Real estate enterprise project management method and system based on edge computing

Abstract

This invention discloses a real estate enterprise project management method and system based on edge computing, specifically relating to the field of data processing technology. It addresses the problem of disordered logical order of cross-node related business events in existing cloud-edge collaborative architectures due to inconsistent clocks and network latency among edge nodes. The method involves acquiring business event data and local clock data collected by each edge computing node, calibrating the local clock based on the inter-node network communication latency threshold to obtain a unified clock reference, extracting and matching resource identifiers from each event data based on the unified clock reference to filter cross-node related business events, and then combining project geographical characteristics and process rules for a rationality assessment to obtain valid related events. Based on predefined state transition rules for resources, the method analyzes the causal logical order between events and generates sequential association results. Based on these results, the business event data is standardized to generate a data summary and uploaded to the cloud platform for integration into a global business status view.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for real estate enterprise project management based on edge computing. Background Technology

[0002] In real estate project management, to overcome the limitations of geographically dispersed data and high latency in cloud processing, existing technologies deploy edge computing nodes at various project sites to build a cloud-edge collaborative architecture. Each edge node is independently responsible for data collection, real-time processing, and local business response within its jurisdiction, and uploads the processed data summary to the cloud platform for aggregation. This aims to achieve rapid perception and response at the project site, while providing management with a centralized data view.

[0003] However, when conducting collaborative management and decision-making across multiple projects and regions based on the aforementioned edge computing architecture, the inherent latency in network communication between nodes, which rely on local clocks for event recording, leads to a lack of consistent and reliable understanding among edge nodes located in different physical locations regarding the logical sequence of cross-node related business events. This results in inherent logical conflicts in the global business status view obtained by the cloud platform integration, severely undermining the reliability and accuracy of core management functions such as resource coordination and scheduling, cross-project process connection, and risk correlation analysis based on the global business status view. Summary of the Invention

[0004] This invention addresses the technical problems existing in the prior art by providing a real estate enterprise project management method and system based on edge computing.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: Edge computing-based project management methods for real estate enterprises include: S1. Obtain business event data and local clock data collected by edge computing nodes at each project site; S2. Based on the network communication latency threshold between edge computing nodes, the local clock data of each edge computing node is calibrated to obtain a unified clock reference; S3. Based on a unified clock reference, extract and match resource identifiers in the data of each business event, determine the resources corresponding to the resource identifiers, and filter out cross-node related business events that are related to the flow of the same resource. S4. Combine the geographical distribution characteristics of real estate projects and the rules for connecting work processes to conduct a rationality assessment of cross-node related business events and obtain valid cross-node related business events. S5. Based on the predefined state transition rules of the resources corresponding to the resource identifiers, analyze and determine the causal logical order between valid cross-node related business events, and generate event sequence association results. S6. Based on the results of event sequence association, standardize the business event data to obtain a data summary, and upload the data summary to the cloud platform for integration.

[0006] Furthermore, S1 includes: From edge computing nodes at each project site, collect business event data reflecting material flow, equipment status changes, or process status changes; When acquiring data for each business event, the local clock data of the corresponding edge computing node is obtained synchronously.

[0007] Furthermore, S2 includes: Select an edge computing node as the master clock node and use its local clock data as the initial reference clock. Based on the network communication latency threshold between edge computing nodes, calculate the clock deviation compensation amount of the local clock data of each other edge computing node relative to the initial reference clock. Based on the clock offset compensation amount, the local clock data of other edge computing nodes are compensated and adjusted so that all edge computing nodes are aligned to a unified clock reference based on the compensated and adjusted local clock data.

[0008] Furthermore, S3 includes: Based on a unified clock reference, business event data from different edge computing nodes are preprocessed in chronological order; From preprocessed business event data, parse and extract resource identifiers that represent material, equipment, or process entities; The extracted resource identifiers are matched and compared across edge computing nodes to identify business event data with the same resource identifiers; Based on the matching results, business event data with the same resource identifier but from different edge computing nodes are aggregated as cross-node associated business events related to the same resource flow.

[0009] Furthermore, S4 includes: Based on the geographical distribution characteristics of real estate projects, the rationality of the physical spatial path of resource flow in cross-node related business events is verified. Based on the process connection rules of real estate projects, verify the rationality of the logical order of business events in cross-node related business events; Based on the verification results of the rationality of the physical space path and the rationality of the logical order, business events that meet the verification conditions are selected from the cross-node associated business events as valid cross-node associated business events.

[0010] Furthermore, S5 includes: Based on predefined state transition rules, valid cross-node associated business events are parsed to reconstruct the complete state evolution chain of the resource corresponding to the resource identifier across the edge computing nodes. Based on a unified clock reference, the occurrence time of all state nodes in the reconstructed complete state evolution chain is reverse-mapped and consistency-verified with the occurrence time of each valid cross-node associated business event. Based on the results of reverse mapping and consistency verification, the causal logical order between each valid cross-node related business event that can uniquely reflect the complete state evolution chain is determined and output as the event sequence association result.

[0011] Furthermore, the predefined state transition rules are established based on the management process of the real estate project resource lifecycle, defining multiple discrete business states of resources and the allowed transition relationships between states.

[0012] Furthermore, the complete state evolution chain of the resource corresponding to the resource identifier across the edge computing nodes is reconstructed, including: The effective cross-node related business events are time-sequentially ordered based on a unified clock reference. Each valid cross-node associated business event after sorting is mapped to the corresponding business state node in the predefined state transition rule; Verify and connect the state nodes mapped to adjacent business events according to predefined state transition rules, forming a complete state evolution chain of resources across the edge computing nodes.

[0013] Furthermore, S6 includes: Based on the causal logical order indicated by the event sequence association results, the business event data is restructured in a structured manner to generate a data summary with resource flow as the main line and marked with time sequence logic; The generated data digest is transmitted from each edge computing node to the cloud platform; In the cloud platform, based on the time sequence logic marked in the data digest, the data digests transmitted by different edge computing nodes are time-series aligned and merged to complete the integration of the global business status view.

[0014] On the other hand, the present invention provides a real estate enterprise project management system based on edge computing, comprising: The data acquisition module is used to acquire business event data and local clock data collected by edge computing nodes at each project site; The clock calibration module is used to calibrate the local clock data of each edge computing node based on the network communication latency threshold between edge computing nodes, so as to obtain a unified clock reference. The association filtering module is used to extract and match resource identifiers in the data of each business event based on a unified clock reference, determine the resources corresponding to the resource identifiers, and filter out cross-node associated business events related to the flow of the same resource. The correlation assessment module is used to assess the rationality of cross-node related business events by combining the geographical distribution characteristics and process connection rules of real estate projects, and to obtain valid cross-node related business events. The timing determination module is used to analyze and determine the causal logical order between valid cross-node related business events based on the predefined state transition rules of the resources corresponding to the resource identifiers, and generate event sequence association results. The data upload module is used to standardize business event data based on the result of event sequence association, obtain data summaries, and upload the data summaries to the cloud platform for integration.

