Business information management method and system for digital office

Abstract

The invention discloses a business information management method and system for digital office, and relates to the technical field of digital office, and the method comprises the steps: collecting multi-source office data, carrying out the preprocessing, and constructing a dynamic evolution space-time knowledge graph; performing resource conflict prediction and conflict risk level labeling on the dynamic evolution space-time knowledge graph through a time sequence diagram convolutional network, and generating a resource conflict thermodynamic diagram; comparing the resource conflict thermodynamic diagram with a conflict threshold value, identifying a high-risk conflict area, performing alternative scheme deduction on the dynamic evolution space-time knowledge graph, and outputting an executable path adjustment suggestion; based on the executable path adjustment suggestion, carrying out alternative adjustment and digital instruction conversion on the initial office process execution chain through a process reconstruction engine to obtain a reconstructed office process execution chain; according to the invention, the cross-department cooperation efficiency of a large-scale organization is improved, the resource vacancy rate and the manpower coordination cost are reduced, and a digital office management closed loop with a self-optimization capability is formed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital office, in particular to a business information management method and system for digital office. BACKGROUND

[0002] With the acceleration of enterprise digital transformation process, intelligent office system gradually adopts resource scheduling technology based on rule engine. The existing representative scheme optimizes the allocation of static resources such as conference rooms and equipment through preset conflict rule library and heuristic search algorithm. Such mapping relationship between resource occupation state and predetermined rules can effectively identify and handle routine resource conflict problems in structured office scenarios, providing a basic automated solution for modern enterprise office resource management.

[0003] The existing technology has fundamental limitations in dealing with dynamic collaboration scenarios: preset rules are difficult to adapt to the spatio-temporal coupling conflicts emerging in cross-department collaboration (such as chain resource preemption, invisible time period overlap, etc.), resulting in a break in the prediction-decision-execution link; resource conflict identification only stays at the post-response level, and the optimization scheme lacks instruction-level penetration with digital workflow, making it difficult to build a closed-loop control loop from risk warning to process reconstruction. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a business information management method for digital office to solve the problems of real-time prediction and dynamic optimization of cross-department resource conflicts and realize closed-loop control.

[0006] To solve the above technical problems, the present application provides the following technical solutions: In a first aspect, the present application provides a business information management method for digital office, which comprises collecting and preprocessing multi-source office data, and constructing a dynamic evolution spatio-temporal knowledge graph; Through a time series diagram convolution network, resource conflict prediction and conflict risk level labeling are performed on the dynamic evolution spatio-temporal knowledge graph to generate a resource conflict heat map; The resource conflict heat map is compared with the conflict threshold to identify high-risk conflict areas, and alternative solution deduction is performed on the dynamic evolution spatio-temporal knowledge graph to output executable path adjustment suggestions; Based on the executable path adjustment suggestions, the initial office process execution chain is replaced and adjusted by a process reconstruction engine, and digital instructions are converted to obtain a reconstructed office process execution chain; The reconstructed office process execution chain is subjected to multi-dimensional efficiency analysis to obtain a high-efficiency business information management scheme.

[0007] As a preferred scheme of the business information management method for digital office, the multi-source office data comprises conference room reservation log data, equipment usage record data, employee schedule data and cross-department collaboration communication data. The preprocessing comprises data cleaning, timestamp standardization, field alignment and missing value interpolation.

[0008] As a preferred scheme of the business information management method for digital office, the dynamic evolution spatio-temporal knowledge graph is constructed by the following steps, The preprocessed multi-source office data is subjected to entity relationship extraction to obtain a time-series entity relationship set, and the time-series entity relationship set is subjected to spatio-temporal dimension enhancement to generate a spatio-temporally enhanced relationship set. The spatio-temporally enhanced relationship set is subjected to spatio-temporal fusion mode definition by a graph mode tool to obtain a spatio-temporal knowledge graph mode definition file. The spatio-temporal knowledge graph mode definition file and the spatio-temporally enhanced relationship set are combined to construct an initial dynamic evolution spatio-temporal knowledge graph. The initial dynamic evolution spatio-temporal knowledge graph is subjected to incremental update and integrity check to obtain a dynamic evolution spatio-temporal knowledge graph.

[0009] As a preferred scheme of the business information management method for digital office, the dynamic evolution spatio-temporal knowledge graph is subjected to resource conflict prediction and conflict risk level labeling by a time-series graph convolution network to generate a resource conflict heat map by the following steps, The time-series graph convolution network is subjected to multi-round iterative training by a back propagation algorithm based on a historical dynamic evolution spatio-temporal knowledge graph to obtain a trained time-series graph convolution network. The dynamic evolution spatio-temporal knowledge graph is input into the trained time-series graph convolution network for spatio-temporal feature extraction and conflict mode recognition to obtain a conflict risk probability distribution. The conflict risk probability distribution is subjected to probability interval division and risk level mapping to obtain conflict risk level labeling results, and subjected to spatial projection and color coding to generate a resource conflict heat map.

