detected and addressed in a timely manner, thus affecting the user experience of other residents and reducing the quality of property management services.
[0005] It accurately identifies various complex hygiene conditions inside elevators, precisely analyzes the information conveyed by people's actions and language, and achieves effective linkage between hygiene monitoring and cleaning mechanisms.
[0006] To address the aforementioned technical shortcomings, a solution is proposed. Summary of the Invention
[0007] The purpose of this invention is to: acquire monitoring data by deploying monitoring nodes for clear differentiation and management, conduct real-time dynamic environmental assessment of elevators, analyze and judge cleaning strategy levels based on pollution level and impact range, generate a cleaning execution ranking dataset, prioritize the treatment of severely polluted elevators, and improve the utilization efficiency of cleaning resources.
[0008] To achieve the above objectives, the present invention adopts the following technical solution: an artificial intelligence-based property service management method, comprising the following steps:
[0009] Step 1: Obtain the property service management area, set up an area management map, mark several monitoring nodes on the area management map based on the number of elevators, assign classification number labels to the monitoring nodes according to their location in the management area, and obtain the actual environmental data of the elevators through sensors set on the monitoring nodes;
[0010] Step 2: Acquire real-time environmental data, divide it into image dataset and behavior dataset, extract features from the actual environment and user behavior based on environmental constraint triggering conditions, and integrate the elevator environmental constraint triggering times to obtain the environmental impact dataset;
[0011] Step 3: Obtain the environmental impact dataset and, in conjunction with the elevator's operating period, conduct a real-time dynamic environmental assessment of the elevator, generate real-time assessment analysis values, make dynamic decision-making judgments based on the real-time assessment analysis values, and generate real-time cleaning response trigger commands.
[0012] Step 4: Obtain real-time cleaning response trigger instructions, mark contaminated elevators based on elevator monitoring node information, analyze and judge the cleaning strategy level based on the degree of elevator contamination and the scope of impact, generate a cleaning execution ranking dataset, rank the cleaning execution of contaminated elevators, and label monitoring nodes with different levels.
[0013] Step 5: Obtain the cleaning execution ranking dataset. Based on the property service area management map and monitoring node coordinates, combined with the current coordinate information of service personnel, perform service personnel matching analysis, and schedule personnel according to time period and path constraints. Generate dispatch personnel information instructions and transmit them to the property service management terminal.
[0014] Furthermore, step one also includes a data acquisition preprocessing mechanism, specifically including the following:
[0015] S100, Property service management area, set up area management map, delineate the management area of each zone on the area management map, set up several monitoring nodes based on the number of elevators and mark them on the area management map, and assign classification number labels to the monitoring nodes according to their location in the management area;
[0016] S101. Deploy sensor equipment for the monitoring nodes, including cameras, temperature and humidity sensors and microphones, to synchronously collect equipment operation data, and set corresponding coordinate points according to the monitoring node markings.
[0017] S102. Based on the timestamp and sensor device coordinates, integrate them into the area management map, and mark each sensor coordinate point based on the area management map;
[0018] The device data is normalized using sensor equipment; the video data is denoised and the frame rate is adjusted; and the audio data is denoised.
[0019] Furthermore, based on the environmental constraint triggering conditions, an environmental impact dataset is obtained, specifically including the following:
[0020] S200: Set environmental constraint triggering conditions, which are feature points of garbage pollution areas, feature points of user behavior actions, and feature points of voice keywords;
[0021] Elevator pollution maps were collected in advance as a comparison pollution database and set as characteristic points of garbage pollution areas;
[0022] Based on the passenger's action of covering their nose, a feature point is marked when the distance from the nose to the hand lasts for ≥2 seconds.
[0023] Obtain relevant words describing a poor environment from the database and set them as keyword feature points. Set these feature points as environmental constraint trigger conditions.
