An evaluation method and device for digital transformation of property equipment, electronic equipment, medium and program product

Abstract

The application belongs to the technical field of data processing, and specifically discloses an evaluation method and device for property equipment digitization reconstruction, electronic equipment, medium and program product, which comprises the following steps: obtaining equipment operation data and property management process data and preprocessing; extracting equipment events containing equipment ID, event type and timestamp, and process data events containing process ID, event type, timestamp and text content; calculating the time alignment, semantic consistency and spatial consistency of the two within a preset time window; analyzing the closed-loop accessibility and interoperability; and obtaining the process coupling degree coefficient based on the above indexes and grading. The application overcomes the limitation of the existing evaluation method which only focuses on equipment hardware, comprehensively considers the adaptability of equipment and management process, and provides a scientific decision basis for digitization reconstruction.

Description

Technical Field

[0001] This invention belongs to the field of data processing technology, and specifically relates to an evaluation method, device, electronic equipment, medium and program product for the digital transformation of property equipment. Background Technology

[0002] With the development of smart property management, property management companies are increasingly attempting to digitally upgrade existing equipment, such as elevators, central air conditioning, water pump rooms, and power distribution systems, to achieve remote monitoring, energy consumption optimization, and intelligent operation and maintenance. However, existing digital upgrade evaluation methods mainly focus on the hardware conditions of the equipment itself, such as whether it has sensor interfaces, network coverage, and system compatibility, neglecting the compatibility issues between digital upgrades and property management processes. Summary of the Invention

[0003] To address this, the present invention provides an evaluation method, apparatus, electronic device, medium, and computer program product for the digital transformation of property equipment, thereby solving the aforementioned technical problems.

[0004] This invention provides an evaluation method for the digital transformation of property equipment, comprising the following steps:

[0005] Step S1: Obtain first data and preprocess the first data, the first data including equipment operation data and property management process data;

[0006] Step S2: Extract device events and process data events from the first data respectively. The device events include device ID, first event type and first timestamp. The process data events include process ID, second event type, second timestamp and text content.

[0007] Step S3: Within a preset time window, calculate the time alignment, semantic consistency, and spatial consistency between device events and process events;

[0008] Step S4: Analyze the closed-loop reachability and interoperability between device events and process events. The closed-loop reachability is at least based on the process completion rate of device events within the service level agreement time limit and whether the device status has returned to normal. The interoperability is at least based on the field mapping success rate of device data and process data, the completeness of key fields, and the interface call success rate.

[0009] Step S5: Based on the time alignment, semantic consistency, spatial consistency, closed-loop reachability and interoperability, the process coupling coefficient is obtained by weighted fusion with preset weights, and the process is classified based on the coupling coefficient.

[0010] In another aspect, this application also provides an assessment device for the digital transformation of property equipment, comprising:

[0011] The first data acquisition module is used to acquire first data and preprocess the first data, the first data including equipment operation data and property management process data;

[0012] The event extraction module is used to extract device events and process data events from the first data, respectively. The device events include a device ID, a first event type, and a first timestamp. The process data events include a process ID, a second event type, a second timestamp, and text content.

[0013] The first calculation module is used to calculate the time alignment, semantic consistency, and spatial consistency between device events and process events within a preset time window.

[0014] The first analysis module is used to analyze the closed-loop reachability and interoperability between device events and process events. The closed-loop reachability is at least based on the process completion rate of device events within the service level agreement time limit and whether the device status has returned to normal. The interoperability is at least based on the field mapping success rate of device data and process data, the completeness of key fields, and the interface call success rate.

[0015] The grading module is used to obtain the process coupling coefficient by weighted fusion with preset weights based on the time alignment, semantic consistency, spatial consistency, closed-loop reachability and interoperability, and to perform grading based on the coupling coefficient.

[0016] In another aspect, this application also provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform an evaluation method for digital transformation of property equipment as described above.

[0017] In another aspect, this application provides a computer-readable storage medium having stored thereon computer program instructions that can be executed by a processor to implement an evaluation method for the digital transformation of property equipment as described above.

[0018] Another aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements an evaluation method for the digital transformation of property equipment as described above.

[0019] This invention acquires and preprocesses equipment operation and property management process data, extracts event information, and calculates and grades the process coupling coefficient from multiple dimensions such as time alignment, semantic consistency, spatial consistency, closed-loop accessibility, and interoperability. This overcomes the limitations of existing assessments that only focus on equipment hardware conditions, comprehensively considers the compatibility between equipment data and management processes, provides a scientific basis for digital transformation decisions, and helps improve the rationality and effectiveness of transformation. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0022] Figure 1 This is a flowchart of an evaluation method for digital transformation of property equipment, provided as an embodiment of the present invention.

