Property sub-house inspection method and device, computer equipment and storage medium

Abstract

This invention relates to the field of building inspection technology, and discloses a method, apparatus, computer equipment, and storage medium for inspecting individual properties. First, anomaly detection is performed on the property data entered, and an intelligent verification mechanism is constructed to identify and intercept erroneous data, obtaining anomaly detection results to ensure data quality. Then, if the anomaly detection results indicate that the property data entered is free of anomalies, automated indicator processing is performed on the property data based on preset building engineering quality indicator processing rules to obtain building engineering quality indicator results, thereby improving processing efficiency and accuracy. Finally, a quality risk early warning model is used to predict the risk of the building engineering quality indicator results, obtaining building risk prediction results to identify potential problems and support early intervention, thereby reducing quality risks. By integrating intelligent verification and risk prediction models, the intelligent transformation of property unit inspection can be promoted, enhancing risk prevention and control capabilities.

Description

Technical Field

[0001] This invention relates to the field of building inspection technology, and in particular to a method, apparatus, computer equipment, and storage medium for inspecting individual properties. Background Technology

[0002] As the real estate industry transforms towards "refined quality control," the processing of individual inspection data is showing a development trend of "intelligentization and early warning."

[0003] In related technologies, the calculation of indicators relies on manual calculation using formulas in Excel. However, manual calculation is prone to introducing systematic errors, resulting in a high error rate and significant deviations. Therefore, a new method for verifying individual properties needs to be proposed. Summary of the Invention

[0004] The embodiments described in this specification aim to at least partially solve one of the technical problems in the related art. To this end, the embodiments described in this specification propose a method, apparatus, computer equipment, and storage medium for inspecting individual units of a property.

[0005] This specification provides a method for inspecting individual units in a property, the method comprising: Anomaly detection is performed on the data entered by the property management company to obtain the anomaly detection results. If the data anomaly detection result indicates that there is no anomaly in the property data, the property data is processed according to the preset housing engineering quality index processing rules to obtain the housing engineering quality index result. Based on the quality risk early warning model, the risk prediction results of the building project quality indicators are obtained.

[0006] In one implementation, the step of performing anomaly detection on the property entry data and obtaining the data anomaly detection result includes: Anomaly detection is performed on the property data entered based on a preset abnormal data verification mechanism to obtain the data anomaly detection result.

[0007] In one implementation, the anomaly detection of the property entry data based on a preset anomaly data verification mechanism includes at least one of the following: Anomaly detection is performed on the property data entered based on preset reasonable rules; Anomaly detection is performed on the property data entered based on a preset reasonable threshold range; The property data entry is anomaly detected based on a preset correlation deviation, where the preset correlation deviation represents the range of deviation between the property data entry and the estimated value.

[0008] In one implementation, the property entry data includes the maximum measured net height of the building and the minimum measured net height of the building. The building construction quality index results include at least one of room qualification rate, building rectification rate, net height range, and measurement deviation. The property entry data is processed based on preset building construction quality index processing rules to obtain building construction quality index results, including: Based on the property entry data, statistics are performed to obtain the total number of qualified entry data corresponding to the first room, the total number of entry data corresponding to the first room, the number of rectified defects corresponding to the first building, and the total number of defects corresponding to the first building. The room pass rate is calculated by comparing the total number of qualified data entries corresponding to the first room with the total number of data entries corresponding to the first room. The building rectification rate is calculated by comparing the number of rectified defects in the first building with the total number of defects in the first building. The difference between the measured maximum and minimum net height of the building is calculated to obtain the net height range. The measurement deviation is obtained by calculating the difference between the property data entered and the preset standard value.

[0009] In one implementation, the risk prediction of the housing project quality indicators based on the quality risk early warning model, to obtain housing risk prediction results, includes: The quality risk early warning model includes a preset risk threshold. The results of the building construction quality indicators are compared with the preset risk threshold to obtain the building risk prediction result; or Based on the quality risk early warning model, feature extraction is performed on the quality index results of the housing project to obtain index features. Based on the index features, risk prediction is performed to obtain the risk prediction result of the housing project.

[0010] In one embodiment, the method further includes: If the data anomaly detection result indicates that the property entry data is abnormal, an anomaly alert message will be pushed. Based on the aforementioned abnormality alerts, conduct inspections and rectifications, and obtain the rectified property data entered into the system. Anomaly detection is performed on the rectified property data, and the above operation is repeated based on the anomaly detection results until the anomaly detection results indicate that there are no anomalies in the property data.

[0011] In one embodiment, the method further includes: When the housing risk prediction results indicate the presence of a warning risk, a warning message is generated and pushed out. The warning information is tracked and processed, and the warning status corresponding to the warning information is updated according to the processing result.