[0024] S201. Obtain real-time environmental data, distinguish between behavioral actions and image data, and divide them into image datasets and behavioral datasets. Extract features from the actual environment and user behavior. During the extraction process, based on environmental constraints and triggering conditions, identify feature points in the image dataset and behavioral dataset using artificial intelligence to obtain feature points in the image dataset and behavioral dataset.
[0025] S202. Integrate the elevator environment constraint trigger counts for feature points in the image dataset and behavior dataset to obtain the environmental impact dataset, and record the total number of feature points in the image dataset and behavior dataset.
[0026] Action recognition sequence A = {A1, A2, ..., A...} T}, where A L This is the Lth identified action label;
[0027] Keyword recognition sequence K = {K1, K2, ..., K} M}, where K J This refers to the Jth identified keyword;
[0028]
[0029]
[0030] In the above formula, C A C represents the total number of feature points for action recognition. K is the total number of feature points identified by the keyword, and I is an indicator function that takes the value of 1 when the condition is met, and 0 otherwise.
[0031] Furthermore, generate real-time cleaning response trigger commands, specifically including the following:
[0032] S300. Obtain the environmental impact dataset and, in conjunction with the elevator's operating period, determine the current personnel density stage of the elevator. Based on the time point, divide the personnel density stage into peak period, frequent period, and off-peak period.
[0033] S301. Based on the current personnel density stage and environmental impact dataset, obtain the total number of feature points in the image dataset and behavior dataset, and perform real-time dynamic environmental assessment of the elevator to generate real-time assessment analysis value. The real-time assessment analysis value is preset to P.
[0034] S302. Obtain the current real-time evaluation and analysis value, and make dynamic decision judgments. The dynamic decision will be based on the set normal evaluation value. When P≥R, it means that the normal evaluation value is exceeded, and it is judged that a cleaning operation is required, and a cleaning response trigger command is generated.
[0035] Furthermore, a clean execution sorting dataset is generated, specifically including the following:
[0036] S400: Obtain real-time cleaning response trigger commands, mark polluted elevators by combining elevator monitoring node information, analyze the type and degree of elevator pollution based on real-time images of pollution inside the elevator, and obtain pollutants and pollution area.
[0037] S401. Based on the pollutants and the contaminated area, conduct an impact analysis on the pollutants and the contaminated area in the elevator, calculate the impact risk of the pollutants in the elevator, and classify them into three levels. Level 1 represents high-risk pollution, which is manifested as: visible stains, food residue, vomit, bloodstains, strong odor, and obvious dirt in high-frequency contact areas.
[0038] Level 2 represents medium-risk pollution, manifested as minor stains, dust, fingerprints, localized odors, and stains on walls and floors in the medium-frequency contact area;
[0039] Level 3 represents low-risk pollution, characterized by: only floating dust, no visible stains, and no odor;
[0040] An elevator pollution impact prediction report is generated based on the above three levels of classification.
[0041] S402. Based on the elevator pollution impact prediction list, conduct cleaning strategy level analysis and judgment, prioritize cleaning execution, generate a cleaning execution ranking dataset, rank the cleaning execution for polluted elevators, and label monitoring nodes with different levels.
[0042] Furthermore, the instructions for generating dispatcher information are transmitted to the property service management terminal, specifically including the following:
[0043] S500: Obtain the cleaning execution sorting dataset, and convert the pollution data reported by the elevator monitoring nodes, including type, level and range, into a three-dimensional pollution heat map in real time based on the area management map of the property service and the coordinates of the monitoring nodes.
[0044] S501. Combine the current coordinate information of the service personnel to perform service personnel matching analysis. Based on the three-dimensional attribute labels of the service personnel, automatically calculate the time and space cost function for the personnel to reach the target elevator: Cost = Travel time + Remaining task time × Congestion coefficient.
[0045] S502. Select the corresponding personnel for scheduling based on the cost value, generate scheduling personnel information instructions and transmit them to the property service management terminal.