[0023] Figure 2 A flowchart for identifying the second event type provided in an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of the structure of an evaluation device for the digital transformation of property equipment, provided in an embodiment of the present invention.

[0025] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] This application proposes an evaluation method for the digital transformation of property equipment. The technical solution of this application will be described in detail below with reference to various embodiments.

[0028] like Figure 1As shown, this embodiment of the invention discloses an evaluation method 100 for the digital transformation of property equipment, comprising the following steps:

[0029] Step S1: Obtain first data and preprocess the first data, which includes equipment operation data and property management process data.

[0030] In some embodiments, equipment operation data includes at least equipment alarm logs, status changes, and energy consumption curves, while property management data includes at least work orders, complaint records, and fee anomalies. The preprocessing of the first data includes: time base unification, location encoding standardization, and text synonym normalization.

[0031] Specifically, regarding equipment operation data, for example, for equipment alarm logs, such as trigger records when equipment failure occurs, including fault codes (such as the code "E01" corresponding to elevator entrapment), alarm trigger time, fault level (emergency / general), etc.

[0032] For status changes, such as the switching information of equipment operating status, such as the time point and status description of an elevator changing from "normal operation" to "stop service" or a water pump changing from "automatic mode" to "manual mode".

[0033] For example, energy consumption curves can be energy consumption data recorded at fixed time granularity (such as every 10 minutes), including electricity, water, gas, etc., reflecting changes in equipment load (such as the peak energy consumption of central air conditioning between 8:00 and 18:00 on weekdays).

[0034] Specifically, regarding property management data, for example, for work orders, such as property operation and maintenance task records, there is work order number, creation time, task type (repair / inspection), associated equipment, handling personnel, completion status, etc.

[0035] For complaint records, such as issues reported by homeowners through channels like the APP, telephone, and WeChat, the complaint content (e.g., "abnormal noise from the elevator in Building 3"), submission time, area involved, and processing result are included.

[0036] For abnormal charges, such as abnormal data related to equipment usage, like incorrect calculation of central air conditioning usage fees or a sudden increase in shared electricity costs, the data includes the time of occurrence of the abnormality, the amount deviation, and the associated user information.

[0037] The preprocessing of the first data, specifically, involves unifying the time base. For example, to address the inconsistency of time sources across different systems, the timestamps of all data can be unified to the same time zone (e.g., Beijing time).

[0038] For example, for device operation data, if the original timestamp is UTC time, it is converted to Beijing time (+8 hours); if the device's local clock has a deviation (such as a difference of more than 5 minutes from the standard time), it is calibrated through an NTP server or corrected by comparing with system logs;

[0039] For example, for property management process data, a unified timestamp is used based on the "event trigger time": the work order takes the "creation time", the complaint record takes the "owner submission time", and the fee exception takes the "time when the system discovered the exception", to ensure the uniqueness of the time field.

[0040] Regarding the standardization of location coding, for example, based on a multi-level principle of regional scope, a unique spatial identifier is assigned to all data. For instance, the coding rules are as follows:

[0041] The system uses a five-level hierarchical coding system: "Park-Building-Floor-Area-Equipment" (e.g., "02-05-03-01-08"). The first two digits, "02", represent the park number (e.g., Park No. 2); the 3rd and 4th digits, "05", represent the building (Building No. 5); the 5th and 6th digits, "03", represent the floor (3rd floor); the 7th and 8th digits, "01", represent the area (public area); and the last two digits, "08", represent the equipment number (Air Conditioner No. 8).

[0042] Regarding text synonym normalization, for example, it unifies the semantic expression of text descriptions in process data, eliminates differences between synonyms, and for instance, constructs a thesaurus for the property management field, including synonymous expressions for scenarios such as equipment malfunctions and process statuses. For example, malfunction descriptions such as "elevator stuck," "elevator trapped," and "elevator out of service and trapped" are unified as "elevator trapped malfunction"; status descriptions such as "work order resolved," "problem resolved," and "task completed" are unified as "process closed loop completed." For text content such as complaint records and work order notes, synonyms are replaced through dictionary matching to ensure consistent textual expressions for the same type of event (e.g., "air conditioner not heating" is unified as "air conditioner heating abnormal").

[0043] Step S2: Extract device events and process data events from the first data respectively. The device events include device ID, first event type and first timestamp. The process data events include process ID, second event type, second timestamp and text content.

[0044] In some embodiments, the goal of device event extraction is to extract structured information, including device ID, first event type, and first timestamp, from preprocessed device operation data to reflect the dynamics of the device.

[0045] Specifically, for the device ID, for example, it is a unique identifier directly associated with the device based on the standardized location code in step S1. For example, the location code "02-05-03-01-08" corresponds to "Air conditioner No. 8 in the public area on the 3rd floor of Building 5 in Park 2". This code is the device ID, ensuring that each device event can be traced back to the specific device.