[0012] In one embodiment, the method further includes: An analysis report is generated based on the results of the aforementioned housing project quality indicators.

[0013] This specification provides a property unit inspection device, the device comprising: The anomaly detection module is used to detect anomalies in the property data and obtain the data anomaly detection results. The indicator processing module is used to process the property data based on preset housing engineering quality indicator processing rules when the data anomaly detection result indicates that there is no anomaly in the property data, and obtain the housing engineering quality indicator result. The risk prediction module is used to predict the risk of the building project quality indicators based on the quality risk early warning model, and obtain the building risk prediction result.

[0014] This specification provides a computer device comprising: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by the one or more processors, the instructions being executed by the one or more processors to cause the one or more processors to perform the steps of the method described in any of the above embodiments.

[0015] This specification provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the above embodiments.

[0016] This specification provides a computer program product that includes instructions that, when executed by a processor of a computer device, enable the computer device to perform the steps of the method described in any of the above embodiments.

[0017] In the above-described implementation method, firstly, anomaly detection is performed on the property data entered, constructing an intelligent verification mechanism to identify and intercept erroneous data, obtaining data anomaly detection results to ensure data quality. Then, if the data anomaly detection results indicate that the property data entered is free of anomalies, automated indicator processing is performed on the property data entered based on preset housing engineering quality indicator processing rules to obtain housing engineering quality indicator results, thereby improving processing efficiency and accuracy. Finally, risk prediction is performed on the housing engineering quality indicator results based on a quality risk early warning model, obtaining housing risk prediction results to identify potential problems, support early intervention, and thus reduce quality risks. By integrating intelligent verification and risk prediction models, the transformation of property unit inspection towards intelligence can be promoted, enhancing risk prevention and control capabilities. Attached Figure Description

[0018] Figure 1 A flowchart of the property unit inspection method provided for the implementation of this specification; Figure 2 A flowchart illustrating the process of obtaining room qualification rate for the implementation of this specification; Figure 3 A flowchart illustrating the process of obtaining housing construction quality indicators for the implementation of this specification; Figure 4 A flowchart illustrating the process of obtaining valid property entry data provided for the implementation of this specification; Figure 5 A flowchart illustrating the update of warning information corresponding to the warning status provided in the embodiments of this specification; Figure 6 A schematic diagram of the property unit inspection device provided for the embodiments of this specification; Figure 7 An internal structural diagram of a computer device provided for embodiments of this specification. Detailed Implementation

[0019] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0020] As the real estate industry transforms towards "refined quality control," the processing of unit-by-unit inspection data is showing a trend towards "intelligent and early warning" capabilities. However, in practice, the following issues still arise: Individual unit inspections involve multiple indicators such as "space dimensional range, deviation rate, and pass rate," which are complex to calculate. For example, the net height range needs to be obtained by subtracting the minimum value from the maximum value, and the inspection pass rate needs to be calculated based on the "proportion of qualified rooms to the total number of rooms."

[0021] In related technologies, the process mainly relies on manual calculations using formulas in Excel. Taking a single building as an example, it requires processing over 500 data points and more than 20 indicators, taking over 8 hours, which is inefficient.

[0022] Manual calculations are prone to introducing systematic errors, resulting in high error rates and significant biases. Common problems include omissions in range calculations and missed rooms in pass rate statistics, leading to an overall error rate exceeding 8%, severely impacting the accuracy of quality assessments. For example, a building's actual pass rate might be 75%, but due to statistical errors, it could be mistakenly classified as 85%.

[0023] A single abnormal data point (such as an incorrectly recorded net height value) may cause deviations in the quality assessment results of the entire building, thereby affecting the management's decision-making.

[0024] In related technologies, only "numerical range prompts" are provided (such as a suggested net height of 2500–3000 mm), lacking an effective mechanism for verifying abnormal data. This allows such outliers to enter the evaluation process, ultimately misleading subsequent rectification directions. For example, the system cannot recognize data that is clearly out of touch with reality, such as "net height 5000 mm" or "net span 1000 mm," leading to deviations in the calculation of the overall building's pass rate (e.g., the pass rate is mistakenly calculated as 60% instead of 80%), which in turn leads to incorrect rectification decisions (such as incurring rectification costs when no rectification is needed).

[0025] Real estate companies have proposed a management requirement of "72-hour advance warning for quality risks," aiming to avoid large-scale rework later through proactive warnings (for example, if a building has widespread substandard ceiling height, construction techniques should be adjusted in advance, rather than remedied afterward). However, the current system can only record historical problem data (such as "10 hollow spots found in a building") and lacks proactive risk warning functions. Risk assessment relies on manual intervention, and the system cannot automatically identify potential risks such as "a building's hollow spot rate exceeds 5%" or "a construction unit's rectification is continuously delayed," resulting in a serious lag in risk detection.