[0015] The beneficial effects of this invention are: 1. By establishing a unified time benchmark through clock calibration between edge computing nodes, a reliable time scale is provided for the comparison of all subsequent cross-node events. By extracting and matching resource identifiers in business events, the chain of related events occurring between multiple nodes around the same physical or logical resource can be accurately identified. Discrete event data is linked into a resource flow trajectory with business significance. Combined with the geographical characteristics of the project and the process rules, the rationality of related events is verified twice, effectively filtering out false associations caused by data errors or collection anomalies, ensuring the business authenticity of the event set entering subsequent analysis. Based on the state transition rules predefined for resources, the causal order of events is determined, and the necessary sequential relationship inherent in the business logic itself is used to infer the order of events. This ensures that even in scenarios where network latency causes timestamp deviations, logically correct and reliable sequential association results can be obtained. Based on this accurate order, a standardized data summary is generated and uploaded to the cloud for integration, so that the global business state view formed in the cloud eliminates the inherent logical conflicts.

[0016] 2. Because the global view has highly consistent temporal logic and reliable business causal relationships, the management can accurately grasp the real-time location and status of resources based on this view, and achieve precise allocation and efficient utilization of cross-project resources. For cross-project process connections, the clear causal event sequence enables reliable analysis and planning of the dependencies and progress coordination between preceding and subsequent processes, avoiding schedule conflicts or resource idleness caused by misjudgment of sequence. In terms of risk correlation analysis, the analysis based on real and reasonable event chains can accurately trace the root cause of problems and assess the risk transmission path, thereby supporting the formulation of effective prevention and response strategies. It fully leverages the local real-time processing advantages of edge computing and achieves high-quality integration of data and logic through the cloud. Attached Figure Description

[0017] Figure 1 The flowchart below shows the real estate enterprise project management method based on edge computing according to the present invention. Figure 2 This is a schematic diagram of the edge computing-based real estate enterprise project management system of the present invention. Figure 3 This is a simulation diagram comparing the clock synchronization errors of edge nodes in a complex network environment according to the present invention. Figure 4 This is a comparison chart showing the interception effect of the "dual rational evaluation" of the present invention on heterogeneous data. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: Figure 1 This invention presents a real estate enterprise project management method based on edge computing, comprising: S1. Obtain business event data and local clock data collected by edge computing nodes at each project site; S2. Based on the network communication latency threshold between edge computing nodes, the local clock data of each edge computing node is calibrated to obtain a unified clock reference; S3. Based on a unified clock reference, extract and match resource identifiers in the data of each business event, determine the resources corresponding to the resource identifiers, and filter out cross-node related business events that are related to the flow of the same resource. S4. Combine the geographical distribution characteristics of real estate projects and the rules for connecting work processes to conduct a rationality assessment of cross-node related business events and obtain valid cross-node related business events. S5. Based on the predefined state transition rules of the resources corresponding to the resource identifiers, analyze and determine the causal logical order between valid cross-node related business events, and generate event sequence association results. S6. Based on the results of event sequence association, standardize the business event data to obtain a data summary, and upload the data summary to the cloud platform for integration.

[0020] S1. Obtain business event data and local clock data collected by edge computing nodes at each project site. The specific implementation is as follows: Specifically, in step S1, edge computing nodes are deployed at key business locations such as concrete mixing plants, rebar processing areas, tower crane operating surfaces, material warehouses, and construction floor working surfaces. Each edge computing node is equipped with a corresponding data acquisition interface, which is used to connect to and read data outputs from sensing devices and business systems deployed at that physical location. Sensing devices and business systems generate raw signals or records reflecting on-site business activities in real time or at preset intervals. The data acquisition program running on the edge computing node obtains raw, unprocessed business activity records from the connected sensing devices and business systems through polling or event-triggered methods. These business activity records constitute the initial business event data. For example, when a truckload of ready-mixed concrete leaves the mixing plant, an RFID reader installed at the plant's exit reads the electronic tag on the concrete truck. The RFID reader sends the read tag identifier, read time, and read location information to the edge computing node where the mixing plant is located. The edge computing node then generates a raw record containing all information from this read operation; this raw record serves as business event data reflecting material flow. For example, when a tower crane completes a lifting cycle, its control system generates an operation completion report containing the lifting task number, completion time, and start and end position information. This report is sent to the edge computing node where the tower crane's working surface is located, forming a business event data reflecting the change in equipment status. Similarly, when a formwork support process on a floor is confirmed by on-site management personnel via a mobile terminal, the mobile terminal sends a confirmation instruction containing the process code, completion confirmation time, confirmer, and their work team information to the edge computing node where that floor is located. The edge computing node then generates a business event data reflecting the change in process status based on the received confirmation instruction. All raw business event data collected by the edge computing nodes is organized and temporarily stored according to a predefined data structure. This predefined data structure includes a sequence number field to uniquely identify the business event, a business event type code field, an event content details field, and a field for recording time information.

[0021] Meanwhile, to assign a traceable local time reference to each business event data record, the edge computing node synchronously queries the local hardware clock maintained by its own operating system the moment it successfully collects and generates a business event data record. The local hardware clock is driven by a clock chip on the edge computing node's motherboard, and its timing is based on the crystal oscillator frequency of the clock chip. The edge computing node establishes a strict correspondence between the current date and time value read from the local hardware clock and this business event data as an independent data item. This correspondence is achieved by filling the timestamp field in the business event data's data structure with date and time values, or by establishing a mapping index between the unique identifier of the business event data record and a separate corresponding timestamp record. For example, when the edge computing node of the mixing plant generates a concrete delivery event record, the edge computing node immediately calls the system clock interface to obtain the current local time, for example, the value 14:30:05 on October 27, 2025, and uses this value as the content of the timestamp field of the concrete delivery event record. When a tower crane completion report arrives at the edge computing node where the tower crane's working face is located, the edge computing node records the local time at which it received the report. When a process completion confirmation instruction arrives at the edge computing node where the construction floor is located, the edge computing node records the local time at which it received the instruction. The acquired local clock data reflects the time indicated by the local hardware clock of the specific edge computing node when the business event data is generated or received. All business event data with accompanying local clock data is then cached in the local storage unit of the edge computing node to provide a unified processing object for the subsequent step S2. In this way, step S1 completes the collection of business-meaningful event data from distributed physical nodes and attaches the local time reference of its generating node to each piece of data, forming the basic data set for all subsequent analysis and processing.