[0046] Specifically, the first event type, for example, covers changes in the operating state of the device. A first event type system is preset based on the characteristics of the device operating data and automatically extracted through rule matching. Specifically, the preset type system includes, for example:

[0047] Alarms: such as "elevator entrapment alarm", "air conditioner temperature exceeding limit alarm", "water pump pressure abnormal alarm"; Status change alarms: such as "elevator running → stopped", "power distribution system closed → opened"; Energy consumption anomalies: such as "sudden increase in central air conditioning energy consumption" and "excessive energy consumption of water supply pump room at night".

[0048] For automatic extraction through rule matching, for example, for equipment alarm logs, the event type is mapped according to the fault code (e.g., fault code "E01" corresponds to "elevator entrapment alarm"); for state change data, the type is determined according to the description before and after the state switch (e.g., "running → shutdown" corresponds to "equipment shutdown event"); for energy consumption curves, they are marked as "abnormal energy consumption events" by judging by threshold (e.g., exceeding the historical average by 20%).

[0049] For example, the extraction of the first timestamp can be achieved by directly using the device event trigger time after unifying the time base in step S1, such as "2024-06-10 09:15:30" in the device alarm log or "2023-06-10 09:18:22" in the status change log, to ensure the accuracy of the time information.

[0050] In some embodiments, the goal of process data event extraction is to extract structured information, including process ID, second event type, second timestamp, and text content, from preprocessed property management process data to reflect elements of the management process.

[0051] Specifically, regarding the generation of process IDs, for example, a unique identifier is assigned to each process event, according to the following rules:

[0052] Work order: Use the format "WO-Date-Serial Number" (e.g., "WO-20240610-001", where WO represents work order).

[0053] Complaint records: Use the format "CP-Date-Serial Number" (e.g., "CP-20240610-002", where CP represents the complaint).

[0054] Abnormal Charge: Adopt the format of "FE-Date-Sequence Number" (e.g., "FE-20240610-003", where FE represents abnormal charge);

[0055] The process ID needs to be associated with the location code in step S1 (e.g., "WO-20240610-001" is bound to "02-05-03-01-08", indicating a work order for this device).

[0056] Among them, regarding the extraction of the second timestamp, for example, based on the trigger time of the process event, directly adopt the unified timestamp in step S1. Exemplarily, for the process work order, take the "creation time" (e.g., "2023-06-10 09:20:15"); for the complaint record, take the "owner submission time" (e.g., "2023-06-10 09:10:00"); for the abnormal charge, take the "system discovery time" (e.g., "2023-06-10 08:30:00").

[0057] Among them, regarding the extraction of text content, for example, retain the text description in the process data for semantic analysis. Exemplarily, for the work order text, retain "The air conditioner in Room 8, 3rd Floor, Building 5 is not cooling and needs to be repaired"; for the complaint text, retain "The heating of the home air conditioner is insufficient, and the room temperature is only 16°C"; preferably, the text content is processed by "text synonym normalization" in step S1 (e.g., "not cooling" is normalized to "refrigeration anomaly").

[0058] Among them, for the recognition of the second event type, it is automatically classified based on a pre-trained first sentence vector model (such as RoBERTa, ALBERT, etc.).

[0059] Preferably, preprocess the text content, input the preprocessed text into the first sentence vector model, extract the sentence vector representation corresponding to the text, and based on a preset standard sentence vector and event type mapping library, calculate the cosine similarity between the text vector to be recognized and each standard sentence vector, and output the event type corresponding to the standard sentence vector with the highest similarity.

[0060] Specifically, exemplarily, as Figure 2 shown, the recognition of the second event type includes the following steps:

[0061] Step S201, text content preprocessing. Specifically, clean and standardize the original text in the process data (such as work order description, complaint content); exemplarily, it includes:

[0062] Text cleaning: Remove meaningless characters (such as special symbols "@, #" and redundant spaces), filter stop words (such as虚词 that do not affect semantics like "de, le, zai", etc.), and retain the core information;

[0063] Synonym unification: Based on the property domain thesaurus built in step S1, replace synonymous expressions in the text. For example, replace "elevator stuck" and "elevator trapped" with "elevator trapped"; replace "not cooling" and "poor cooling effect" with "abnormal cooling".

[0064] Length standardization: Truncate excessively long texts (preserving the core semantic parts), and supplement excessively short texts (e.g., only 1-2 words) with contextual information (e.g., supplement with "[device type]" by combining positional encoding) to ensure that the length of the model input meets the requirements of the pre-trained model.