[0010] As a preferred scheme of the business information management method for digital office, the resource conflict heat map is compared with a conflict threshold value to identify a high-risk conflict area, and the dynamic evolution spatio-temporal knowledge graph is subjected to alternative solution deduction to output executable path adjustment suggestions by the following steps, The spatio-temporal distribution characteristics of historical resource conflict heat maps are extracted by a moving window statistical analysis method to obtain a conflict threshold value, and the resource conflict heat map is compared with the conflict threshold value region by region to identify a high-risk conflict area. Based on the high-risk conflict area, the resource scheduling path of the dynamic evolution space-time knowledge graph is simulated and deduced through the counterfactual reasoning algorithm, and an optional optimized resource scheduling path is output. The cost-benefit of the optional optimized resource scheduling path is evaluated, the optional optimized path with the highest comprehensive score is selected, and an executable path adjustment suggestion is generated.

[0011] As a preferred scheme of the business information management method for digital office, based on the executable path adjustment suggestion, the initial office process execution chain is adjusted and the digital instruction is converted through the process reconstruction engine to obtain the reconstructed office process execution chain, and the specific steps are as follows, The resource scheduling path sequence of the dynamic evolution space-time knowledge graph is extracted to generate the initial office process execution chain. Based on the dynamic evolution space-time knowledge graph, the resource availability of the executable path adjustment suggestion is verified to obtain the adjusted instruction set. Based on the adjusted instruction set, the initial office process execution chain is adjusted and the digital instruction is converted through the process reconstruction engine to output the reconstructed office process execution chain.

[0012] As a preferred scheme of the business information management method for digital office, based on the executable path adjustment suggestion, the initial office process execution chain is adjusted and the digital instruction is converted through the process reconstruction engine to obtain the reconstructed office process execution chain, and the specific steps are as follows, The time stamp data of the reconstructed office process execution chain is extracted, the execution time of each process approval node and the end-to-end delay are calculated, and a time efficiency analysis report is generated. The resource utilization rate of the reconstructed office process execution chain and the initial office process execution chain is compared, a resource utilization rate comparison table is generated, the number and density change of the high-risk conflict area are quantified, and a conflict improvement analysis chart is output. The cost-benefit of the optional optimized resource scheduling path is evaluated, the optional optimized path with the highest comprehensive score is selected, and an executable path adjustment suggestion is generated. The time efficiency analysis report, the resource utilization rate comparison table, the conflict improvement analysis chart and the cost-benefit evaluation list are subjected to multi-dimensional data fusion and cross-validation, and an efficient business information management scheme is output.

[0013] In a second aspect, the present application provides a business information management system for digital office, comprising, The acquisition module is used for collecting and preprocessing multi-source office data, and constructing a dynamic evolution space-time knowledge graph. The prediction module is used for predicting resource conflicts and conflict risk levels of the dynamic evolution space-time knowledge graph through a time series graph convolution network, and generating a resource conflict heat map. The deduction module is configured for comparing the resource conflict heat map with the conflict threshold, identifying a high-risk conflict area, and performing alternative solution deduction on the dynamic evolution spatiotemporal knowledge graph to output an executable path adjustment suggestion. The conversion module is configured for performing alternative adjustment and digital instruction conversion on the initial office process execution chain through the process reconfiguration engine based on the executable path adjustment suggestion, to obtain a reconfigured office process execution chain. The evaluation module is configured for performing multi-dimensional efficiency analysis on the reconfigured office process execution chain to obtain a high-efficiency business information management scheme.

[0014] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and wherein the computer program, when executed by the processor, implements any step of the business information management method for digital office according to the first aspect of the present application.

[0015] In a fourth aspect, the present application provides a computer readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any step of the business information management method for digital office according to the first aspect of the present application.

[0016] The present application has the following beneficial effects: through the construction of the dynamic evolution spatiotemporal knowledge graph and the conflict prediction based on the time series graph convolution network, the spatiotemporal coupling modeling of all elements of office resources and the intelligent conflict early warning are realized, the discrete conference room reservation, equipment scheduling and personnel schedule data form an organic knowledge network, the real-time sensing of resource state changes and the prediction of potential conflicts are realized, the visual decision support is provided for managers, the passive response management is changed into the active prevention type intelligent scheduling, the cross-department collaboration efficiency of large organizations is improved, the resource idle rate and the human coordination cost are reduced, and a self-optimizing digital office management closed loop is formed. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0018] Fig. 1 The flowchart of the business information management method for digital office.