[0022] S2. Based on the network communication latency threshold between edge computing nodes, the local clock data of each edge computing node is calibrated to obtain a unified clock reference. The specific implementation is as follows: One edge computing node is selected as the master clock node from all participating edge computing nodes. The master clock node can be selected by specifying an edge computing node located at a central geographical location, specifying an edge computing node with the most stable local hardware clock source, or dynamically electing an edge computing node through a negotiation protocol between nodes. The local clock data acquired by the selected edge computing node in step S1 is defined as the initial reference clock. The value of the initial reference clock, such as a date and time combination like October 27, 2025, 10:00:00:00 milliseconds, will serve as the reference origin for subsequent calibration operations.

[0023] Determining the network communication latency threshold requires first understanding its meaning. The network communication latency threshold between edge computing nodes refers to the maximum allowable time delay for round-trip communication of network packets between different edge computing nodes. This threshold can be obtained during the system deployment phase through actual network measurements. During actual network measurements, multiple time-stamped test data packets are continuously sent between the master clock node and each other edge computing node, and corresponding responses are awaited. The complete round-trip time for each request and response is recorded. After collecting multiple round-trip time samples, these sample values ​​are sorted, and a larger sample value is selected as the network communication latency threshold. For example, the sample value at the 95th percentile after sorting can be selected as the network communication latency threshold for that node pair. Network communication latency thresholds can also be estimated and set based on theoretical models of the network architecture. For example, for edge computing nodes directly connected via wired LANs, the network communication latency threshold can be set to 1 to 10 milliseconds; for edge computing nodes connected via metropolitan wide area networks (WANs), the threshold can be set to 20 to 100 milliseconds; and for edge computing nodes connected via mobile cellular networks, the threshold can be set to 100 to 500 milliseconds. A specific network communication latency threshold needs to be determined between each other edge computing node and the master clock node.

[0024] The clock offset compensation amount of the local clock data of each other edge computing node relative to the initial reference clock is calculated based on the network communication latency threshold. The calculation process follows the basic principle of clock synchronization. For each edge computing node that needs calibration, the master clock node sends a synchronization request message to the edge computing node that needs calibration. The synchronization request message carries the local clock data read by the master clock node at the time of sending this synchronization request message, which is denoted as T1. The edge computing node that needs calibration immediately reads its own local clock data when it receives the synchronization request message, which is denoted as T2. The edge computing node that needs calibration replies to the master clock node with an acknowledgment message. The acknowledgment message carries T2 and the local clock data read by the edge computing node that needs calibration at the time of sending this acknowledgment message, which is denoted as T3. The master clock node immediately reads its own local clock data when it receives the acknowledgment message, which is denoted as T4. Assuming that the network path is symmetrical, that is, the propagation delay of the message from the master clock node to the edge computing node that needs calibration is equal to the propagation delay from the edge computing node that needs calibration back to the master clock node, this propagation delay is denoted as D. The relationship between T2 and T1 can be expressed as: T2 = T1 + D + θ; where T2 represents the local time when the edge computing node to be calibrated receives the synchronization request message, T1 represents the local time when the master clock node sends the synchronization request message, D represents the one-way propagation delay of the network message between the master clock node and the edge computing node to be calibrated, and θ represents the clock deviation of the edge computing node to be calibrated relative to the master clock node. The relationship between T4 and T3 can be expressed as: T4 = T3 + D - θ; where T4 represents the local time when the master clock node receives the acknowledgment message, and T3 represents the local time when the edge computing node to be calibrated sends the acknowledgment message. From these two relationships, the clock deviation can be derived as: θ = (T2 - T1 - T3 + T4) / 2. Since a single measurement may be affected by instantaneous network jitter, multiple measurements are required. After each measurement, check whether the calculated round-trip time, T4 - T1, is less than or equal to twice the predetermined network communication delay threshold. Only measurement results that meet this condition are retained for subsequent calculations. The arithmetic mean of the clock deviation θ values ​​calculated from the retained multiple measurements is taken. The final average value is the clock deviation compensation amount of the local clock data of the edge computing node that needs to be calibrated relative to the initial reference clock. The clock deviation compensation amount can be positive or negative. For example, a positive 5.2 milliseconds means that the local clock of the edge computing node that needs to be calibrated is 5.2 milliseconds faster than the initial reference clock, and a negative 3.8 milliseconds means that the local clock of the edge computing node that needs to be calibrated is 3.8 milliseconds slower than the initial reference clock.

[0025] The system compensates for and adjusts the local clock data of other edge computing nodes based on the clock skew compensation amount. This compensation adjustment is implemented at the application logic level and does not directly modify the physical hardware clock of the edge computing nodes. The system maintains a corresponding compensation adjustment value for each edge computing node; this value is the clock skew compensation amount calculated for that node. When subsequent steps require the time information accompanying the business event data collected by a particular edge computing node, the system reads the original local clock data from that business event data record and adds the compensation adjustment value corresponding to that edge computing node to obtain the compensated and adjusted local clock data. The compensated and adjusted local clock data represents the time point at which the business event should be recorded under a unified clock benchmark. By applying their respective compensation adjustment values ​​to the business event data collected by all edge computing nodes, all edge computing nodes are logically aligned to the same unified clock benchmark based on the compensated and adjusted local clock data. This unified clock benchmark provides a common and consistent time measurement standard for business event data from different geographical locations and different edge computing nodes, providing a reliable basis for subsequent steps involving comparisons, sorting, and correlation analysis based on chronological order. After the entire calibration process is completed, the information of the unified clock reference and the compensation adjustment value corresponding to each edge computing node are persistently stored in the system configuration or database for continuous use.

[0026] To verify the technical effectiveness of the clock calibration scheme in step S2 under complex network environments, such as Figure 3 As shown, a network simulation environment including a master clock node and edge nodes was constructed. The basic network latency was set to 20ms, and random jitter following a normal distribution and occasional high-latency pulses up to 500ms were introduced to simulate the unstable network environment of the construction site. The network communication latency threshold of this invention was set to 50ms. In a continuous test of 60 minutes, the control group using the conventional NTP synchronization strategy and the experimental group using the latency threshold filtering and deviation compensation of this invention were compared. The simulation data showed that when a network congestion peak of 420ms occurred at T+30 minutes, the control group failed to identify the abnormal latency, causing the clock synchronization error to soar to 208.5ms, resulting in severe drift. The experimental group successfully eliminated the abnormal message using the threshold mechanism, and stabilized the synchronization error at 4.5ms. Throughout the test period, the average synchronization error of the experimental group remained within 5ms, which was significantly better than the average error level of more than 50ms of the control group. This strongly proves the stability and accuracy of the unified clock reference established by this application under harsh network conditions.