[0065] Step S202: The pre-trained first-sentence vector model is used to convert the preprocessed text into high-dimensional vectors. For example, based on the characteristics of property management process text—primarily short sentences, containing numerous equipment and service terms, and with relatively fixed semantic scenarios—RoBERTa-base is selected as the first-sentence vector model. For example, the workflow based on RoBERTa-base includes the following steps:

[0066] First, the RoBERTa-base model is loaded. Specifically, the pre-trained weight file and word segmenter are loaded (for example, using HuggingFace's transformers library to load the RoBERTa-base model and its corresponding word segmenter). The weight file contains the parameters obtained by training the model on a large amount of general text (such as books and web pages), recording language rules (for example, "repair" and "repair" are semantically similar). The word segmenter has a built-in predefined vocabulary (such as "elevator" and "work order"), which is responsible for converting the text into a sequence of sub-words that the model can recognize.

[0067] In practice, publicly available resources can be accessed directly through open-source tools (such as the HuggingFace ecosystem), eliminating the need for training from scratch;

[0068] Next, the preprocessed text is transformed into the model input format. Specifically, firstly, the word segmenter splits the preprocessed text into subwords according to predefined rules, and then splits the text into subwords (e.g., "elevator trapped" is split into "electric / elevator / trapped / person" or subwords predefined by the model); secondly, [CLS] (used to extract sentence vectors) and [SEP] (sentence separator) are added to the beginning and end of the text respectively; then, the subwords are mapped to unique IDs in the model vocabulary, and padding is added to ensure uniform length.

[0069] Finally, the model outputs semantic vectors. Specifically, the encoded text is input into the pre-trained RoBERTa-base model, which analyzes the text layer by layer through a multi-layer neural network (e.g., a 12-layer Transformer). The model outputs vector representations at each position, where the output vector corresponding to the [CLS] marker has been aggregated with the semantics of the whole sentence, and the output vector corresponding to the [CLS] position serves as the sentence vector representation of the entire text.

[0070] Step S203: Construct a standard sentence vector and event type mapping library. Specifically, firstly, based on the property management scenario, preset a second event type, including but not limited to:

[0071] Equipment maintenance requests: procedures involving equipment malfunction repair (e.g., "elevator entrapment repair" or "air conditioning cooling malfunction handling").

[0072] Service quality complaints: Feedback on the property management's response efficiency and service attitude (e.g., "complaint about untimely repairs" or "poor attitude of staff").

[0073] Feedback on abnormal charges: Charges related to equipment (e.g., "incorrect calculation of central air conditioning costs" or "abnormal shared electricity charges").

[0074] Routine inspection tasks: planned equipment inspections (e.g., monthly elevator safety inspections, quarterly water pump maintenance).

[0075] Next, select 3-5 typical texts for each event type as "standard samples". For example, the standard texts for "equipment maintenance request" include "elevator malfunction requires emergency repair" and "air conditioner not cooling, please handle".

[0076] Finally, the standard text is processed according to the above process to generate corresponding sentence vectors, and a "standard sentence vector-event type" mapping library is constructed (e.g., dictionary structure: {type1:[vector1,vector2],type2:[vector3,vector4]...}).

[0077] Step S204: Determine the event type of the text to be identified by comparing vector similarity. Specifically, firstly, calculate the cosine similarity (range 0~1, the higher the value, the closer the semantics) between the sentence vector of each text to be identified and all standard sentence vectors in the mapping library.

[0078] Next, the average similarity between the text to be identified and all standard vectors of a certain type of event is taken as the matching score for that type; the type with the highest score is selected as the final classification result.

[0079] Step S3: Within a preset time window, calculate the time alignment, semantic consistency, and spatial consistency between device events and process events.

[0080] The time window serves as a benchmark for determining whether there is a correlation between equipment events and process events, and can be dynamically set based on the urgency of the event. Specifically, for emergency events (e.g., elevator entrapment, power distribution failure): the time window Δt is set based on the "emergency response time limit" in the SLA, ensuring that events requiring rapid response are included in the matching range; for general events (e.g., abnormal air conditioning temperature, water supply pressure fluctuations): the time window Δt can be set to 24~72 hours (allowing for normal processing cycles); for low-priority events (e.g., slight fluctuations in energy consumption curves): the time window can be relaxed to 7 days. It is understood that the specific values ​​of the above time windows are only examples, and this invention does not limit them.

[0081] In some embodiments, time alignment is used to evaluate the matching efficiency of device events and process events in the time dimension. Specifically, the delay time of the matched event pairs is calculated and its distribution is statistically analyzed to obtain the delay distribution. The hit rate is obtained by calculating the proportion of the number of device events that match the process events within a preset time to the total number of all device events. The time alignment is obtained by weighting and fusing the delay distribution and the hit rate with preset weights.