[0046] Furthermore, it also includes an elevator pollution tracing mechanism, specifically including the following:
[0047] S700: Based on the elevator monitoring video and artificial intelligence recognition, conduct behavioral analysis on pollution events inside the elevator, and pre-set intentional and unintentional behaviors based on user behavior characteristics;
[0048] S701. Identify i key features of real-time pollution events, determine the probability of intentional or unintentional pollution, generate pollution event determination results, establish a dynamic database of user pollution records based on the pollution event determination results, and set corresponding intervention strategies based on the number of pollution events, divided into three risk levels.
[0049] Furthermore, it also includes a cleaning implementation feedback mechanism, specifically including the following:
[0050] S600: Elevator image data during the cleaning period is obtained through camera equipment to obtain the cleaning service actions of the cleaner, and the cleaning action capture is identified. In normal and abnormal states, if the dwell time is less than the preset standard time of the cleaning task, it is marked as an abnormal state.
[0051] S601. Obtain a panoramic image of the elevator after cleaning using a camera. Based on the recorded location, area, and type of stains, compare the images before and after cleaning using an image difference analysis algorithm to calculate the stain removal rate.
[0052] S602. Set up an intelligent evaluation mechanism for service personnel, including evaluation based on the time taken from triggering the instruction to the start of cleaning, stain removal rate, and user rating.
[0053] An artificial intelligence-based property service management system includes the following modules:
[0054] The area management module obtains the property service management area, sets up an area management map, sets up several monitoring nodes on the area management map based on the number of elevators, assigns classification number labels to the monitoring nodes according to their location in the management area, and obtains the actual environmental data of the elevators through sensors set on the monitoring nodes.
[0055] The data acquisition module acquires real-time environmental data, which is divided into image datasets and behavior datasets. Based on the environmental constraint triggering conditions, it extracts features from the actual environment and user behavior, and integrates the elevator environmental constraint triggering times to obtain the environmental impact dataset.
[0056] The assessment and decision-making module acquires the environmental impact dataset and, in conjunction with the elevator's operating period, performs a real-time dynamic environmental assessment of the elevator, generating real-time assessment and analysis values. Based on these values, it makes dynamic decision-making judgments and generates real-time cleaning response trigger commands.
[0057] The cleaning execution module acquires real-time cleaning response trigger commands, marks contaminated elevators based on elevator monitoring node information, analyzes and judges the cleaning strategy level based on the degree of elevator contamination and the scope of impact, generates a cleaning execution ranking dataset, ranks the contaminated elevators for cleaning execution, and labels monitoring nodes with different levels.
[0058] The dispatch and communication module acquires the cleaning execution ranking dataset, performs service personnel matching analysis based on the property service area management map and monitoring node coordinates, combined with the current coordinate information of service personnel, and performs dispatch based on time period and path constraints, generating dispatch personnel information instructions to be transmitted to the property service management terminal.
[0059] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:
[0060] This AI-based property service management method and system acquires and clearly distinguishes monitoring data through the deployment of monitoring nodes. The constraint-triggered processing method effectively filters key information. The environmental impact dataset comprehensively reflects the impact of the actual environment and user behavior on the elevator environment. Combined with the environmental impact mechanism, the system performs real-time dynamic assessment of the elevator environment, generating real-time cleaning response trigger commands. Based on these commands and elevator monitoring node information, polluted elevators are marked. The system analyzes and judges cleaning strategy levels based on the degree of pollution and the scope of impact, generating a cleaning execution ranking dataset. Monitoring nodes are labeled with different levels, prioritizing severely polluted elevators, thus improving the utilization efficiency of cleaning resources. Based on the area management map, monitoring node coordinates, and the current coordinates of service personnel, scheduling is performed with time periods and paths as constraints. This comprehensive consideration of various factors optimizes the work paths and time arrangements of service personnel, improving response speed and service efficiency. Attached Figure Description
[0061] Figure 1 A schematic diagram of the method flow structure of the present invention is shown;
[0062] Figure 2 A schematic diagram of the overall structure of the system steps of the present invention is shown. Detailed Implementation
[0063] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0064] Example 1:
[0065] like Figure 1 As shown, the AI-based property service management method includes the following steps:
[0066] Step 1: Obtain the property service management area, set up an area management map, mark several monitoring nodes on the area management map based on the number of elevators, assign classification number labels to the monitoring nodes according to their location in the management area, and obtain the actual environmental data of the elevators through sensors set on the monitoring nodes;
[0067] Step 2: Acquire real-time environmental data, divide it into image dataset and behavior dataset, extract features from the actual environment and user behavior based on environmental constraint triggering conditions, and integrate the elevator environmental constraint triggering times to obtain the environmental impact dataset;
[0068] Step 3: Obtain the environmental impact dataset and, in conjunction with the elevator's operating period, conduct a real-time dynamic environmental assessment of the elevator, generate real-time assessment analysis values, make dynamic decision-making judgments based on the real-time assessment analysis values, and generate real-time cleaning response trigger commands.