[0027] S3. Based on a unified clock reference, extract and match resource identifiers from the data of each business event, determine the resources corresponding to the resource identifiers, and filter out cross-node related business events related to the same resource flow. The specific implementation is as follows: Business event data from different edge computing nodes is preprocessed based on a unified clock reference. The core operation of preprocessing is chronological sorting. From the processing result of step S2, each business event data entry includes compensated and adjusted local clock data, which is the timestamp of that business event data under the unified clock reference. The preprocessing program reads all the business event data to be processed, reads the timestamp field from each data entry, and then sorts all the business event data in ascending order from earliest to latest based on the date and time values ​​recorded in the timestamp field. The sorting algorithm used is quicksort. The sorting operation produces a sequence of business event data arranged in an ordered manner according to the unified time reference. This ordered sequence ensures that the temporal relationship of events is clear and consistent in subsequent processing.

[0028] From a preprocessed, chronologically ordered sequence of business event data, resource identifiers representing material, equipment, or process entities are parsed and extracted. The parsing process relies on the event content details field of the business event data. This field stores structured or semi-structured data reflecting specific business activities. The system predefines a parsing template for each type of business event. This template is a data structure that specifies the specific key name, position offset within the string, or specific format pattern used for identification of the resource identifier in the event content details field for that type of business event. For example, for an event type reflecting material arrival and acceptance, the parsing template specifies finding the key-value pair with the key name "batch number" in the event content details field and extracting the corresponding value as the resource identifier. For instance, a string in such an event content details field might be: Material type = rebar, specification = HRB400E 25mm, batch number = GJ20251027001, quantity = 50 tons, supplier = a certain steel company. Based on the parsing template, the character sequence GJ20251027001 following the batch number is extracted as the resource identifier. The extraction process is implemented using a string splitting function. First, the string is split into multiple key-value pairs based on semicolons. Then, for each pair, the key and value are separated using an equals sign. When the key matches the batch number, the corresponding value is extracted. The extracted resource identifier is associated with the source business event data. Specifically, an extended entry is created in memory for each business event data record. This extended entry contains the unique identifier of the source record and the extracted resource identifier string.

[0029] The extracted resource identifiers are matched and compared across edge computing nodes to identify business event data with the same resource identifier. This matching and comparison operation requires aggregating extended entries from all business event data processed by the edge computing nodes. After extracting resource identifiers from their local business event data, each edge computing node sends extended entry records containing the unique identifier of the business event data, the extracted resource identifier string, and the edge computing node's own identifier to a centralized association analysis service. The association analysis service collects all extended entry records. The core of the matching and comparison is the equivalence judgment of the resource identifier strings. Before comparison, each resource identifier string is standardized and cleaned, including removing leading and trailing whitespace characters and converting all English letters in the string to uppercase. The system iterates through all standardized and cleaned resource identifier strings, using identical string values ​​as the aggregation key, and groups all extended entry records with the same string value into the same logical group. For example, an extended entry record from the edge computing node of the mixing plant carries the resource identifier string C30-20251027-001, and an extended entry record from the edge computing node of the construction site also carries the resource identifier string C30-20251027-001. After standardization cleaning, the two strings are determined to be completely identical, so the two extended entry records are grouped into the same logical group.

[0030] Based on the matching and comparison results, business event data with the same resource identifier but originating from different edge computing nodes are aggregated as cross-node associated business events related to the same resource flow. The aggregation operation is based on the logical grouping results generated by the matching and comparison. The system checks the extended entry records within each logical group. For a logical group, the system checks the edge computing node's own identifier field contained in all extended entry records within the group. If there are two or more different edge computing node's own identifiers within the group, it is determined that the original business event data corresponding to all extended entry records within that logical group constitute a set of cross-node associated business events. For example, a logical group has a resource identifier string of B789XYZ. This logical group contains three extended entry records: Extended entry record A has an edge computing node's own identifier of Node_Plant, and its corresponding original business event data content is "rebar cutting completed"; Extended entry record B has an edge computing node's own identifier of Node_Logistics, and its corresponding original business event data content is "rebar cage loading and shipping"; Extended entry record C has an edge computing node's own identifier of Node_Site, and its corresponding original business event data content is "rebar cage arrival and reception". Since the edge computing node identifiers Node_Plant, Node_Logistics, and Node_Site in the three extended entry records are different, the three original business event data corresponding to them are grouped into a set of cross-node related business events. The system assigns a unique association group identifier, such as a UUID format string, to each such set of cross-node related business events, and establishes a mapping relationship between the unique identifiers of all original business event data within the group and this association group identifier. The resulting set of mapping relationships is the output of step S3, which is then processed by subsequent steps S4 and S5. Through the above process, step S3 accurately discovers and organizes sets of related events that are essentially related to the same physical or logical resource from multi-source business event data.