[0082] Specifically, within a preset time window, candidate matching pairs of device events and process events are filtered based on the same location code or adjacent areas. For example, the "cooling abnormality alarm" of device ID "02-05-03-01-08" (air conditioner No. 8 on the 3rd floor of Building 5) is matched with the "air conditioner repair work order" with the same location code;

[0083] For the selected matching pairs, calculate the time delay ΔT = process event timestamp (second timestamp) - device event timestamp (first timestamp), and statistically analyze the distribution characteristics of the delay, exemplarily including:

[0084] Average delay: the arithmetic mean of all ΔT values ​​(reflecting the overall response speed);

[0085] Median delay: Sort ΔT and take the median value (to avoid the influence of extreme values);

[0086] 90th percentile delay: 90% of event pairs have a delay of no more than this value (reflecting response efficiency in most cases);

[0087] Meanwhile, for the selected matching pairs, the hit rate is calculated as follows: (Number of device events that match the process event within the time window ÷ Total number of device events) × 100%.

[0088] Finally, weights are assigned to the delay distribution and the hit rate (e.g., each accounting for 50%), and the weighted average is used to obtain Ts (with a value of 0 to 1). The shorter the delay and the higher the hit rate, the closer Ts is to 1.

[0089] In one embodiment, semantic consistency measures the semantic correlation between device events and process events through text vector similarity. Preferably, semantic consistency is obtained based on a pre-trained second sentence vector model. Specifically, a standard description is generated based on the device event type and input into the second sentence vector model to obtain a device event text vector. The text content is input into the second sentence vector model to obtain a process event text vector. The cosine similarity between the device event text vector and the process time text vector is calculated, and the semantic consistency is obtained based on the cosine similarity.

[0090] Specifically, firstly, the text of equipment events and process events is standardized. For example, for equipment event text, a standard description is generated based on the first event type (e.g., the text corresponding to "air conditioner cooling abnormal alarm" is "air conditioner has cooling abnormality fault"); for process event text, the preprocessed text extracted in step S2 is used (e.g., the work order description "Air conditioner No. 8 on the 3rd floor of Building 5 is not cooling and needs maintenance").

[0091] Next, the second vector model, specifically the Sentence-BERT model, transforms the two types of text into high-dimensional vectors. The device event text vector reflects the core semantics of device failure (e.g., "cooling abnormality" or "equipment type is air conditioner"); the process event text vector reflects the core semantics of process request (e.g., "repair request" or "involves air conditioning cooling problem").

[0092] Finally, semantic consistency (Ss) is calculated. For example, semantic consistency is equal to the cosine similarity (value 0~1) between the device event text vector and the process event text vector. The closer the similarity is to 1, the more consistent the semantics are; the less than 0.5 the similarity is, the weaker the semantic association.

[0093] In one embodiment, Sentence-BERT is a variant of the BERT model optimized for sentence vector generation, whose structure is based on BERT with the addition of a sentence vector output layer, exemplarily including:

[0094] Pre-trained BERT: Employs a multi-layer Transformer encoder (e.g., the base version contains 12 layers) to obtain contextual semantic features of the text, supporting input of single sentences or sentence pairs;

[0095] Sentence vector output layer: The output of BERT is aggregated. For example, for a single sentence, the hidden state corresponding to the [CLS] label is taken as the sentence vector; for sentence pairs, a comprehensive representation is generated by averaging, max pooling or concatenating sentence pair vectors, and finally a fixed-dimensional sentence vector is output.

[0096] In one embodiment, the pre-training process of the Sentence-BERT model includes, for example:

[0097] Construct a sentence pair dataset for the property management domain, including: positive examples: semantically related device event text and process event text (e.g., "air conditioner cooling malfunction alarm" and "air conditioner not cooling repair request"); negative examples: semantically unrelated text pairs (e.g., "elevator entrapment alarm" and "water bill anomaly work order").

[0098] The loss function can be set, for example, by using triplet loss, which optimizes the semantic discriminativeness of sentence vectors by bringing positive example pairs closer together and increasing the distance between negative example pairs.

[0099] Input text into the model and adjust the top-level parameters of BERT through backpropagation to make semantically similar text vectors closer in high-dimensional space;

[0100] The model performance is verified by calculating cosine similarity (e.g., positive example similarity ≥ 0.8, negative example similarity ≤ 0.3), and the training data and hyperparameters are iteratively adjusted to finally obtain the pre-trained Sentence-BERT model in this embodiment.

[0101] In some embodiments, spatial consistency is based on standardized location coding to determine whether device events and process events point to the same spatial range, wherein the location coding forms a multi-level coding according to the regional range, and the judgment is made level by level according to the level.

[0102] For example, based on the standardized multi-level location coding in step S1, it is determined whether the device event and the process event point to the same spatial range. Specifically, for example, the location coding is matched level by level from high to low according to a five-level structure of "campus-building-floor-area-equipment" (e.g., "02-05-03-01-08"); a weight score (value 0~1) is assigned according to the matching level.