[0026] Due to the lack of early warning systems, quality problems often only surface during the later rectification phase, requiring remedial measures and significantly increasing rectification costs. Statistics show that such delayed rectification leads to an average cost increase of approximately 30% (for example, problems that could have been resolved through process adjustments ultimately necessitate demolition and reinstallation). For instance, managers must periodically review defect statistics charts (such as bar charts showing the number of defects in each building) and manually assess risks. This method results in quality risks being discovered on average approximately 24 hours later, and measures can only be implemented after the rectification phase (e.g., rework and reinstallation after walls have been completed), failing to achieve preventative measures.

[0027] In summary, due to the lack of intelligent algorithm support, it is difficult to achieve efficient, accurate, and forward-looking quality control.

[0028] Based on the above analysis, this specification provides a method for inspecting individual property units. First, anomaly detection is performed on the property entry data. An intelligent verification mechanism is constructed to identify and intercept erroneous data, obtaining anomaly detection results to ensure data quality. Then, if the anomaly detection results indicate that the property entry data is free of anomalies, automated indicator processing is performed on the property entry data based on preset housing engineering quality indicator processing rules to obtain housing engineering quality indicator results, thereby improving processing efficiency and accuracy. Finally, a quality risk early warning model is used to predict the risk of the housing engineering quality indicator results, obtaining housing risk prediction results to identify potential problems and support early intervention, thereby reducing quality risks. By integrating intelligent verification and risk prediction models, the intelligent transformation of property unit inspection can be promoted, enhancing risk prevention and control capabilities.

[0029] This specification provides a method for inspecting individual units in a property. Please refer to [link / reference]. Figure 1 The method for inspecting individual units in a property may include the following steps: S110. Perform anomaly detection on the property data entered and obtain the data anomaly detection results.

[0030] Specifically, based on pre-defined housing data collection specifications and parameter lists, on-site measurements and condition inspections of the houses are conducted, and the results are entered into the system to form structured property data. This property data can include measurement data (such as net depth, floor height, and area) and inspection data (such as the presence of wall cracks, leaks, and hollow areas). Subsequently, anomaly detection is performed on the property data. Using methods such as range verification, the system identifies whether the data exceeds a reasonable detection range, obtaining the anomaly detection results.

[0031] S120. If the data anomaly detection result indicates that there is no anomaly in the property data, the property data is processed according to the preset housing engineering quality index processing rules to obtain the housing engineering quality index result.

[0032] S130. Based on the quality risk early warning model, risk prediction is performed on the results of housing project quality indicators to obtain housing risk prediction results.

[0033] Specifically, if the data anomaly detection result indicates that there are no anomalies in the property entry data, it means that the measurement data meets the standards and quality requirements, and the indicator processing flow will be automatically triggered. Then, according to the preset housing engineering quality indicator processing rules, the property entry data is automatically processed to obtain the corresponding housing engineering quality indicator results (such as deviation value, range, and pass rate). Finally, the quality risk early warning model is used to predict the risk of the housing engineering quality indicator results, which can identify potential fluctuations or abnormal trends and obtain housing risk prediction results.

[0034] It should be noted that a cloud-based intelligent analysis platform can be used to achieve closed-loop management from data verification and indicator generation to risk warning, thereby improving the timeliness, accuracy, and intelligence of quality monitoring.

[0035] In the above implementation, firstly, anomaly detection is performed on the property data entered, and an intelligent verification mechanism is constructed to identify and intercept erroneous data, obtaining data anomaly detection results to ensure data quality. Then, if the data anomaly detection results indicate that the property data entered is free of anomalies, automated indicator processing is performed on the property data entered based on preset housing engineering quality indicator processing rules to obtain housing engineering quality indicator results, thereby improving processing efficiency and accuracy. Finally, risk prediction is performed on the housing engineering quality indicator results based on a quality risk early warning model to obtain housing risk prediction results, identifying potential problems and supporting early intervention, thereby reducing quality risks. By integrating intelligent verification and risk prediction models, the transformation of property unit inspection towards intelligence can be promoted, enhancing risk prevention and control capabilities.

[0036] In some implementations, anomaly detection is performed on the property entry data to obtain anomaly detection results. This may include: performing anomaly detection on the property entry data based on a preset anomaly data verification mechanism to obtain anomaly detection results.