[0031] S4. Based on the geographical distribution characteristics and process connection rules of real estate projects, a rationality assessment is conducted on cross-node related business events to obtain valid cross-node related business events. The specific implementation is as follows: Based on the geographical distribution characteristics of real estate projects, the rationality of the physical spatial path of resource flow in cross-node associated business events is verified. The geographical distribution characteristics of real estate projects refer to the actual geographical coordinates of each functional area or construction site in the project and the topological connections between these locations. This information is predefined and stored in the project management system, recording the latitude and longitude coordinates of each edge computing node deployment location and the functional area type to which that location belongs in the form of a database table. When verifying the rationality of the physical spatial path, the system obtains a set of cross-node associated business events to be verified. This set of events comes from the output of step S3 and includes an association group identifier. The system parses the edge computing node identifier associated with each business event data occurrence from this set of events. Based on the edge computing node identifier, the system queries the geographical distribution characteristic database to obtain the geographical coordinates of each edge computing node deployment location. The system connects these edge computing node locations according to the time sequence of occurrence of the business event data under a unified clock reference, forming a hypothetical physical movement path for resource flow. To verify the rationality of this hypothetical path, the system calculates the geographical straight-line distance between two adjacent locations in the path. When calculating the straight-line distance, the system uses the latitude and longitude coordinates of two locations and applies spherical trigonometry formulas or simplified planar projection distance formulas. The results are expressed in kilometers. The system presets a maximum allowable transportation distance threshold. This threshold is set based on the resource type and actual project conditions, taking into account the physical characteristics of the resource, the reasonable transport distance of commonly used transportation tools, and the actual road network conditions at the project site. For example, for easily setting materials like ready-mixed concrete, the maximum allowable transportation distance threshold can be set to 30 kilometers. For ordinary steel reinforcement, the threshold can be set to 100 kilometers. The verification logic requires that the calculated straight-line distance between adjacent locations must be less than or equal to the preset maximum allowable transportation distance threshold for that type of resource. If all adjacent distances meet the threshold condition, and the path direction generally conforms to the expected direction from the production area to the storage area and then to the construction area, then the physical spatial path of this group of cross-node related business events is deemed reasonable. For example, in a set of related events concerning concrete, the path is sequentially the coordinates of the batching plant node and the coordinates of the construction site A area node. The calculated distance between the two points is 25 kilometers, which is less than the maximum allowable transportation distance threshold of 30 kilometers for easily setting materials like ready-mixed concrete, so the verification is passed. If the distance exceeds the maximum allowable transportation distance threshold, or if the path shows a contradictory route from the construction site back to the batching plant, then the physical path is determined to be unreasonable.

[0032] Based on the process connection rules of real estate projects, the rationality of the logical order of business events in cross-node related business events is verified. The process connection rules of real estate projects define the inherent and irreversible sequential dependencies between various operational activities in project management. These rules are pre-formally stored in a rule base. Each rule in the rule base states that process A must be completed before process B begins and is associated with a specific resource type or engineering part. When verifying the rationality of the logical order, the system first identifies the core resource types involved in the group of cross-node related business events, such as concrete, formwork, and reinforcement. Then, based on the resource type, it retrieves the standard process chain applicable to that resource from the process connection rule base. The standard process chain is an ordered list of process nodes. The system maps each business event data within the group to one or more process nodes in the standard process chain. The mapping is based on the semantic matching between the business action described in the event content details field of the business event data and the process node. For example, an event content of "concrete pouring begins" maps to the "pouring" node in the process chain. For example, an event content of "formwork removal completed" maps to the "formwork removal" node. After mapping, the system checks whether the actual occurrence order of these business event data, sorted according to a unified clock reference, is consistent with the logical order specified in the standard process chain. The principle of consistency checking is that if the process chain stipulates that process X must precede process Y, then the occurrence time of any business event mapped to process X must be earlier than the occurrence time of any business event mapped to process Y. If the actual time sequence of all events conforms to the logical order specified by the rules, then the logical order of the group of events is determined to be reasonable. For example, for rebar resources, the standard process chain is rebar processing, rebar transportation, and rebar binding. The corresponding business event time sequence is processing completion event time t1, loading and shipping event time t2, and on-site binding start event time t3, and t1 is earlier than t2, and t2 is earlier than t3, then the logical order is reasonable. If the binding start event time is earlier than the processing completion event time, then the logical order is unreasonable.

[0033] Based on the verification results of the physical path rationality and the logical order rationality, business events that meet the verification conditions are selected as valid cross-node associated business events from the cross-node associated business events. The selection process is executed by a decision logic. The input of the decision logic is a combination of Boolean values ​​of the physical path rationality verification result and the logical order rationality verification result. The verification result is a Boolean value, i.e., reasonable or unreasonable. The decision rule is that only when a group of cross-node associated business events passes both the physical path rationality verification and the logical order rationality verification will all member business event data of that group be marked as valid cross-node associated business events. The system records a validity status flag for these marked valid business event data and sets the flag bit to true. For associated event groups that fail to pass both verifications simultaneously, the system marks them as invalid. Invalid event groups will be logged for subsequent manual verification, but will not enter the subsequent processing flow of step S5. Through this dual verification mechanism, step S4 can efficiently filter out false or erroneous associations that, although linked by identifiers at the data level, are impossible or unreasonable in the actual physical world or project management logic. This ensures that the event set on which subsequent analysis is based has high business authenticity and reliability, laying a solid data quality foundation for generating accurate sequential association results.

[0034] To verify the technical effectiveness of the rationality assessment scheme based on geographical distribution and process rules in step S4 under real working conditions, such as... Figure 4 As shown, a linear simulation scenario is constructed, comprising a concrete mixing plant (node ​​A, coordinate 0km), a construction site gate (node ​​B, coordinate 15km), and a pouring operation surface (node ​​C, coordinate 15.5km). The maximum transport speed threshold is preset to 60km / h, and the standard process sequence is A production—B arrival—C pouring. A test set containing 1000 sets of resource flow data is constructed, which includes 10% physical instantaneous movement anomaly data (the flow time from A to B is set to only 5 minutes, and the calculated average speed is 180km / h, far exceeding the speed threshold) and 10% process reversal anomaly data (the event time of node C is set to occur earlier than that of node B). (10 minutes); comparative tests showed that the control group, which relied solely on resource identifier matching, erroneously marked all 200 sets of dirty data as valid due to a lack of semantic verification capabilities, resulting in a data purity of 80%. In contrast, the experimental group, which adopted step S4 of this application, successfully intercepted 97% of physical anomalies and 98% of logical anomalies under the premise of simulating real sensor noise environment and allowing extremely low false alarms (<1%), through multi-dimensional rule verification. This significantly improved the validity of downstream data from 80% to 99.5%, thereby quantitatively demonstrating the significant technical effect and robustness of this solution in eliminating false associations and ensuring the authenticity of business data in complex on-site noise environments.

[0035] S5. Based on the predefined state transition rules of the resources corresponding to the resource identifiers, analyze and determine the causal logical order between valid cross-node related business events, and generate event sequence association results. The specific implementation is as follows: The establishment of predefined state transition rules is the primary foundation for implementation. These rules are based on the abstraction and formalization of the full lifecycle management process for specific types of resources in a real estate project. For each type of managed resource, such as ready-mixed concrete, reinforcing steel, formwork, or large equipment, all key business stages experienced by the resource from its entry into the project to its final consumption or installation are analyzed. Each key business stage is defined as a discrete business state. For example, for ready-mixed concrete resources, the defined discrete business states may include: reserved, in production, loaded, in transit, arrived, awaiting inspection, inspected and approved, poured, curing, and cured. The definitions of business states are explicit and unambiguous; each business state corresponds to a specific stage in the physical form or management procedures of the resource. The permissible transition relationships between states define the legal paths for a resource to change from one business state to another. These transition relationships are defined based on business logic and physical constraints and are typically unidirectional. For example, permitted transition relationships include transitioning from the "produced" state to the "loaded" state, from the "loaded" state to the "in transit" state, and from the "in transit" state to the "arrived" state. Reverse transitions from the "arrived" state to the "in transit" state are not permitted. These business states and the permitted transition relationships between states are stored as a state transition diagram or a state transition matrix. This state transition diagram or matrix constitutes a predefined state transition rule base. The predefined state transition rule base can be established by project managers entering information according to construction specifications through a configuration interface, or it can be derived by analyzing historical project data.