[0069] Step 4: Obtain real-time cleaning response trigger instructions, mark contaminated elevators based on elevator monitoring node information, analyze and judge the cleaning strategy level based on the degree of elevator contamination and the scope of impact, generate a cleaning execution ranking dataset, rank the cleaning execution of contaminated elevators, and label monitoring nodes with different levels.
[0070] Step 5: Obtain the cleaning execution ranking dataset. Based on the property service area management map and monitoring node coordinates, combined with the current coordinate information of service personnel, perform service personnel matching analysis, and schedule personnel according to time period and path constraints. Generate dispatch personnel information instructions and transmit them to the property service management terminal.
[0071] Step one also includes a data acquisition preprocessing mechanism, specifically including the following:
[0072] S100, Property service management area, set up area management map, delineate the management area of each zone on the area management map, set up several monitoring nodes based on the number of elevators and mark them on the area management map, and assign classification number labels to the monitoring nodes according to their location in the management area;
[0073] S101. Deploy sensor equipment for the monitoring nodes, including cameras, temperature and humidity sensors and microphones, to synchronously collect equipment operation data, and set corresponding coordinate points according to the monitoring node markings.
[0074] S102. Based on the timestamp and sensor device coordinates, integrate them into the area management map, and mark each sensor coordinate point based on the area management map;
[0075] The device data is normalized using sensor equipment; the video data is denoised and the frame rate is adjusted; and the audio data is denoised.
[0076] Based on the environmental constraint triggering conditions, the environmental impact dataset is obtained, which specifically includes the following:
[0077] S200: Set environmental constraint triggering conditions, which are feature points of garbage pollution areas, feature points of user behavior actions, and feature points of voice keywords;
[0078] Elevator pollution maps were collected in advance as a comparison pollution database and set as characteristic points of garbage pollution areas;
[0079] Based on the passenger's action of covering their nose, a feature point is marked when the distance from the nose to the hand lasts for ≥2 seconds.
[0080] Obtain relevant words from the database that describe a poor environment and set them as keyword feature points. Set these feature points as environmental constraint trigger conditions.
[0081] S201. Obtain real-time environmental data, distinguish between behavioral actions and image data, and divide them into image datasets and behavioral datasets. Extract features from the actual environment and user behavior. During the extraction process, based on environmental constraints and triggering conditions, identify feature points in the image dataset and behavioral dataset using artificial intelligence to obtain feature points in the image dataset and behavioral dataset.
[0082] S202. Integrate the elevator environment constraint trigger counts for feature points in the image dataset and behavior dataset to obtain the environmental impact dataset, and record the total number of feature points in the image dataset and behavior dataset.
[0083] Action recognition sequence A = {A1, A2, ..., A...} T}, where A L This is the Lth identified action label;
[0084] Keyword recognition sequence K = {K1, K2, ..., K} M}, where K J This refers to the Jth identified keyword;
[0085]
[0086]
[0087] In the above formula, C A C represents the total number of feature points for action recognition. K is the total number of feature points identified by the keyword, and I is an indicator function that takes the value of 1 when the condition is met, and 0 otherwise.