[0019] Fig. 2 The schematic diagram of the business information management system for digital office.

[0020] Fig. 3A flowchart for generating executable path adjustment suggestions.

[0021] Fig. 4 A flowchart for outputting an efficient business information management scheme. DETAILED DESCRIPTION

[0022] In order to make the above objectives, features and advantages of the present application more apparent, specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0023] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be apparent to one skilled in the art that the present application can be practiced without the specific details given herein, that the present application can be practiced with other than the described embodiments, and that variations from the particular embodiments described herein can be made and still be within the scope of the present application.

[0024] Secondly, the term "one embodiment" or "an embodiment" as used herein means that a particular implementation can include a particular feature, structure, or characteristic, but every embodiment can not necessarily include the particular feature, structure, or characteristic. Furthermore, the following terms can be used throughout the specification and claims: includes, has, contains, etc. means that since a total of these terms can be open-ended, the terms does not exclude the possibility that any additional information, undefined in the specification, can be present in an embodiment.

[0025] SUMMARY Figs. 1-4 For one embodiment of the present application, the embodiment provides a business information management method for digitized office, comprising the following steps: S1, collecting multi-source office data for preprocessing.

[0026] S1.1, the multi-source office data includes conference room reservation log data, equipment usage record data, employee schedule data, and cross-department collaboration communication data.

[0027] S1.2, the preprocessing includes data cleaning, timestamp standardization, field alignment, and missing value interpolation.

[0028] Specifically, data cleaning: outliers in the multi-source office data are detected and removed using the interquartile range method, missing fields are filled with the historical record mean value of the same entity, duplicate entries are removed by primary key hash comparison, and unstructured text is extracted into structured fields by regular expression; Timestamp standardization: the time information in the multi-source office data is uniformly converted to ISO 8601 format, relative time descriptions are parsed into absolute timestamps, and non-uniform time zones are normalized according to the UTC reference; Field alignment: heterogeneous fields of multi-source office data are mapped to a unified naming specification, numerical values are unified in dimension, categorical variables are established with a standard coding table, and spatial location information is converted to a standard coordinate system code; Missing value imputation: the missing values in the multi-source office data are filled with a sliding window mean, the categorical variables are completed with the mode, the time series data are generated by linear interpolation, and the fields that cannot be completed are marked as specific enumeration values.

[0029] S2, constructing a dynamic evolving spatio-temporal knowledge graph.

[0030] S2.1, extracting entity relationships from the pre-processed multi-source office data to obtain a time-series entity relationship set, and enhancing the spatio-temporal dimensions to generate a spatio-temporally enhanced relationship set.

[0031] Specifically, the association relationships between conference room reservation log data, equipment usage record data, employee schedule data, and cross-department collaboration communication data entities are extracted from the pre-processed multi-source office data. The "reservation-participation" relationship is obtained by matching the employee ID between the conference room reservation log data and the employee schedule data. The "conference room-equipment" relationship is formed by associating the equipment number between the equipment usage record data and the conference room reservation log data. The "communication-schedule" relationship is formed by matching the session ID and meeting topic between the cross-department collaboration communication data and the employee schedule data. The time-series entity relationship set is formed by adding the original time stamp to the "reservation-participation" relationship, the "conference room-equipment" relationship, and the "communication-schedule" relationship. The time information in the time-series entity relationship set is converted to Unix time stamp with millisecond precision, the spatial position is converted to WGS84 coordinate system latitude and longitude, the conference room reservation is added with the duration attribute, the equipment usage record is supplemented with the movement trajectory sequence, and the employee schedule is marked with the periodic task time pattern to generate the spatio-temporally enhanced relationship set.

[0032] S2.2, defining spatio-temporal fusion patterns for the spatio-temporally enhanced relationship set using a graph pattern tool to obtain a spatio-temporal knowledge graph pattern definition file.

[0033] Specifically, the Protégé graph tool is used to receive the spatio-temporally enhanced relationship set, define the "conference room" entity class to include "location coordinates", "capacity", and other spatio-temporal attributes, and the "employee" entity class to include "department", "job level", and other business attributes. The "reservation" relationship class is formed and set with "start time constraint attribute time" and "end time", and the "use" relationship class is defined to include "device location trajectory" spatial attribute. The "reservation-participation" relationship is added with time validity constraint conditions, and an exemplary time validity constraint condition is that the reservation time must be within the employee's working hours. The "conference room-equipment" relationship is set with spatial co-location verification rules, and an exemplary spatial co-location verification rule requires that the device location must match the conference room coordinates. The defined classes, attributes, and constraints are exported as an OWL format spatio-temporal knowledge graph pattern definition file.