[0103] Step S4: Analyze the closed-loop reachability and interoperability between device events and process events. The closed-loop reachability is at least based on the process completion rate of device events within the service level agreement time limit and whether the device status has returned to normal. The interoperability is at least based on the field mapping success rate of device data and process data, the completeness of key fields, and the interface call success rate.

[0104] In some embodiments, regarding closed-loop accessibility, for example, closed-loop completion time limits are set for different types of device events based on the service standards (Service Level Agreement SLA time limits) agreed upon by the property management and the owner, such as:

[0105] Emergency events (such as elevator entrapment, power distribution failure): SLA time limit is 2-4 hours (immediate response and resolution are required);

[0106] For general events (such as abnormal air conditioning temperature or insufficient water supply pressure): the SLA time limit is 24 to 48 hours (normal processing cycle is allowed).

[0107] Low-priority events (e.g., slight fluctuations in the energy consumption curve): SLA timeframe extended to 7 days (can be planned).

[0108] Among these, the closed-loop status of process events triggered by device events is verified in stages, for example including:

[0109] Creation phase: Confirm whether process events (such as maintenance work orders) are initiated after an equipment event occurs;

[0110] Processing phase: Check whether the work order has been assigned to maintenance personnel and whether there is a clear processing record (e.g., "on-site inspection has been carried out").

[0111] Completion phase: Verify whether the work order is marked as "processed and completed" and includes the resolution result (e.g., "air conditioning cooling has returned to normal").

[0112] Follow-up phase: Confirm whether to conduct a satisfaction survey with the homeowner (e.g., "The homeowner reported that the problem has been resolved").

[0113] The process completion rate is calculated as follows: (Number of device events that complete the entire process loop within the SLA timeframe ÷ Total number of device events) × 100%.

[0114] Next, after the loop is closed, the status is confirmed to have returned to normal by checking the equipment operation data. For example, this includes:

[0115] For alarm-related events: check whether the equipment alarm has been cleared (e.g., the elevator entrapment alarm has been cleared).

[0116] For status change events: Verify whether the device has returned to normal operating status (e.g., switching from "out of service" to "operating");

[0117] For abnormal energy consumption events: confirm whether the energy consumption curve has returned to the normal range (e.g., the energy consumption of central air conditioning has dropped to the level of the same period in history).

[0118] Wherein, the state recovery rate = (number of events in which the equipment state returns to normal after the loop is closed ÷ number of events in which the loop is closed) × 100%;

[0119] Finally, the comprehensive calculation of closed-loop reachability (Cs): Cs is the weighted average of process completion rate and state recovery rate (each accounting for 50%), with a value range of 0 to 1.

[0120] In one embodiment, regarding interoperability, for example, interoperability measures the ability of a device data system to interact with a property management system, including, for example, three metrics: field mapping success rate, interface call success rate, and key field completeness.

[0121] Specifically, regarding field mapping success rate, for example, evaluating the accuracy of matching key fields between device data and process data includes:

[0122] Preset fields to be mapped: Device ID and work order associated device ID, event timestamp and work order creation time, location code and work order location, fault type and work order problem description, etc.; count the number of successfully matched fields (e.g., 4 out of 5 key fields are successfully matched); field mapping success rate = (number of successfully matched fields ÷ total number of fields to be mapped) × 100%.

[0123] For the interface call success rate, for example, the interface types include the device system pushing alarm information to the property system (triggering work order creation), the property system querying the real-time status of the equipment (e.g., viewing equipment parameters during maintenance), and data synchronization between systems (e.g., work order status feedback to the device system); the total number of interface calls and the number of successful calls are counted (e.g., 100 calls, 92 successful); the interface call success rate = (number of successful calls ÷ total number of calls) × 100%.

[0124] Regarding the completeness of key fields, for example, verifying whether the key information related to device events in the process data is complete, specifically including:

[0125] Key information includes: equipment model, specific location of the fault (e.g., "button malfunction inside the elevator car"), qualifications of the personnel handling the fault (e.g., "holding an elevator maintenance certificate"), fault handling plan, etc.; the number of process events with complete information (e.g., 85 out of 100 work orders have complete information); key field completeness = (number of process events with complete information ÷ total number of process events) × 100%.

[0126] Finally, the comprehensive calculation of interoperability (Is): Is is the weighted sum of the above three indicators (the weights can be set according to the system characteristics, such as field mapping 0.4, interface call 0.3, and field completeness 0.3), with a value range of 0 to 1.

[0127] Step S5: Based on the time alignment, semantic consistency, spatial consistency, closed-loop reachability and interoperability, the process coupling coefficient is obtained by weighted fusion with preset weights, and the process is classified based on the coupling coefficient.