[0088] Generate real-time cleaning response trigger commands, specifically including the following:
[0089] S300. Obtain the environmental impact dataset and, in conjunction with the elevator's operating period, determine the current personnel density stage of the elevator. Based on the time point, divide the personnel density stage into peak period, frequent period, and off-peak period.
[0090] S301. Based on the current personnel density stage and environmental impact dataset, obtain the total number of feature points in the image dataset and behavior dataset, and perform real-time dynamic environmental assessment of the elevator to generate real-time assessment analysis value. The real-time assessment analysis value is preset to P.
[0091] S302. Obtain the current real-time evaluation and analysis value, and make dynamic decision judgments. The dynamic decision will be based on the set normal evaluation value. When P≥R, it means that the normal evaluation value is exceeded, and it is judged that a cleaning operation is required, and a cleaning response trigger command is generated.
[0092] Generate a clean execution sorting dataset, specifically including the following:
[0093] S400: Obtain real-time cleaning response trigger commands, mark polluted elevators by combining elevator monitoring node information, analyze the type and degree of elevator pollution based on real-time images of pollution inside the elevator, and obtain pollutants and pollution area.
[0094] S401. Based on the pollutants and the contaminated area, conduct an impact analysis on the pollutants and the contaminated area in the elevator, calculate the impact risk of the pollutants in the elevator, and classify them into three levels. Level 1 represents high-risk pollution, which is manifested as: visible stains, food residue, vomit, bloodstains, strong odor, and obvious dirt in high-frequency contact areas.
[0095] Level 2 represents medium-risk pollution, manifested as minor stains, dust, fingerprints, localized odors, and stains on walls and floors in the medium-frequency contact area;
[0096] Level 3 represents low-risk pollution, characterized by: only floating dust, no visible stains, and no odor;
[0097] An elevator pollution impact prediction report is generated based on the above three levels of classification.
[0098] S402. Based on the elevator pollution impact prediction list, conduct cleaning strategy level analysis and judgment, prioritize cleaning execution, generate a cleaning execution ranking dataset, rank the cleaning execution for polluted elevators, and label monitoring nodes with different levels.
[0099] The instruction to generate dispatcher information is transmitted to the property service management terminal, specifically including the following:
[0100] S500: Obtain the cleaning execution sorting dataset, and convert the pollution data reported by the elevator monitoring nodes, including type, level and range, into a three-dimensional pollution heat map in real time based on the area management map of the property service and the coordinates of the monitoring nodes.
[0101] S501. Combine the current coordinate information of the service personnel to perform service personnel matching analysis. Based on the three-dimensional attribute labels of the service personnel, automatically calculate the time and space cost function for the personnel to reach the target elevator: Cost = Travel time + Remaining task time × Congestion coefficient.
[0102] S502. Select the corresponding personnel for scheduling based on the cost value, generate scheduling personnel information instructions and transmit them to the property service management terminal.
[0103] It also includes an elevator pollution traceability mechanism, specifically including the following:
[0104] S700: Based on the elevator monitoring video and artificial intelligence recognition, conduct behavioral analysis on pollution events inside the elevator, and pre-set intentional and unintentional behaviors based on user behavior characteristics;
[0105] S701. Identify i key features of real-time pollution events, determine the probability of intentional or unintentional pollution, generate pollution event determination results, establish a dynamic database of user pollution records based on the pollution event determination results, and set corresponding intervention strategies based on the number of pollution events, divided into three risk levels.
[0106] It also includes a cleaning performance feedback mechanism, specifically including the following:
[0107] S600: Elevator image data during the cleaning period is obtained through camera equipment to obtain the cleaning service actions of the cleaner, and the cleaning action capture is identified. In normal and abnormal states, if the dwell time is less than the preset standard time of the cleaning task, it is marked as an abnormal state.
[0108] S601. Obtain a panoramic image of the elevator after cleaning using a camera. Based on the recorded location, area, and type of stains, compare the images before and after cleaning using an image difference analysis algorithm to calculate the stain removal rate.