[0034] S2.3, combining the spatio-temporal knowledge graph pattern definition file with the spatio-temporally enhanced relationship set to construct an initial dynamic evolving spatio-temporal knowledge graph.

[0035] Specifically, the spatiotemporal knowledge graph schema definition file is imported into the TDB database of the Apache Jena framework, and the spatiotemporal enhanced relationship set is inserted into the spatiotemporal knowledge graph schema definition file according to the schema definition using the SPARQL INSERT statement; each record in the conference room reservation log data is mapped to an instance of the "conference room" class, and the "location coordinates" attribute value is filled in as the WGS84 longitude and latitude; the device usage record data is converted into an instance of the "device" class, and the "device location trajectory" attribute is set as a coordinate sequence; the employee schedule data is instantiated as a "employee" class node; the cross-department collaboration communication data is obtained as an instance of the "communication" class; the "reservation-participation" relationship instance is added with a time constraint condition through the SPARQL UPDATE, and an exemplary time constraint condition is that the "start time" must be greater than or equal to the "working hours" of the employee; the "conference room-device" relationship instance is executed with a spatial verification, and an exemplary spatial verification is a spatial inclusion calculation of the device coordinates and the conference room polygon area; after the loading of the spatiotemporal enhanced relationship set is completed, the REASONER verification is executed to ensure that the spatiotemporal enhanced relationship set meets the time constraint condition in the spatiotemporal knowledge graph schema definition file, and an initial dynamic evolution spatiotemporal knowledge graph is generated.

[0036] S2.4, incrementally updating and checking the integrity of the initial dynamic evolution spatiotemporal knowledge graph to obtain a dynamic evolution spatiotemporal knowledge graph.

[0037] Specifically, through the change event of the conference room reservation log data, exemplary events include adding a new reservation record or modifying the reservation time; after detecting the change, the timestamp and entity ID of the changed data are extracted, a delete statement is executed in the TDB database of the Apache Jena framework to delete the associated old triple, and the updated triple is inserted using the SPARQL INSERT; the device usage record data change triggers a spatial trajectory update, and an exemplary operation is to append a new coordinate point to the "device location trajectory" attribute; when the employee schedule data is changed, the time constraint associated with the reservation record is verified, and an exemplary verification is to check whether the new schedule conflicts with the existing reservation; after each update, the SHACL integrity check is executed, and an exemplary spatiotemporal constraint rule is that the number of reservations for a conference room in the same period does not exceed the upper limit of the capacity; if the spatiotemporal constraint rule is violated, the incremental update is rolled back, and an error log is recorded; the SPARQL SELECT is used to periodically scan the isolated nodes, and an exemplary scanning interval is 1 hour; the update log records the timestamp, change content and execution result of each operation to form a traceable evolution history; finally, the dynamic evolution spatiotemporal knowledge graph that passes all integrity checks is output.

[0038] S3. Resource conflict prediction and conflict risk level labeling are performed on the dynamic evolution spatio-temporal knowledge graph by a time series graph convolution network to generate a resource conflict heat map.

[0039] S3.1. Based on the historical dynamic evolution spatio-temporal knowledge graph, the time series graph convolution network is trained by a back propagation algorithm in multiple rounds to obtain a trained time series graph convolution network.

[0040] Specifically, the example is to extract the graph snapshot of the last 90 days from the historical dynamic evolution spatio-temporal knowledge graph; map the entity nodes in each graph snapshot into the vertices of the spatio-temporal graph structure, map the conference room entity into the vertex with WGS84 coordinate attribute, map the device entity into the vertex with moving trajectory sequence, and map the employee entity into the vertex with department attribute; map the "reservation-participation" relationship into the edge with time interval attribute, map the "conference room-device" relationship into the edge with space trajectory, and map the "reservation-participation" relationship into the edge with role attribute; The input layer dimension of the time series graph convolution network is configured to be consistent with the feature dimension of the conference room entity, the device entity, and the employee entity, and the example dimension is 64; a 3-layer graph convolution architecture is constructed, and the receptive field range of each layer of the graph convolution architecture is set to 5; cross entropy is used to measure the difference between the predicted conflict probability and the actual conflict label; the parameter update step is initialized to 0.01, and the Adam optimization strategy is selected; the parameter adjustment process is performed, the prediction result is output after each forward processing, and the parameters are adjusted in reverse; the example condition is that the loss change of the verification set is less than 0.001 for 10 consecutive rounds; after the parameter adjustment of the time series graph convolution network is completed, the network state is saved, and a time series graph convolution network with spatio-temporal graph processing capability is obtained.

[0041] S3.2. The dynamic evolution spatio-temporal knowledge graph is input into the trained time series graph convolution network for spatio-temporal feature extraction and conflict pattern recognition to obtain a conflict risk probability distribution.