[0037] Specifically, to achieve effective quality control of property data entry, a systematic and clearly defined pre-defined abnormal data verification mechanism is established. This mechanism, through pre-defined multi-dimensional verification rules, performs anomaly detection on the property data entry data after obtaining it, determining whether the data exceeds a reasonable range and obtaining the anomaly detection result.

[0038] In the above implementation, anomaly detection is performed on the property data based on a preset abnormal data verification mechanism to obtain the data anomaly detection results, which provides a data basis for whether to process the indicators in the future.

[0039] In some implementations, anomaly detection of property data entry based on a preset anomaly data verification mechanism may include: anomaly detection of property data entry based on preset reasonable rules.

[0040] Specifically, property data entry can include multiple logically related data entries. To ensure the logical rationality of the data, preset rationality rules are set based on the logical relationships within the data. For example, in unit information, "net depth" is greater than "net width". Then, anomaly detection will be performed on the corresponding property data entry according to the preset rationality rules to improve the accuracy and reliability of the data. It should be noted that multiple preset rationality rules can exist, each targeting different logical relationships.

[0041] For example, the property data entered may include net depth and net width. A preset reasonable rule can be that the net depth is greater than the net width. For example, if the net depth is 3600mm and the net width is 3000mm, it meets the rule and is judged as reasonable; conversely, if the net depth is not greater than the net width, it is considered as abnormal data.

[0042] In the above implementation, anomaly detection is performed on the property data based on a preset reasonable threshold, providing a data basis for whether to process the indicators in the future.

[0043] In some implementations, anomaly detection of property data entry based on a preset abnormal data verification mechanism may include: anomaly detection of property data entry based on a preset reasonable threshold range.

[0044] Specifically, to ensure the reasonableness of the scope of property data entry, corresponding preset reasonable threshold ranges are set based on the business characteristics and regulatory requirements of the data. Then, anomaly detection is performed on the corresponding property data according to the preset reasonable threshold ranges to improve the accuracy and reliability of the data. It should be noted that multiple preset reasonable threshold ranges can exist, each targeting different types of data.

[0045] For example, the data entered by the property management company may include a mandatory range. The preset reasonable threshold range corresponding to the mandatory range can be 2500-3200mm. If the mandatory range exceeds the corresponding preset reasonable threshold range, it is determined to be abnormal data; conversely, if the mandatory range does not exceed the corresponding preset reasonable threshold range, it is considered to be reasonable data.

[0046] In the above implementation, anomaly detection is performed on the property data based on a preset reasonable threshold range, providing a data basis for whether to process the indicators in the future.

[0047] In some implementations, anomaly detection of property data entry based on a preset abnormal data verification mechanism may include: anomaly detection of property data entry based on a preset correlation deviation.

[0048] Among them, the preset correlation deviation represents the range of deviation between the property data entered and the estimated value.

[0049] Specifically, a preset correlation deviation is set based on the property data entered and its corresponding reasonableness, reflecting the acceptable range of difference between the entered data and the theoretical values ​​calculated based on business rules. Then, anomaly detection is performed on the corresponding property data entered based on the preset correlation deviation to improve the accuracy and reliability of the data. It should be noted that multiple preset correlation deviations can exist, each targeting different entered data.

[0050] For example, the data entered into the property management system may include the actual measured net height, and the estimated value may include the estimated net height of the unit type. The preset correlation deviation can be that the deviation between the actual measured net height and the estimated net height of the unit type does not exceed ±10%. For example, if the estimated net height of a certain unit type is 2800mm, then the actual measured net height should be between 2520mm and 3080mm; if the actual measured net height is lower than 2520mm or higher than 3080mm, it will be judged as data anomaly.

[0051] In the above implementation, anomaly detection is performed on the property data based on a preset correlation deviation, providing a data basis for whether to perform indicator processing in the future.

[0052] In some implementations, please refer to Figure 2 The property data entered includes the maximum and minimum measured net height of the building. The building construction quality index results include at least one of the following: room qualification rate, building rectification rate, net height range, and measurement deviation. Based on the preset building construction quality index processing rules, the property data entered is processed to obtain the building construction quality index results, which may include the following steps: S210. Based on the property entry data, perform statistics to obtain the total number of qualified entry data corresponding to the first room, the total number of entry data corresponding to the first room, the number of rectified defects corresponding to the first building, and the total number of defects corresponding to the first building.

[0053] S220. Calculate the room pass rate by comparing the total number of qualified data entries corresponding to the first room with the total number of data entries corresponding to the first room.