[0109] S602. Set up an intelligent evaluation mechanism for service personnel, including evaluation based on the time taken from triggering the instruction to the start of cleaning, stain removal rate, and user rating.
[0110] Example 2:
[0111] like Figure 2 As shown, an artificial intelligence-based property service management system includes the following modules:
[0112] The area management module obtains the property service management area, sets up an area management map, sets up several monitoring nodes on the area management map based on the number of elevators, assigns classification number labels to the monitoring nodes according to their location in the management area, and obtains the actual environmental data of the elevators through sensors set on the monitoring nodes.
[0113] The data acquisition module acquires real-time environmental data, which is divided into image datasets and behavior datasets. Based on the environmental constraint triggering conditions, it extracts features from the actual environment and user behavior, and integrates the elevator environmental constraint triggering times to obtain the environmental impact dataset.
[0114] The assessment and decision-making module acquires the environmental impact dataset and, in conjunction with the elevator's operating period, performs a real-time dynamic environmental assessment of the elevator, generating real-time assessment and analysis values. Based on these values, it makes dynamic decision-making judgments and generates real-time cleaning response trigger commands.
[0115] The cleaning execution module acquires real-time cleaning response trigger commands, marks contaminated elevators based on elevator monitoring node information, analyzes and judges the cleaning strategy level based on the degree of elevator contamination and the scope of impact, generates a cleaning execution ranking dataset, ranks the contaminated elevators for cleaning execution, and labels monitoring nodes with different levels.
[0116] The dispatch and communication module acquires the cleaning execution ranking dataset, performs service personnel matching analysis based on the property service area management map and monitoring node coordinates, combined with the current coordinate information of service personnel, and performs dispatch based on time period and path constraints, generating dispatch personnel information instructions to be transmitted to the property service management terminal.
[0117] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value.
[0118] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.
[0119] In the two embodiments provided in this application, it should be understood that the disclosed methods and systems can be implemented in other ways; for example, the system embodiments described above are merely illustrative, and the division of modules is merely a logical functional division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed; furthermore, the coupling or direct coupling or communication connection shown or discussed between each other can be indirect coupling or communication connection through some interfaces, methods or modules, and can be electrical, mechanical or other forms.
[0120] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.
Claims
1. A property service management method based on artificial intelligence, characterized in that, Includes the following steps: Step 1: Obtain the property service management area, set up an area management map, mark several monitoring nodes on the area management map based on the number of elevators, assign classification number labels to the monitoring nodes according to their location in the management area, and obtain the actual environmental data of the elevators through sensors set on the monitoring nodes; Step 2: Acquire real-time environmental data, divide it into image dataset and behavior dataset, extract features from the actual environment and user behavior based on environmental constraint triggering conditions, and integrate the elevator environmental constraint triggering times to obtain the environmental impact dataset; Step 3: Obtain the environmental impact dataset and, in conjunction with the elevator's operating period, conduct a real-time dynamic environmental assessment of the elevator, generate real-time assessment analysis values, make dynamic decision-making judgments based on the real-time assessment analysis values, and generate real-time cleaning response trigger commands. Step 4: Obtain real-time cleaning response trigger instructions, mark contaminated elevators based on elevator monitoring node information, analyze and judge the cleaning strategy level based on the degree of elevator contamination and the scope of impact, generate a cleaning execution ranking dataset, rank the cleaning execution of contaminated elevators, and label monitoring nodes with different levels. Step 5: Obtain the cleaning execution ranking dataset. Based on the property service area management map and monitoring node coordinates, combined with the current coordinate information of service personnel, perform service personnel matching analysis, and schedule personnel according to time period and path constraints. Generate dispatch personnel information instructions and transmit them to the property service management terminal.