[0042] Specifically, the entity nodes in the dynamic evolving spatio-temporal knowledge graph are converted into entity feature vectors, the WGS84 coordinates of an exemplary conference room entity are converted into 64-dimensional space encodings, the role relationship features are output, the moving trajectory sequence of a device entity is generated through an LSTM network to obtain 32-dimensional trajectory features, the department attribute of an employee entity is converted into 16-dimensional one-hot encoding to obtain employee attribute and role relationship features; the time interval attribute of the "reservation-participation" relationship is converted into the relative values of the duration and the start time, the Hausdorff distance of the "conference room-device" relationship is calculated, and the role attribute of the "reservation-participation" relationship is mapped into a 3-dimensional classification vector; the spatial convolution is performed on the conference room node features in the first layer of the time series graph convolution network, and the exemplary convolution kernel size is 5x5; the device trajectory features and the reservation relationship time features are fused in the second layer, and the gating mechanism is used to adjust the spatio-temporal feature weights; the employee attribute and role relationship features are aggregated in the third layer, and the correlation strength between employee nodes is calculated through the attention mechanism; the conflict probability value is generated for each conference room node in the output layer, and the exemplary output range is [0, 1]; the probability normalization processing is performed on the nodes of the dynamic evolving spatio-temporal knowledge graph to generate the conflict risk probability distribution covering all resource nodes; It should be noted that the expression for calculating the Hausdorff distance of the "conference room-device" relationship is as follows: ; wherein, is the Hausdorff distance of the conference room A device B relationship, a is a single element in the conference room A, b is a single element in the device B, and d(a, b) is the Euclidean distance between the points a and b.

[0043] S3.3, the conflict risk probability distribution is divided into probability intervals and risk level mapping to obtain the conflict risk level labeling result, and the spatial projection and color coding are performed to generate the resource conflict heat map.

[0044] Specifically, the conflict risk probability distribution is divided into four risk intervals, for example, low risk (0, 0.3), medium risk (0.3, 0.6), high risk (0.6, 0.9), and emergency risk (0.9, 1); the Jenks natural break algorithm is used to optimize the risk interval boundary to ensure the continuity of similar risks in spatial distribution; the risk level is mapped to four-color coding, for example, green represents low risk, yellow represents medium risk, orange represents high risk, and red represents emergency risk; based on the WGS84 coordinates in the dynamic evolution spatio-temporal knowledge graph, the four-color coding is projected onto the two-dimensional plane using the QGIS spatial analysis tool; point feature rendering is performed on the conference room nodes, line feature rendering is performed on the device trajectories, and surface feature buffer rendering is performed on the scheduled relationships; the Alpha blending algorithm is used to process the color mixing of the element overlapping area, and the example transparency is set to 60%; and the resource conflict heat map is output. It should be noted that the specific process of optimizing the risk interval boundary using the Jenks natural break algorithm is as follows: sort all conference room node probability values and calculate the inter-class variance, find the three partition points that minimize the intra-class variance when the number of classes is 4, for example, the partition points may appear at probability values of 0.28, 0.59, and 0.87, ensure that the probability value difference within each risk interval is minimized and the interval difference is maximized, and finally output the optimized risk interval boundary.

[0045] S4, compare the resource conflict heat map with the conflict threshold, identify the high-risk conflict area, and perform alternative scheme deduction on the dynamic evolution spatio-temporal knowledge graph to output executable path adjustment suggestions.

[0046] S4.1, use the moving window statistical analysis method to extract the spatio-temporal distribution characteristics of the historical resource conflict heat map, obtain the conflict threshold, and compare it with the resource conflict heat map region by region to identify the high-risk conflict area.

[0047] Specifically, extract the historical resource conflict heat map of the last 30 days, divide it into a spatio-temporal cube according to an example 1-hour time slice; define a 500m x 500m grid as a moving window unit in QGIS, and calculate the proportion of pixel points of high-risk and emergency-risk levels in each moving window unit; calculate the historical risk mean value of each moving window unit in the same period, and take the 90th percentile value as the conflict threshold; superimpose the current resource conflict heat map on the grid coordinate system to identify the moving window units with a risk level above high risk and exceeding the conflict threshold; merge the continuous over-limit moving window units, for example, the center distance between adjacent moving window units is not more than 700 meters; label the spatio-temporal range of the merged area, for example, record it as "Area A-20230705T14:00-16:00"; and output the high-risk conflict area containing the geographic location, time window, and risk intensity.

[0048] S4.2、based on the high-risk conflict area, a multi-scheme simulation and deduction is performed on the resource scheduling path of the dynamic evolution space-time knowledge graph by a counterfactual reasoning algorithm, and an optional optimized resource scheduling path is output.