[0054] Specifically, if the data anomaly detection result indicates that the property management's entered data is free of anomalies, it means that the measurement data meets the standards and quality requirements. However, some inspection data (such as whether there are wall cracks, leaks, or hollow areas) may still be substandard. Therefore, after confirming that the data anomaly detection result indicates that the property management's entered data is free of anomalies, statistics can be compiled based on the entered data for the first room to obtain the total number of qualified entered data for the first room and the total number of entered data for the first room. Then, the ratio of the total number of qualified entered data for the first room to the total number of entered data for the first room is calculated to obtain the room pass rate. For example, the room pass rate = (total number of qualified entered data / total number of entered data) × 100%. For instance, if room A has 8 entered data points, and 7 of them are qualified, then the room pass rate = (7 / 8) × 100% = 87.5%.

[0055] In addition, the inspection data can also include records of defect status. For example, if a defect is "wall crack exists," and the wall crack has been repaired during subsequent inspections, it will be recorded as "not present," indicating that the defect has been rectified; if it still exists, it will be recorded as "not rectified." Based on this, statistics are compiled from the data entered by the property management company for the first building to obtain the number of rectified defects for the first building and the total number of defects for the first building.

[0056] In the above implementation, statistics are performed based on the property entry data to obtain the total number of qualified entry data corresponding to the first room, the total number of entry data corresponding to the first room, the number of rectified defects corresponding to the first building, and the total number of defects corresponding to the first building. The room qualification rate is calculated based on the ratio of the total number of qualified entry data corresponding to the first room and the total number of entry data corresponding to the first room, providing a data basis for subsequent housing risk prediction.

[0057] In some implementations, please refer to Figure 3 The property data entered includes the maximum and minimum measured net height of the building. The building construction quality index results include at least one of the following: room qualification rate, building rectification rate, net height range, and measurement deviation. Based on the preset building construction quality index processing rules, the property data entered is processed to obtain the building construction quality index results, which may include the following steps: S310. Based on the property entry data, perform statistics to obtain the total number of qualified entry data corresponding to the first room, the total number of entry data corresponding to the first room, the number of rectified defects corresponding to the first building, and the total number of defects corresponding to the first building.

[0058] S320. The building rectification rate is obtained by calculating the ratio between the number of rectified defects corresponding to the first building and the total number of defects corresponding to the first building.

[0059] Specifically, if the data anomaly detection result indicates that the property management's entered data is free of anomalies, it means that the measurement data meets the standards and quality requirements. However, some inspection data (such as whether there are wall cracks, leaks, or hollow areas) may still be substandard. Therefore, after confirming that the data anomaly detection result indicates that the property management's entered data is free of anomalies, statistics can be compiled based on the data entered for the first room to obtain the total number of qualified data entries for the first room and the total number of data entries for the first room. In addition, the inspection data can also record the status of defects. For example, if a defect is "existing wall crack," and the crack has been repaired in subsequent inspections, it is recorded as "not present," indicating that the defect has been rectified; if it still exists, it is recorded as "not rectified." Based on this, statistics can be compiled based on the data entered for the first building to obtain the number of rectified defects for the first building and the total number of defects for the first building. Then, the ratio of the number of rectified defects for the first building to the total number of defects for the first building is calculated to obtain the building rectification rate. For example, the building rectification rate = (number of rectified defects / total number of defects) × 100%.

[0060] In the above implementation, statistics are performed based on the property entry data to obtain the total number of qualified entry data corresponding to the first room, the total number of entry data corresponding to the first room, the number of rectified defects corresponding to the first building, and the total number of defects corresponding to the first building. The building rectification rate is calculated based on the ratio of the number of rectified defects corresponding to the first building to the total number of defects corresponding to the first building, providing a data basis for subsequent housing risk prediction.

[0061] In some implementations, the property data entered includes the maximum and minimum measured net height of the building, and the building engineering quality index results include at least one of the following: room qualification rate, building rectification rate, net height range, and measurement deviation. The property data entered is processed based on preset building engineering quality index processing rules to obtain the building engineering quality index results, which may include: calculating the difference between the maximum and minimum measured net height of the building to obtain the net height range.

[0062] Specifically, the data entered by the property management includes the maximum and minimum measured net height of the rooms. The difference between these two values ​​is calculated to reflect the uniformity of the room's net height, resulting in the net height range. This indicator is commonly used to assess floor levelness, construction accuracy, and the comfort of living spaces; a smaller range indicates a more uniform indoor net height. For example, if room A has a maximum measured net height of 2860mm and a minimum measured net height of 2840mm, then the net height range = maximum measured net height - minimum measured net height = 2860mm - 2840mm = 20mm.

[0063] In the above implementation method, the difference between the measured maximum and minimum net height of the building is calculated to obtain the net height range, which provides a data basis for subsequent building risk prediction.