[0128] In some embodiments, the preset weights can be determined using expert scoring or the Analytic Hierarchy Process (AHP). For example,

[0129] Time alignment (Ts): Weight w1=0.2 (emphasizing the timeliness of event response, which directly affects service efficiency);

[0130] Semantic consistency (Ss): weight w2=0.2 (reflects the matching accuracy of the text description);

[0131] Spatial consistency (Ls): weight w3 = 0.15 (ensures the spatial accuracy of event associations);

[0132] Closed-loop reachability (Cs): Weight w4 = 0.25 (the closed-loop process is the core of service delivery, and has the highest weight).

[0133] Interoperability (Is): Weight w2=0.2 (System interoperability is the foundation of data flow and affects overall coupling efficiency).

[0134] The total weight is 1 (0.2+0.2+0.15+0.25+0.2=1). It should be noted that each weight can be flexibly adjusted according to the actual needs of the property (for example, if more attention is paid to response speed, the weight of Ts will be increased). This invention does not limit this adjustment.

[0135] Next, based on the five indicators obtained in the previous steps (all standardized to values ​​between 0 and 1), a weighted sum is calculated according to preset weights. The calculation formula can be simplified to:

[0136]

[0137] Finally, based on the calculated CPC value, the system is classified according to a preset threshold. Preferably, modification suggestions can also be output based on the classification. For example, CPC ≥ 0.75 indicates high coupling, which can be directly modified; 0.50 ≤ CPC < 0.75 indicates medium coupling, which is recommended to be optimized before modification; and CPC < 0.50 indicates low coupling, which requires process reengineering or system modification first.

[0138] Figure 3 An assessment device 300 for the digital transformation of property equipment is shown. This device embodiment is similar to... Figure 1 Corresponding to the illustrated method embodiments, this device can be specifically applied to various electronic devices. Specifically, it includes:

[0139] The first data acquisition module 301 is used to acquire first data and preprocess the first data, the first data including equipment operation data and property management process data;

[0140] The event extraction module 302 is used to extract device events and process data events from the first data respectively. The device events include a device ID, a first event type, and a first timestamp. The process data events include a process ID, a second event type, a second timestamp, and text content.

[0141] The first calculation module 303 is used to calculate the time alignment, semantic consistency, and spatial consistency between device events and process events within a preset time window.

[0142] The first analysis module 304 is used to analyze the closed-loop reachability and interoperability between device events and process events. The closed-loop reachability is at least based on the process completion rate of the device event within the service level agreement time limit and whether the device status has returned to normal. The interoperability is at least based on the field mapping success rate of device data and process data, the completeness of key fields, and the interface call success rate.

[0143] The grading module 305 is used to obtain the process coupling coefficient by weighted fusion with preset weights based on the time alignment, semantic consistency, spatial consistency, closed-loop reachability and interoperability, and to perform grading based on the coupling coefficient.

[0144] Based on the same inventive concept, this application also provides an electronic device. The method corresponding to the electronic device can be the method in the foregoing embodiments, and its problem-solving principle is similar to that method. The electronic device provided in this application includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the methods and / or technical solutions of the foregoing embodiments of this application.

[0145] The electronic device can be a user device, or a device formed by integrating user devices and network devices through a network, or it can be an application running on the aforementioned devices. The user device includes, but is not limited to, various terminal devices such as computers, mobile phones, tablets, smartwatches, and smart bands. The network device includes, but is not limited to, network hosts, single network servers, multiple network server sets, or cloud computing-based computer sets, and can be used to implement some processing functions when setting an alarm clock. Here, the cloud consists of a large number of hosts or network servers based on cloud computing. Cloud computing is a type of distributed computing, consisting of a virtual computer composed of a group of loosely coupled computer sets.

[0146] Figure 4The diagram illustrates the structure of an apparatus suitable for implementing the methods and / or technical solutions in the embodiments of this application. The apparatus 400 includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes based on a program stored in a read-only memory (ROM) 402 or a program loaded from a storage portion 408 into a random access memory (RAM) 403. The RAM 403 also stores various programs and data required for system operation. The CPU 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0147] The following components are connected to I / O interface 405: input section 406 including keyboard, mouse, touch screen, microphone, infrared sensor, etc.; output section 407 including cathode ray tube (CRT), liquid crystal display (LCD), LED display, OLED display, etc., and speakers, etc.; storage section 408 including one or more computer-readable media such as hard disk, optical disk, magnetic disk, semiconductor memory, etc.; and communication section 409 including network interface card such as LAN (local area network) card, modem, etc. Communication section 409 performs communication processing via a network such as the Internet.

[0148] In particular, the methods and / or embodiments in this application can be implemented as computer software programs. For example, the embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart. When the computer program is executed by the central processing unit (CPU) 401, it performs the functions defined in the methods of this application.

[0149] Another embodiment of this application provides a computer-readable storage medium having computer program instructions stored thereon, which can be executed by a processor to implement the methods and / or technical solutions of any one or more embodiments of this application described above.