[0049] Specifically, for each high-risk conflict area, the conference room entity, device entity and scheduled relationship involved in the conflict are located in the dynamic evolution space-time knowledge graph; the counterfactual condition is obtained, and an exemplary counterfactual condition includes "delaying the scheduled start time by 30 minutes" and "allocating the device entity to an idle conference room within a range of 500 meters"; a temporary copy is generated by copying the current graph state in the dynamic evolution space-time knowledge graph, the original attributes of all entities and relationships are retained, and the scheduled relationship time attribute and device entity location attribute are allowed to be modified to obtain a resource scheduling deduction temporary copy. The time attribute of the "scheduled-participation" relationship or the location attribute of the device entity; check if the new time exceeds the employee's working hours; perform a shortest path algorithm to calculate the device entity movement cost, and an exemplary device entity movement path distance weight is set to 1.5 times the time weight; evaluate the conflict risk reduction degree for each feasible scheme, and an exemplary requirement is that the high-risk area area is reduced by more than 50%; retain the top 3 feasible schemes that meet the counterfactual condition, record the adjustment content, implementation cost and expected effect of the top 3 feasible schemes; and output the optional optimized resource scheduling path.

[0050] S4.3, cost-benefit evaluation is performed on the optional optimized resource scheduling path, the optional optimized path with the highest comprehensive score is screened, and an executable path adjustment suggestion is generated.

[0051] Specifically, evaluation indexes are defined for each optional optimized resource scheduling path, and exemplary evaluation indexes include time adjustment cost (minutes), device movement distance (meters), and conflict risk reduction rate (percent); the evaluation index weight is set, and an exemplary weight distribution is that the time cost accounts for 40%, the movement distance accounts for 30%, and the risk reduction accounts for 30%; the time adjustment cost score, device movement distance score, risk reduction rate score and comprehensive score in the evaluation indexes of each optional optimized resource scheduling path are calculated; the comprehensive scores of all optional optimized resource scheduling paths are sorted, and the paths with scores exceeding an exemplary 85 points are screened; the resource availability of the highest-score optional optimized resource scheduling path is verified, and it is checked whether the conference room is occupied by other scheduled during the adjustment period; an executable path adjustment suggestion containing specific adjustment time, device movement route and expected risk value change is generated; It should be noted that the expression for calculating the time adjustment cost score of the optional optimized resource scheduling path is: ; wherein T is the time adjustment cost score of the optional optimized resource scheduling path, t is the actual time adjustment cost, is the allowed maximum time adjustment cost amount (an exemplary value of 120 minutes); The expression for calculating the alternative optimized resource scheduling path equipment movement distance score is: ; wherein D is the alternative optimized resource scheduling path equipment movement distance score, d is the actual equipment movement distance, is the allowed maximum movement distance (an exemplary value of 500 meters); The expression for calculating the alternative optimized resource scheduling path risk reduction rate score is: ; wherein R is the alternative optimized resource scheduling path risk reduction rate score, r is the actual risk reduction rate, is the expected maximum risk reduction rate (an exemplary value of 50%); The expression for calculating the alternative optimized resource scheduling path comprehensive score is: ; wherein S is the alternative optimized resource scheduling path comprehensive score, is the time adjustment cost score weight exemplary 0.4, is the equipment movement distance score weight exemplary 0.3, is the risk reduction rate score weight exemplary 0.3.

[0052] S5, based on the executable path adjustment suggestion, the initial office process execution chain is subjected to alternative adjustment and digital instruction conversion through a process reconfiguration engine, to obtain a reconfigured office process execution chain.

[0053] S5.1, resource scheduling path sequence extraction is performed on the dynamic evolution spatiotemporal knowledge graph, to generate an initial office process execution chain.

[0054] Specifically, all "scheduled-participation" relationships in the dynamic evolution spatiotemporal knowledge graph are located, and are sorted according to scheduled start time to form a time sequence; conference room entity location coordinates and equipment entity use records associated with the "scheduled-participation" relationships are extracted to generate "scheduled ID-time window-conference room coordinates-equipment trajectory" basic data units; it is checked whether adjacent scheduled events exist in spatiotemporal overlap, and an exemplary determination condition is that the time overlap exceeds 15 minutes and the spatial distance is less than 200 meters; a dependency relationship is established for scheduled events in conflict to form a conflict constraint chain of "scheduled A scheduled B"; the equipment movement path and the scheduled time window are combined to generate a movement constraint of "the equipment must be completed from location X to location Y within time T"; the dynamic evolution spatiotemporal knowledge graph, the conflict constraint chain, and the movement constraint are integrated according to the time line, and are output as an initial office process execution chain containing spatiotemporal constraint conditions.