[0064] In some implementations, the property data entered includes the maximum and minimum measured net height of the building, and the building engineering quality index results include at least one of the following: room qualification rate, building rectification rate, net height range, and measurement deviation. The property data entered is processed based on preset building engineering quality index processing rules to obtain the building engineering quality index results, which may include: calculating the difference between the property data entered and preset standard values ​​to obtain the measurement deviation.

[0065] Specifically, the difference between the property data entered and the preset standard values ​​is calculated to objectively reflect the degree of conformity between the actual construction or building status and the design standards, thus obtaining the measurement deviation to clarify the direction (positive / negative) and specific value of the deviation. For example, the measurement deviation corresponding to the net height = 2850mm - 2800mm = +50mm.

[0066] In the above implementation method, the difference between the property data and the preset standard value is calculated to obtain the measurement deviation, which provides a data basis for subsequent housing risk prediction.

[0067] In some implementations, risk prediction of housing project quality indicators based on a quality risk early warning model can be used to obtain housing risk prediction results. This may include: the quality risk early warning model includes a preset risk threshold, and the housing project quality indicator results are compared with the preset risk threshold to obtain housing risk prediction results.

[0068] Specifically, the quality risk early warning model includes preset risk thresholds. Since different indicators have their own measurement standards and risk characteristics, corresponding preset risk thresholds can be set for different housing project quality indicators. By comparing the housing project quality indicator results with the corresponding preset risk thresholds, the housing risk prediction results are obtained. It should be noted that users can customize multi-dimensional and hierarchical early warning thresholds according to actual management needs, thereby flexibly adapting to the control requirements of different projects and different stages.

[0069] For example, the quality indicators for housing construction projects can be: 1) Defect incidence rate, with a preset risk threshold of 5%. If the defect incidence rate exceeds 5%, a warning risk is indicated. 2) Rectification period, with a preset risk threshold of 3 days. If rectification is not completed within 3 days, a warning risk is indicated. 3) Inspection progress, with a preset risk threshold of 80%. If the inspection progress is below 80%, a warning risk is indicated. 4) Building pass rate, with a preset risk threshold of 75%. If the building pass rate is below 75%, a warning risk is indicated.

[0070] In the above embodiments, the quality risk early warning model includes a preset risk threshold. The results of housing engineering quality indicators are compared with the preset risk threshold to obtain housing risk prediction results, thereby identifying potential problems, supporting early intervention, and reducing quality risks. By integrating intelligent verification and risk prediction models, the transformation of property unit-by-unit inspection towards intelligence can be promoted, enhancing risk prevention and control capabilities.

[0071] In some implementations, risk prediction of housing project quality indicators based on a quality risk early warning model to obtain housing risk prediction results may include: extracting features from housing project quality indicators based on the quality risk early warning model to obtain indicator features, and then performing risk prediction based on the indicator features to obtain housing risk prediction results.

[0072] In some cases, a quality risk early warning model is constructed. The core of this model lies in the systematic analysis and feature extraction of various quality indicators of a building project. By deeply mining the quality indicator data, key features that reflect potential risks in the construction process and material usage of the building project are identified and extracted, thereby obtaining a set of indicator features.

[0073] Specifically, a quality risk early warning model is used to process housing project quality indicator data to extract features and obtain indicator features reflecting the quality status. Subsequently, based on the extracted indicator features, risk prediction algorithms (such as regression analysis and decision trees) are used to assess the quality risk of the housing project and obtain housing risk prediction results.

[0074] In the above implementation, feature extraction is performed on the quality indicators of housing projects based on a quality risk early warning model to obtain indicator features. Risk prediction is then performed based on these indicator features to obtain housing risk prediction results, thereby identifying potential problems, supporting early intervention, and reducing quality risks. By integrating intelligent verification and risk prediction models, the transformation of property unit-by-unit inspection towards intelligence can be promoted, enhancing risk prevention and control capabilities.

[0075] In some implementations, please refer to Figure 4 The method may also include the following steps: S410. If the data anomaly detection result indicates that there is an anomaly in the data entered by the property management, push an anomaly prompt message.

[0076] S420. Based on the abnormal prompt information, conduct inspection and rectification, and obtain the property data entered after rectification.

[0077] S430. Perform anomaly detection on the rectified property data, and repeat the above operation based on the anomaly detection results until the anomaly detection results indicate that there are no anomalies in the property data.