[0036] The system parses valid cross-node associated business events according to predefined state transition rules to reconstruct the complete state evolution chain of the resource corresponding to the resource identifier across the edge computing nodes. During implementation, the system acquires a set of valid cross-node associated business events from step S4, which share the same resource identifier. The system sorts all valid cross-node associated business events within the group according to a unified clock reference. The sorting is based on the timestamp represented by the compensated and adjusted local clock data obtained in step S2, which is attached to each valid cross-node associated business event data. The sorting algorithm generates a list of valid cross-node associated business events arranged from earliest to latest timestamp. The system maps each sorted valid cross-node associated business event to the corresponding business state node in the predefined state transition rules. The mapping process is based on the semantic association rules between the business actions described in the event content details field and the business state nodes. The semantic association rules are predefined. For example, if the event content of a valid cross-node associated business event is "Concrete mixing completed batch number C30-001," the mapping rule will match it to the "produced" state. For example, another valid cross-node associated business event data has the event content of concrete truck departure batch number C30-001. The mapping rule will match it to the loaded or in transit state. The system verifies and connects the state nodes mapped to adjacent valid cross-node associated business events according to predefined state transition rules, forming a complete state evolution chain of resources across the edge computing nodes. The system starts from the first valid cross-node associated business event in the time-series sorted list and obtains the business state mapped to the first valid cross-node associated business event. The system checks the business state mapped to the next valid cross-node associated business event. The system queries the predefined state transition rules to determine whether the transition from the previous business state to the next business state is allowed. If the predefined state transition rules allow the transition from the previous business state to the next business state, the system connects the next business state to the previous business state, forming a preliminary chain. The system traverses all valid cross-node associated business events in chronological order, verifies the legality of the transition between the mapped states of adjacent valid cross-node associated business events, and connects the legal business state nodes in chronological order. The final sequence formed, from the initial business state to the final business state, and composed of multiple business state nodes connected in chronological order, is considered the complete state evolution chain of the reconstructed resource across the edge computing nodes. For example, for resource identifier C30-001, after the effective cross-node associated business events are sorted and mapped by time, the complete state evolution chain formed by connecting them may be from the produced state to the loaded state to the transported state to the arrived state to the inspected and qualified state to the poured state.

[0037] Based on a unified clock reference, the occurrence times of all state nodes in the reconstructed complete state evolution chain are reverse-mapped and their consistency verified with the occurrence times of each valid cross-node associated business event. Here, the occurrence time of a state node refers to the point in time when the business state is logically considered to have occurred. When reconstructing the complete state evolution chain, each business state node is associated with a specific valid cross-node associated business event; therefore, the occurrence time of a business state node is defaulted to the timestamp of the associated valid cross-node associated business event. The purpose of the reverse mapping is to verify the consistency of this default setting. Specifically, the system traverses each business state node in the complete state evolution chain. For a current business state node in the complete state evolution chain, the system finds the original valid cross-node associated business event mapped to that current business state node. The system reads the occurrence time of this valid cross-node associated business event under the unified clock reference. Simultaneously, based on the allowed transition rules between states and business knowledge, there are reasonable constraints on the time interval between adjacent business states. For example, from the "loaded" state to the "arrived" state, considering the transportation distance, the time difference between the two business states should be within a reasonable range. The system presets a maximum state interval time threshold; for example, for concrete transportation, the maximum state interval time threshold is set to 2 hours. Consistency verification checks the timestamps of any pair of adjacent business state nodes in the complete state evolution chain. It calculates the time difference between the occurrence time of the subsequent business state node and the occurrence time of the preceding business state node. It verifies whether this time difference is greater than 0 to ensure time increment; and it also verifies whether this time difference is less than or equal to the preset maximum state interval time threshold for this specific business state transition. If the time intervals of all adjacent business state nodes satisfy the condition of being greater than 0 and less than or equal to the corresponding maximum state interval time threshold, the consistency verification passes. This means that the reconstructed complete state evolution chain is continuous and reasonable in the time dimension and is consistent with the observed event timestamps. If a time difference does not meet the conditions, such as a negative time difference or a time difference exceeding the maximum state interval time threshold, it indicates a significant deviation between the effective cross-node associated business event timestamps and the logical expectation of the complete state evolution chain, and the consistency verification fails.

[0038] The maximum state interval time threshold is set comprehensively based on the physical and chemical characteristics of the resource itself, as well as the efficiency standards of routine operations on the project site. For example, a threshold is set for the transportation stage based on the setting characteristics of concrete, and a threshold is set for the processing stage based on the historical average operation time of each process, ensuring that the thresholds conform to objective laws and are in line with the actual management level of the project.

[0039] Based on the results of reverse mapping and consistency verification, the system determines and outputs the causal logical order between each valid cross-node associated business event that uniquely reflects the complete state evolution chain, as the event sequence association result. The determination logic is based on the conclusion of the consistency verification. If the consistency verification passes, the system confirms that the reconstructed complete state evolution chain is valid. At this time, the order of the business state nodes in the complete state evolution chain directly and uniquely determines the causal logical order between each valid cross-node associated business event mapped to these business state nodes. This is because the allowed transition rules between states define causality, i.e., the preceding business state is the cause and the succeeding business state is the effect, and the complete state evolution chain is a legal sequence under the rules. The system extracts the corresponding valid cross-node associated business event identifiers according to the order of the business state nodes in the complete state evolution chain and generates an ordered list. This ordered list is the event sequence association result. The event sequence association result explicitly records causal relationships such as event A occurring before event B and event B occurring before event C. The output format can be a list, a directed graph, or a data structure containing temporal relationships. If the consistency verification fails, the system determines that it cannot generate a definite event sequence association result based on the current data. At this point, the system can output an anomaly flag, marking the problematic business status node and its corresponding valid cross-node associated business events, indicating the need for manual review or data cleaning. If the process proceeds normally, the generated event sequence association results will be passed to step S6 to guide the standardization of the data summary.