[0055] S5.2, verify the resource availability of the executable path adjustment suggestion based on the dynamic evolving spatio-temporal knowledge graph, and obtain an over-check adjustment instruction set.

[0056] Specifically, the pre-determined ID involved in the executable path adjustment suggestion is located in the dynamic evolving spatio-temporal knowledge graph, and the latest state of the associated conference room entity and device entity is queried. It is verified whether the conference room entity is occupied by other pre-determinations during the adjustment period, and it is exemplarily determined to be in conflict if the time window overlap exceeds 5 minutes. The device movement path feasibility is checked, and it is exemplarily required that the movement distance within 500 meters must be completed within 15 minutes. The employee schedule data is checked again to confirm that the adjusted pre-determined time is still within the employee working period, and the exemplarily defined working period is 8:00-18:00. It is detected by SPARQL query whether it violates the spatio-temporal constraint rules in the dynamic evolving spatio-temporal knowledge graph, and it is exemplarily detected whether the conference room entity capacity is over limit. The executable path adjustment suggestion that passes all verification conditions is marked as a valid instruction, and the executable path adjustment suggestion that exists conflict or violation is filtered. An over-check adjustment instruction set containing valid pre-determined time modification, device movement path and associated constraint conditions is generated.

[0057] S5.3, based on the over-check adjustment instruction set, the initial office process execution chain is replaced by the alternative adjustment and digital instruction conversion through the process reconfiguration engine, and the reconfigured office process execution chain is output.

[0058] Specifically, the over-check adjustment instruction set is processed by a structured query method to identify data nodes containing "pre-determined number"; by extracting the pre-determined number that needs to be modified (exemplary format is "pre-determined-year-month-day-number" such as "pre-determined-20230705-001"), while obtaining the modification corresponding adjustment parameter combination, including time parameter (adjusted start / end time), space parameter (new device position coordinates) and coverage parameter; the target pre-determined event node is located in the initial office process execution chain, and it is exemplarily quickly found by pre-determined ID hash matching; the original time window attribute is replaced by the adjusted time interval, and the exemplarily time format is kept in ISO 8601 standard; the device movement path node coordinate sequence is updated, and an exemplarily trajectory point is recorded every 5 seconds; an adjustment mark field is added to record the change reason, and it is exemplarily marked as "conflict avoidance-time adjustment" or "conflict avoidance-path optimization"; the spatio-temporal constraint conditions are converted into machine instructions, and it is exemplarily converted into "move_speed=33.33 meters / minute" that "the device must move 500 meters within 15 minutes"; the time margin of each key path node in the process chain is recalculated, and it is exemplarily required that the key path node margin is not less than 10 minutes; and the reconfigured office process execution chain is generated.

[0059] S6. Perform multi-dimensional efficiency analysis on the reconstructed office process execution chain to obtain an efficient business information management scheme.

[0060] S6.1. Extract timestamp data from the reconstructed office process execution chain, calculate the execution time of each process approval node and the end-to-end delay, and generate a time efficiency analysis report.

[0061] Specifically, the start time stamp and end time stamp of each approval node are extracted from the reconstructed office process execution chain, the example time stamp format is ISO 8601 standard; the processing time of each approval node is obtained, the starting node and the terminating node on the critical path of the process are identified, and the end-to-end time range is determined; the time consumption of each functional department is counted, the example statistics include the average processing time of the administrative department and the maximum processing time of the financial department; the time data of the initial office process execution chain is compared, the time difference percentage before and after optimization is calculated; the time data is classified and summarized according to the process stage, the example stage is divided into "application submission-department approval-resource allocation-execution confirmation", and the time efficiency analysis report is generated.

[0062] S6.2. Compare the resource utilization rate of the reconstructed office process execution chain with the initial office process execution chain, generate a resource utilization rate comparison table, and quantify the number and density changes of high-risk conflict areas, output conflict improvement analysis charts.

[0063] Specifically, the use frequency of each conference room entity in the reconstructed office process execution chain during the period of 8:00-20:00 is counted, and the example is counted by the number of times per hour interval; the conference room utilization rate of the initial office process execution chain during the same period is calculated; the total moving distance of equipment entities in the reconstructed office process execution chain and the initial office process execution chain is compared; identify high-risk conflict areas in the resource conflict heat map, count the number of conflict events in the same area in the reconstructed process chain; obtain the conflict density per unit area, for example, count the number of conflict events per 100 square meters; draw a superimposed comparison chart of the heat map, for example, use red gradient to represent the conflict density change before and after optimization; generate a structured table containing conference room utilization rate comparison data, equipment movement efficiency improvement percentage and conflict area improvement degree; output conflict improvement analysis charts.

[0064] S6.3. Calculate the human work hours and losses of the areas to be optimized in the conflict improvement analysis chart to obtain a cost-benefit evaluation list.