[0078] Specifically, when the data anomaly detection result indicates an anomaly in the property management's data entry, an anomaly alert will be automatically triggered. This alert may include the content of the abnormal data and its corresponding physical or logical location (e.g., a wall crack exists in the load-bearing wall of room A) to help responsible personnel quickly pinpoint the problem. After receiving the alert, the responsible personnel must verify the actual situation of the relevant data to determine if the anomaly is caused by measurement error or a mistake in the testing process. If it is confirmed that it is not a measurement or testing error, the responsible personnel should rectify the corresponding problem within the rectification period stipulated in the regulations and re-collect the data as the rectified property management data. If it is confirmed that the anomaly is caused by a measurement or testing error, the data should be re-collected as the rectified property management data. Then, the anomaly detection process is repeated on the rectified property management data. If the data anomaly detection result still indicates an anomaly, the above alert, rectification, and anomaly detection steps are repeated until the data anomaly detection result confirms that the data complies with the specifications and there are no anomalies. For example, the error message could be "The net height of 5000mm exceeds the reasonable range of 2500–3200mm. Please check."

[0079] In the above implementation, when the data anomaly detection result indicates that there is an anomaly in the property entry data, an anomaly prompt message is pushed, and inspection and rectification are carried out based on the anomaly prompt message. The rectified property entry data is then obtained, and anomaly detection is performed on the rectified property entry data. The above operation is repeated according to the data anomaly detection result until the data anomaly detection result indicates that there is no anomaly in the property entry data, so as to ensure data quality.

[0080] In some implementations, please refer to Figure 5 The method may also include the following steps: S510. When the housing risk prediction results indicate that there is a risk of early warning, generate and push early warning information.

[0081] S520: Track and process the early warning information, and update the early warning status corresponding to the early warning information according to the processing results.

[0082] Specifically, when a building risk prediction indicates a potential early warning risk, the system automatically generates corresponding push notifications based on the different types of early warning risks and sends them to relevant personnel through pre-configured push channels (such as app push notifications, SMS, and email). For example, if the early warning risk is determined based on the defect incidence rate, the notification will be sent to the project engineer; if it is determined based on the rectification deadline, the notification will be sent to the construction unit and supervisor; if it is based on the inspection progress, the notification will be sent to the head of the building inspector; and if it is determined based on the building's pass rate, the notification will be sent to the project quality manager. Upon receiving the notification, the responsible personnel must track and process it according to the content and fill in the processing results in the system. The system will automatically update the status of the early warning information based on the submitted processing results, such as changing "unprocessed" to "processed," forming a complete closed-loop management process from risk discovery to notification, processing, and status update. For example, the early warning information may include the early warning type, the building / room involved, the current value, the threshold, and recommended measures. For instance, if the hollow wall rate of Building 3 is 6%, exceeding the threshold of 5%, it is recommended to prioritize the rectification of the walls in this building.

[0083] In the above implementation, when the housing risk prediction results indicate the existence of a warning risk, a warning information is generated and pushed, the warning information is tracked and processed, and the warning status corresponding to the warning information is updated according to the processing results. This realizes automatic risk push and early intervention control, thereby improving the efficiency of risk identification and response and enhancing the overall effectiveness of housing safety management.

[0084] In some implementations, the method may further include generating an analysis report based on the results of housing construction quality indicators.

[0085] Specifically, based on the results of housing construction quality indicators (including inspection progress, pass rate, rectification rate, and defect incidence rate), detailed analysis reports can be automatically generated. Users can customize the time period as needed to generate reports at different frequencies, such as daily, weekly, and monthly reports. It should be noted that the number of times the property unit-by-unit inspection method is implemented may vary within a specified time period, which means that the data and analysis results included in the report will be dynamically adjusted according to the actual situation.

[0086] In some implementations, the analysis report may include various charts, such as trend graphs and comparison charts, to visually present changes in indicators over a selected time range. For example, a weekly report may display the pass rate trend over the past seven days, a comparison of defect rates for each building, etc., to identify potential problems and make assessments. The report may also include a risk summary section, which lists the total number of warnings for the period, the number of warnings that have been processed, and a list of unprocessed risks, so that managers can follow up in a timely manner and take countermeasures.

[0087] It should be noted that, for ease of use and information sharing, the system supports exporting analysis reports to PDF or Excel format and allows for direct sharing with management within the system. For example, it can be set to automatically send weekly reports to the project's main email address regularly, ensuring that relevant personnel receive the latest information in a timely manner.

[0088] In the above implementation method, an analysis report is generated based on the results of housing project quality indicators, which improves the efficiency, accuracy and risk control capabilities of project management and provides strong support for ensuring the quality of housing projects.

[0089] This specification provides a property unit inspection device 600. Please refer to [link / reference]. Figure 6 The property unit inspection device 600 includes: an anomaly detection module 610, an indicator processing module 620, and a risk prediction module 630.