2. The property service management method based on artificial intelligence according to claim 1, characterized in that, Step one also includes a data acquisition preprocessing mechanism, specifically including the following: S100, Property service management area, set up area management map, delineate the management area of each zone on the area management map, set up several monitoring nodes based on the number of elevators and mark them on the area management map, and assign classification number labels to the monitoring nodes according to their location in the management area; S101. Deploy sensor equipment for the monitoring nodes, including cameras, temperature and humidity sensors and microphones, to synchronously collect equipment operation data, and set corresponding coordinate points according to the monitoring node markings. S102. Based on the timestamp and sensor device coordinates, integrate them into the area management map, and mark each sensor coordinate point based on the area management map; The device data is normalized using sensor equipment; the video data is denoised and the frame rate is adjusted; and the audio data is denoised.
3. The property service management method based on artificial intelligence according to claim 1, characterized in that, Based on the environmental constraint triggering conditions, the environmental impact dataset is obtained, which specifically includes the following: S200: Set environmental constraint triggering conditions, which are feature points of garbage pollution areas, feature points of user behavior actions, and feature points of voice keywords; Elevator pollution maps were collected in advance as a comparison pollution database and set as characteristic points of garbage pollution areas; Based on the passenger's action of covering their nose, a feature point is marked when the distance from the nose to the hand lasts for ≥2 seconds. Obtain relevant words from the database that describe a poor environment and set them as keyword feature points. Set these feature points as environmental constraint trigger conditions. S201. Obtain real-time environmental data, distinguish between behavioral actions and image data, and divide them into image datasets and behavioral datasets. Extract features from the actual environment and user behavior. During the extraction process, based on environmental constraints and triggering conditions, identify feature points in the image dataset and behavioral dataset using artificial intelligence to obtain feature points in the image dataset and behavioral dataset. S202. Integrate the elevator environment constraint trigger counts for feature points in the image dataset and behavior dataset to obtain the environmental impact dataset, and record the total number of feature points in the image dataset and behavior dataset. Action recognition sequence A = {A1, A2, ..., A...} T }, where A L This is the Lth identified action label; Keyword recognition sequence K = {K1, K2, ..., K} M }, where K J This refers to the Jth identified keyword; In the above formula, C A C represents the total number of feature points for action recognition. K is the total number of feature points identified by the keyword, and I is an indicator function that takes the value of 1 when the condition is met, and 0 otherwise.
4. The property service management method based on artificial intelligence according to claim 1, characterized in that, Generate real-time cleaning response trigger commands, specifically including the following: S300. Obtain the environmental impact dataset and, in conjunction with the elevator's operating period, determine the current personnel density stage of the elevator. Based on the time point, divide the personnel density stage into peak period, frequent period, and off-peak period. S301. Based on the current personnel density stage and environmental impact dataset, obtain the total number of feature points in the image dataset and behavior dataset, and perform real-time dynamic environmental assessment of the elevator to generate real-time assessment analysis value. The real-time assessment analysis value is preset to P. S302. Obtain the current real-time evaluation and analysis value, and make dynamic decision judgments. The dynamic decision will be based on the set normal evaluation value. When P≥R, it means that the normal evaluation value is exceeded, and it is judged that a cleaning operation is required, and a cleaning response trigger command is generated.
5. The property service management method based on artificial intelligence according to claim 1, characterized in that, Generate a clean execution sorting dataset, specifically including the following: S400: Obtain real-time cleaning response trigger commands, mark polluted elevators by combining elevator monitoring node information, analyze the type and degree of elevator pollution based on real-time images of pollution inside the elevator, and obtain pollutants and pollution area. S401. Based on the pollutants and the contaminated area, conduct an impact analysis on the pollutants and the contaminated area in the elevator, calculate the impact risk of the pollutants in the elevator, and classify them into three levels. Level 1 represents high-risk pollution, which is manifested as: visible stains, food residue, vomit, bloodstains, strong odor, and obvious dirt in high-frequency contact areas. Level 2 represents medium-risk pollution, manifested as minor stains, dust, fingerprints, localized odors, and stains on walls and floors in the medium-frequency contact area; Level 3 represents low-risk pollution, characterized by: only floating dust, no visible stains, and no odor; Based on the above three levels of classification, an elevator pollution impact prediction report is generated. S402. Based on the elevator pollution impact prediction list, conduct cleaning strategy level analysis and judgment, prioritize cleaning execution, generate a cleaning execution ranking dataset, rank the cleaning execution for polluted elevators, and label monitoring nodes with different levels.