[0150] Specifically, this embodiment may employ any combination of one or more computer-readable media. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0151] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0152] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0153] The flowcharts or block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of devices, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-specific system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0154] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a device claim may also be implemented by a single unit or device through software or hardware. The terms "first," "second," etc., are used to indicate names and do not indicate any specific order.

Claims

1. An evaluation method for digital retrofitting of property equipment, characterized in that, The method comprises the following steps: Step S1, obtaining first data and pre-processing the first data, wherein the first data comprises equipment operation data and property management process data; Step S2, extracting equipment events and process data events from the first data, wherein the equipment events comprise an equipment ID, a first event type, and a first timestamp, and the process data events comprise a process ID, a second event type, a second timestamp, and text content; Step S3, calculating the time alignment, semantic consistency, and spatial consistency between the equipment events and the process events within a preset time window; Step S4, analyzing the closed-loop accessibility and interoperability between the equipment events and the process events, wherein the interoperability is used to measure the interaction capability between the equipment data system and the property management system, and the closed-loop accessibility is obtained based on at least the process completion rate of the equipment events within a service level agreement time limit and whether the equipment state is restored to normal; The interoperability is obtained based on at least the field mapping success rate, the key field completeness, and the interface call success rate of the equipment data and the process data; Step S5, obtaining a process coupling degree coefficient by weighted fusion based on the time alignment, the semantic consistency, the spatial consistency, the closed-loop accessibility, and the interoperability through a preset weight, and classifying based on the coupling degree coefficient. The time alignment is used to evaluate the matching efficiency of the equipment events and the process events in the time dimension, specifically, the delay time of the matched event pairs is calculated and the delay distribution is counted to obtain the hit rate of the equipment events matched to the process events within a preset time, and the time alignment is obtained by weighted fusion based on the delay distribution and the hit rate through a preset weight.

2. The evaluation method for property equipment digitization reconstruction according to claim 1, wherein The equipment operation data comprises at least equipment alarm logs, state changes, and energy consumption curves, and the property management process data comprises at least process work orders, complaint records, and charging abnormalities. The pre-processing of the first data comprises time reference unification, location coding standardization, and text synonym normalization.

3. The evaluation method for property equipment digitization reconstruction according to claim 1, wherein The second event type is obtained based on a pre-trained first sentence vector model, specifically, the text content is pre-processed, the pre-processed text is input into the first sentence vector model, the sentence vector representation corresponding to the text is extracted, the cosine similarity between the text vector to be recognized and each standard sentence vector is calculated based on a preset standard sentence vector and event type mapping library, and the event type corresponding to the standard sentence vector with the highest similarity is output.

4. The evaluation method for property equipment digitization reconstruction according to claim 1, wherein The semantic consistency is obtained based on a pre-trained second sentence vector model, specifically, a device event type is input into the second sentence vector model to obtain a device event text vector, a text content is input into the second sentence vector model to obtain a process event text vector, a cosine similarity of the device event text vector and the process event text vector is calculated, and the semantic consistency is obtained based on the cosine similarity.

5. The evaluation method for digital transformation of property equipment according to claim 1, characterized in that, The spatial consistency is based on standardized position encoding to determine whether the device event and the process event point to the same spatial range, wherein the position encoding forms a multi-level encoding according to the area range, and is determined step by step according to the level.

6. An evaluation device for digital retrofitting of property equipment, characterized by Comprise: The first data acquisition module is used for acquiring the first data and pre-processing the first data, and the first data comprises equipment operation data and property management process data; The event extraction module is used for extracting device events and process data events from the first data, respectively, the device events comprising device ID, first event type and first timestamp, and the process data events comprising process ID, second event type, second timestamp and text content; The first calculation module is used for calculating the time alignment, semantic consistency and spatial consistency between the device events and the process events within a preset time window; The first analysis module is used for analyzing the closed-loop accessibility and interoperability between the device events and the process events, wherein the closed-loop accessibility is obtained based on at least the process completion rate of the device events within the service level agreement time limit and whether the device state is restored to normal; The interoperability is obtained based on at least the field mapping success rate, key field completeness and interface call success rate of the device data and the process data; The grading module is used for obtaining a process coupling degree coefficient by weighting and fusing the time alignment, semantic consistency, spatial consistency, closed-loop accessibility and interoperability through a preset weight, and grading based on the coupling degree coefficient.

7. An electronic device, the electronic device comprising: Comprise: At least one processor; And a memory connected with the processor in communication; wherein, The memory stores instructions executable by the processor, and the instructions are executed by the processor to enable the processor to execute the method of any one of claims 1-5.

8. A computer readable medium having stored thereon computer program instructions, characterized in that, The computer program instructions can be executed by the processor to implement the method of any one of claims 1-5.

9. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the method of any one of claims 1-5.

Images