[0090] The anomaly detection module 610 is used to perform anomaly detection on the property data entered and obtain the data anomaly detection results. The indicator processing module 620 is used to process the property data based on preset housing engineering quality indicator processing rules when the data anomaly detection result indicates that there is no anomaly in the property data, so as to obtain the housing engineering quality indicator result. The risk prediction module 630 is used to predict the risk of the housing project quality indicators based on the quality risk early warning model, and obtain the housing risk prediction result.

[0091] For a detailed description of the property unit inspection device, please refer to the description of the property unit inspection method above, which will not be repeated here.

[0092] In some embodiments, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a property unit inspection method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0093] Those skilled in the art will understand that Figure 7 The structures shown are merely block diagrams of some structures related to the solutions disclosed in this specification, and do not constitute a limitation on the computer device to which the solutions disclosed in this specification are applied. Specifically, the computer device may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements.

[0094] In some embodiments, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method steps described above.

[0095] This specification provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0096] One embodiment of this specification provides a computer program product including instructions that, when executed by a processor of a computer device, enable the computer device to perform the steps of the method described in any of the above embodiments.

[0097] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

Claims

1. A method for inspecting individual units in a property, characterized in that, The method includes: Anomaly detection is performed on the data entered by the property management company to obtain the anomaly detection results. If the data anomaly detection result indicates that there is no anomaly in the property data, the property data is processed according to the preset housing engineering quality index processing rules to obtain the housing engineering quality index result. Based on the quality risk early warning model, the risk prediction results of the building project quality indicators are obtained.

2. The method according to claim 1, characterized in that, The anomaly detection of the property entry data, and the resulting data anomaly detection results, include: Anomaly detection is performed on the property data entered based on a preset abnormal data verification mechanism to obtain the data anomaly detection result.

3. The method according to claim 2, characterized in that, The anomaly detection of the property data entered based on the preset anomaly data verification mechanism includes at least one of the following: Anomaly detection is performed on the property data entered based on preset reasonable rules; Anomaly detection is performed on the property data entered based on a preset reasonable threshold range; The property data entry is anomaly detected based on a preset correlation deviation, where the preset correlation deviation represents the range of deviation between the property data entry and the estimated value.

4. The method according to claim 1, characterized in that, The property data entered includes the maximum and minimum measured net height of the building. The building construction quality index results include at least one of the following: room pass rate, building rectification rate, net height range, and measurement deviation. The property data entered is processed based on preset building construction quality index processing rules to obtain the building construction quality index results, including: Based on the property entry data, statistics are performed to obtain the total number of qualified entry data corresponding to the first room, the total number of entry data corresponding to the first room, the number of rectified defects corresponding to the first building, and the total number of defects corresponding to the first building. The room pass rate is calculated by comparing the total number of qualified data entries corresponding to the first room with the total number of data entries corresponding to the first room. The building rectification rate is calculated by comparing the number of rectified defects in the first building with the total number of defects in the first building. The difference between the measured maximum and minimum net height of the building is calculated to obtain the net height range. The measurement deviation is obtained by calculating the difference between the property data entered and the preset standard value.

5. The method according to claim 1, characterized in that, The risk prediction results of the building project quality indicators based on the quality risk early warning model are obtained, including: The quality risk early warning model includes a preset risk threshold. The results of the building construction quality indicators are compared with the preset risk threshold to obtain the building risk prediction result; or Based on the quality risk early warning model, feature extraction is performed on the quality index results of the housing project to obtain index features. Based on the index features, risk prediction is performed to obtain the risk prediction result of the housing project.

6. The method according to claim 1, characterized in that, The method further includes: If the data anomaly detection result indicates that the property entry data is abnormal, an anomaly alert message will be pushed. Based on the aforementioned abnormality alerts, conduct inspections and rectifications, and obtain the rectified property data entered into the system. Anomaly detection is performed on the rectified property data, and the above operation is repeated based on the anomaly detection results until the anomaly detection results indicate that there are no anomalies in the property data.

7. The method according to claim 1, characterized in that, The method further includes: When the housing risk prediction results indicate the presence of a warning risk, a warning message is generated and pushed out. The warning information is tracked and processed, and the warning status corresponding to the warning information is updated according to the processing result.

8. The method according to claim 1, characterized in that, The method further includes: An analysis report is generated based on the results of the aforementioned housing project quality indicators.

9. A property unit inspection device, characterized in that, The device includes: The anomaly detection module is used to detect anomalies in the property data and obtain the data anomaly detection results. The indicator processing module is used to process the property data based on preset housing engineering quality indicator processing rules when the data anomaly detection result indicates that there is no anomaly in the property data, and obtain the housing engineering quality indicator result. The risk prediction module is used to predict the risk of the building project quality indicators based on the quality risk early warning model, and obtain the building risk prediction result.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images