6. The property service management method based on artificial intelligence according to claim 1, characterized in that, The instruction to generate dispatcher information is transmitted to the property service management terminal, specifically including the following: S500: Obtain the cleaning execution sorting dataset, and convert the pollution data reported by the elevator monitoring nodes, including type, level and range, into a three-dimensional pollution heat map in real time based on the area management map of the property service and the coordinates of the monitoring nodes. S501. Combine the current coordinate information of the service personnel to perform service personnel matching analysis. Based on the three-dimensional attribute labels of the service personnel, automatically calculate the time and space cost function for the personnel to reach the target elevator: Cost = Travel time + Remaining task time × Congestion coefficient. S502. Select the corresponding personnel for scheduling based on the cost value, generate scheduling personnel information instructions and transmit them to the property service management terminal.
7. The property service management method based on artificial intelligence according to claim 1, characterized in that, It also includes an elevator pollution traceability mechanism, specifically including the following: S700: Based on the elevator monitoring video and artificial intelligence recognition, conduct behavioral analysis on pollution events inside the elevator, and pre-set intentional and unintentional behaviors based on user behavior characteristics; S701. Identify i key features of real-time pollution events, determine the probability of intentional or unintentional pollution, generate pollution event determination results, establish a dynamic database of user pollution records based on the pollution event determination results, and set corresponding intervention strategies based on the number of pollution events, divided into three risk levels.
8. The property service management method based on artificial intelligence according to claim 1, characterized in that, It also includes a cleaning performance feedback mechanism, specifically including the following: S600: Elevator image data during the cleaning period is obtained through camera equipment to obtain the cleaning service actions of the cleaner, and the cleaning action capture is identified. In normal and abnormal states, if the dwell time is less than the preset standard time of the cleaning task, it is marked as an abnormal state. S601. Obtain a panoramic image of the elevator after cleaning using a camera. Based on the recorded location, area, and type of stains, compare the images before and after cleaning using an image difference analysis algorithm to calculate the stain removal rate. S602. Set up an intelligent evaluation mechanism for service personnel, including evaluation based on the time taken from triggering the instruction to the start of cleaning, stain removal rate, and user rating.
9. A property service management system based on artificial intelligence, characterized in that, Includes the following modules: The area management module obtains the property service management area, sets up an area management map, sets up several monitoring nodes on the area management map based on the number of elevators, assigns classification number labels to the monitoring nodes according to their location in the management area, and obtains the actual environmental data of the elevators through sensors set on the monitoring nodes. The data acquisition module acquires real-time environmental data, which is divided into image datasets and behavior datasets. Based on the environmental constraint triggering conditions, it extracts features from the actual environment and user behavior, and integrates the elevator environmental constraint triggering times to obtain the environmental impact dataset. The assessment and decision-making module acquires the environmental impact dataset and, in conjunction with the elevator's operating period, performs a real-time dynamic environmental assessment of the elevator, generating real-time assessment and analysis values. Based on these values, it makes dynamic decision-making judgments and generates real-time cleaning response trigger commands. The cleaning execution module acquires real-time cleaning response trigger commands, marks contaminated elevators based on elevator monitoring node information, analyzes and judges the cleaning strategy level based on the degree of elevator contamination and the scope of impact, generates a cleaning execution ranking dataset, ranks the contaminated elevators for cleaning execution, and labels monitoring nodes with different levels. The dispatch and communication module acquires the cleaning execution ranking dataset, performs service personnel matching analysis based on the property service area management map and monitoring node coordinates, combined with the current coordinate information of service personnel, and performs dispatch based on time period and path constraints, generating dispatch personnel information instructions to be transmitted to the property service management terminal.