[0040] S6. Based on the event sequence association results, the business event data is standardized to obtain a data summary, which is then uploaded to the cloud platform for integration. The specific implementation is as follows: Based on the causal logical order indicated by the event sequence association results, the business event data is structurally reorganized to generate a data summary with resource flow as the main thread and time-series logic annotations. The event sequence association results are derived from the output of step S5, defining the sequential causal relationships between a set of valid cross-node related business events for a specific resource in the form of an ordered list. The system retrieves the complete record of the corresponding original business event data from storage according to the order of event identifiers in this ordered list. The structured reorganization operation is based on a predefined data summary template. The data summary template specifies the fields that the data summary should include, aiming to extract the core information of the original business event data and establish a clear resource flow context. The system iterates through each event identifier in the event sequence association result list. For the first event identifier in the event sequence association result list, the system reads the original business event data corresponding to the first event identifier, extracting the resource identifier, business event type, event location, key operation results in the event content, and the occurrence time of the original business event data corresponding to the first event identifier under a unified clock reference. Information extracted from the original business event data corresponding to the first event identifier is filled into the starting paragraph of the data summary template. For subsequent event identifiers in the event sequence association result list, the system performs the same extraction operation and adds the information extracted from the original business event data corresponding to the subsequent event identifiers as a new paragraph order to the data summary. Each new paragraph in the data summary explicitly references the resource status of the previous paragraph in the data summary to reflect causal logic. The final generated data summary is a structured document or data object. The core content of the data summary includes a unified resource identifier, a chronologically ordered sequence of events, and each entry in the event sequence containing the event type, location, key operation results, and precise timestamp, as well as the implicit or explicitly declared state transition relationships between the event sequence entries. For example, for a batch of concrete, the generated data digest might include the resource identifier C30-20251027-001 and an event sequence: Event 1, type: production completed, location: mixing plant, result: produced, timestamp: October 27, 2025, 08:00:00; Event 2, type: shipment and departure, location: mixing plant, result: loaded, timestamp: October 27, 2025, 08:30:00; Event 3, type: on-site receipt, location: construction site A area, result: arrived and awaiting inspection, timestamp: October 27, 2025, 09:20:00. The data digest follows the resource flow as the main thread and clearly marks the temporal logic of each step.

[0041] The generated data digests are transmitted from each edge computing node to the cloud platform. The processing unit responsible for executing step S6 logic, which may be located on an edge computing node or a dedicated aggregation node, immediately initiates the upload to the cloud platform after generating a data digest for a resource. The transmission process is based on network communication protocols. The system serializes the structured and recombined data digest object into a standard data exchange format, such as JSON. The serialized data digest is encapsulated into a network request or message via HTTP or a lightweight publish-subscribe messaging protocol. In addition to the data digest itself, the transmission request includes necessary metadata, such as the source edge computing node identifier, the timestamp of data digest generation, and the associated resource type. The system presets a data transmission timeout threshold and a retry count threshold. The data transmission timeout threshold is set based on an assessment of the average latency and reliability of the network link; for example, after assessing the typical response time of a wide area network, the data transmission timeout threshold is set to 30 seconds. The retry threshold is set based on an estimate of the probability of temporary network failures. To ensure transmission reliability while avoiding infinite retries, for example, the retry threshold is set to 3 times. When a transmission is initiated, a timer is started. If no successful response is received from the cloud platform within the data transmission timeout threshold, the transmission is retried according to the retry policy until success or the retry threshold is reached. The network transmission guarantee mechanism ensures that the data digest can be reliably delivered from the edge to the message receiving endpoint of the cloud platform.

[0042] In the cloud platform, based on the timing logic marked in the data digests, the data digests transmitted from different edge computing nodes are time-aligned and merged to complete the integration of the global business status view. The message receiving endpoint of the cloud platform continuously listens for and receives data digests transmitted from various edge computing nodes. After receiving the data digest, the message receiving endpoint of the cloud platform first performs format verification and parsing to restore the structured data digest object. The core of the integration is timing alignment and merging. Timing alignment refers to arranging the resource status change events described in different data digests on a unified, global timeline. The global timeline is the cloud-based representation of the unified clock reference that runs through the entire method. The cloud platform reads the precise timestamp field of each event sequence entry in each data digest. Based on the precise timestamp field, the cloud platform inserts all received event sequence entries from data digests belonging to different resources into a global event timeline data structure according to the order of the timestamps. The global event timeline data structure can be a time series table. The merging operation is performed on different data digest fragments of the same resource. In distributed processing, the flow of the same resource may involve multiple edge computing nodes generating local data summaries and reporting them separately. The cloud platform identifies these data summary fragments belonging to the same resource using resource identifiers. The cloud platform maintains a merge workspace with resource identifiers as keys. When the cloud platform receives a data summary, it checks the resource identifier of the data summary. If the merge workspace does not contain an entry for the resource identifier of the data summary, a new merge entry is initialized with the content of the data summary. If the merge workspace already contains an entry for the resource identifier of the data summary, the event sequence entries in the data summary are merged with the existing event sequence entries in the merge entry. During merging, insertion is performed based on the precise timestamps of the event sequence entries, and the consistency of event types and states is checked to avoid duplication or contradictions. For example, edge computing node U reports a data summary of resource Q from state W1 to state W2, and edge computing node V reports a data summary of resource Q from state W2 to state W3. The cloud platform identifies the data digests reported by edge computing node U and edge computing node V as the same resource based on the resource identifier Q, and merges the two event sequences into a complete chain from W1 to W2 to W3 based on timestamps. Ultimately, for each managed resource, the cloud platform maintains a latest, complete, time-ordered sequence of state evolution events in its merged workspace. The complete event sequences of all resources together constitute the project's global business status view. This global business status view can be provided externally in the form of a database or a graphical timeline. It accurately reflects the real-time location, status, and historical flow of all resources within the project, thereby supporting high-level resource coordination, schedule analysis, and decision-making.Through the implementation of step S6, the local perceptions generated by decentralized processing are effectively integrated into a consistent, complete, and time-accurate global view, achieving the ultimate management goal of cloud-edge collaboration.

[0043] Example 2: Figure 2 A schematic diagram of the edge computing-based real estate enterprise project management system of the present invention is provided. The edge computing-based real estate enterprise project management system includes: The data acquisition module is used to acquire business event data and local clock data collected by edge computing nodes at each project site; The clock calibration module is used to calibrate the local clock data of each edge computing node based on the network communication latency threshold between edge computing nodes, so as to obtain a unified clock reference. The association filtering module is used to extract and match resource identifiers in the data of each business event based on a unified clock reference, determine the resources corresponding to the resource identifiers, and filter out cross-node associated business events related to the flow of the same resource. The correlation assessment module is used to assess the rationality of cross-node related business events by combining the geographical distribution characteristics and process connection rules of real estate projects, and to obtain valid cross-node related business events. The timing determination module is used to analyze and determine the causal logical order between valid cross-node related business events based on the predefined state transition rules of the resources corresponding to the resource identifiers, and generate event sequence association results. The data upload module is used to standardize business event data based on the result of event sequence association, obtain data summaries, and upload the data summaries to the cloud platform for integration.

[0044] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0045] It should be noted that this invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0046] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wireless or wired transmission; wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; wireless transmission includes infrared, microwave, etc. Computer-readable storage media can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0047] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0048] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0049] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0050] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0051] If a function is implemented as a software module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0052] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0053] In conclusion, the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A real estate enterprise project management method based on edge computing, characterized in that, include: S1. Obtain business event data and local clock data collected by edge computing nodes at each project site; S2. Based on the network communication latency threshold between edge computing nodes, the local clock data of each edge computing node is calibrated to obtain a unified clock reference; S3. Based on a unified clock reference, extract and match resource identifiers in the data of each business event, determine the resources corresponding to the resource identifiers, and filter out cross-node related business events that are related to the flow of the same resource. S4. Combine the geographical distribution characteristics of real estate projects and the rules for connecting work processes to conduct a rationality assessment of cross-node related business events and obtain valid cross-node related business events. S5. Based on the predefined state transition rules of the resources corresponding to the resource identifiers, analyze and determine the causal logical order between valid cross-node related business events, and generate event sequence association results. S6. Based on the results of event sequence association, standardize the business event data to obtain a data summary, and upload the data summary to the cloud platform for integration.

2. The real estate enterprise project management method based on edge computing according to claim 1, characterized in that, S1 includes: From edge computing nodes at each project site, collect business event data reflecting material flow, equipment status changes, or process status changes; When acquiring data for each business event, the local clock data of the corresponding edge computing node is obtained synchronously.

3. The real estate enterprise project management method based on edge computing according to claim 1, characterized in that, S2 includes: Select an edge computing node as the master clock node and use its local clock data as the initial reference clock. Based on the network communication latency threshold between edge computing nodes, calculate the clock deviation compensation amount of the local clock data of each other edge computing node relative to the initial reference clock. Based on the clock offset compensation amount, the local clock data of other edge computing nodes are compensated and adjusted so that all edge computing nodes are aligned to a unified clock reference based on the compensated and adjusted local clock data.

4. The real estate enterprise project management method based on edge computing according to claim 1, characterized in that, S3 includes: Based on a unified clock reference, business event data from different edge computing nodes are preprocessed in chronological order; From preprocessed business event data, parse and extract resource identifiers that represent material, equipment, or process entities; The extracted resource identifiers are matched and compared across edge computing nodes to identify business event data with the same resource identifiers; Based on the matching results, business event data with the same resource identifier but from different edge computing nodes are aggregated as cross-node associated business events related to the same resource flow.

5. The real estate enterprise project management method based on edge computing according to claim 1, characterized in that, S4 include: Based on the geographical distribution characteristics of real estate projects, the rationality of the physical spatial path of resource flow in cross-node related business events is verified. Based on the process connection rules of real estate projects, verify the rationality of the logical order of business events in cross-node related business events; Based on the verification results of the rationality of the physical space path and the rationality of the logical order, business events that meet the verification conditions are selected from the cross-node associated business events as valid cross-node associated business events.

6. The real estate enterprise project management method based on edge computing according to claim 1, characterized in that, S5 includes: Based on predefined state transition rules, valid cross-node associated business events are parsed to reconstruct the complete state evolution chain of the resource corresponding to the resource identifier across the edge computing nodes. Based on a unified clock reference, the occurrence time of all state nodes in the reconstructed complete state evolution chain is reverse-mapped and consistency-verified with the occurrence time of each valid cross-node associated business event. Based on the results of reverse mapping and consistency verification, the causal logical order between each valid cross-node related business event that can uniquely reflect the complete state evolution chain is determined and output as the event sequence association result.

7. The real estate enterprise project management method based on edge computing according to claim 6, characterized in that, The predefined state transition rules are established based on the management process of the real estate project resource lifecycle, defining multiple discrete business states of resources and the allowed transition relationships between states.

8. The real estate enterprise project management method based on edge computing according to claim 6, characterized in that, The reconstructed complete state evolution chain of the resource corresponding to the resource identifier across the edge computing nodes includes: The effective cross-node related business events are time-sequentially ordered based on a unified clock reference. Each valid cross-node associated business event after sorting is mapped to the corresponding business state node in the predefined state transition rule; Verify and connect the state nodes mapped to adjacent business events according to predefined state transition rules, forming a complete state evolution chain of resources across the edge computing nodes.

9. The real estate enterprise project management method based on edge computing according to claim 1, characterized in that, S6 include: Based on the causal logical order indicated by the event sequence association results, the business event data is restructured in a structured manner to generate a data summary with resource flow as the main line and marked with time sequence logic; The generated data digest is transmitted from each edge computing node to the cloud platform; In the cloud platform, based on the time sequence logic marked in the data digest, the data digests transmitted by different edge computing nodes are time-series aligned and merged to complete the integration of the global business status view.

10. A real estate enterprise project management system based on edge computing, used to implement the real estate enterprise project management method based on edge computing as described in any one of claims 1-9, characterized in that, include: The data acquisition module is used to acquire business event data and local clock data collected by edge computing nodes at each project site; The clock calibration module is used to calibrate the local clock data of each edge computing node based on the network communication latency threshold between edge computing nodes, so as to obtain a unified clock reference. The association filtering module is used to extract and match resource identifiers in the data of each business event based on a unified clock reference, determine the resources corresponding to the resource identifiers, and filter out cross-node associated business events related to the flow of the same resource. The correlation assessment module is used to assess the rationality of cross-node related business events by combining the geographical distribution characteristics and process connection rules of real estate projects, and to obtain valid cross-node related business events. The timing determination module is used to analyze and determine the causal logical order between valid cross-node related business events based on the predefined state transition rules of the resources corresponding to the resource identifiers, and generate event sequence association results. The data upload module is used to standardize business event data based on the result of event sequence association, obtain data summaries, and upload the data summaries to the cloud platform for integration.

Images