[0065] Specifically, in the conflict improvement analysis chart, the area to be optimized is located, and the conflict improvement analysis chart area boundary and time range are extracted; the hourly wage standard of the associated post is queried from the enterprise human resource database, and the example data includes 50 yuan / hour for administrative positions and 80 yuan / hour for technical positions; the total length of time of human waiting caused by the conflict events in the conflict improvement analysis chart area boundary before optimization is counted, and the example usage rate standard of conference rooms and equipment is 200 yuan / hour for conference rooms and 30 yuan / hour for projection equipment; the saved resource idle time after optimization is obtained, the total cost saving brought by process optimization is calculated, and the optimization input cost is compared, and the example includes transformation cost allocation and training cost; and a cost-benefit evaluation list including annual saving amount, investment return period and net present value is generated.

[0066] S6.4, multi-dimensional data fusion and cross verification are performed on the time efficiency analysis report, the resource utilization rate comparison table, the conflict improvement analysis chart and the cost-benefit evaluation list, and an efficient business information management scheme is output.

[0067] Specifically, data correlation mapping of the time efficiency analysis report, the resource utilization rate comparison table, the conflict improvement analysis chart and the cost-benefit evaluation list is established, and an example correlation key is established by a predetermined number and a conference room number; the end-to-end delay improvement rate in the time efficiency analysis report and the department approval efficiency improvement value are extracted; the conference room usage rate improvement percentage in the resource utilization rate comparison table is matched with the time efficiency data; the number of high-risk areas in the conflict improvement analysis chart is cross-verified with the cost saving amount in the cost-benefit evaluation list; a multi-dimensional evaluation matrix is constructed, and an example multi-dimensional evaluation matrix includes a time efficiency dimension (weight 40%), a resource utilization dimension (weight 30%) and an economic benefit dimension (weight 30%); a comprehensive optimization index of the time efficiency dimension (weight 40%), the resource utilization dimension (weight 30%) and the economic benefit dimension (weight 30%) is calculated, and special cases of time efficiency improvement but resource utilization rate decrease are checked; and an efficient business information management scheme including global evaluation of the optimization scheme, implementation priority suggestion and risk warning is generated.

[0068] The embodiment also provides a business information management system for digital office, comprising: a collection module configured to collect multi-source office data for preprocessing and construct a dynamic evolution spatio-temporal knowledge graph; a prediction module configured to perform resource conflict prediction and conflict risk level labeling on the dynamic evolution spatio-temporal knowledge graph through a time series graph convolution network, and generate a resource conflict heat map; a deduction module configured to compare the resource conflict heat map with a conflict threshold, identify a high-risk conflict area, and deduce a replacement scheme for the dynamic evolution spatio-temporal knowledge graph, and output an executable path adjustment suggestion; The conversion module is configured to perform alternative adjustment and digital instruction conversion on the initial office process execution chain based on the executable path adjustment suggestion through the process reconfiguration engine, so as to obtain a reconfigured office process execution chain. The evaluation module is configured to perform multi-dimensional efficiency analysis on the reconfigured office process execution chain, so as to obtain a high-efficiency business information management scheme.

[0069] The embodiment also provides a computer device suitable for the business information management method for digital office, which comprises a memory and a processor.

[0070] The computer device can be a terminal, and the computer device comprises a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is configured to perform wired or wireless communication with an external terminal. The wireless communication can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0071] The embodiment also provides a storage medium on which a computer program is stored, the program being executed by a processor to implement the service information management method for digital office proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk, or an optical disk.

[0072] To sum up, the application realizes the spatio-temporal coupling modeling of office resources and the intelligent conflict early warning by constructing a dynamic evolution spatio-temporal knowledge graph and a conflict prediction based on a time series graph convolution network, makes discrete conference room reservation, equipment scheduling, and personnel schedule data form an organic knowledge network, provides visual decision support for managers by real-time sensing of resource state changes and prediction of potential conflicts, changes passive response management into active prevention type intelligent scheduling, improves the cross-department collaboration efficiency of large organizations, reduces the resource idle rate and human coordination cost, and forms a self-optimizing digital office management closed loop.

[0073] It should be noted that the above embodiments are only used to illustrate the technical solutions of the application rather than limit the application. Although the application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the application, and all should be covered in the scope of the claims of the application.

Claims

1. A method for managing business information for digitized office, characterized by: The application relates to an office resource conflict prediction and adjustment method based on dynamic evolution spatio-temporal knowledge graph. The method comprises the following steps: Collecting multi-source office data for preprocessing and constructing a dynamic evolution spatio-temporal knowledge graph; Performing resource conflict prediction and conflict risk level labeling on the dynamic evolution spatio-temporal knowledge graph through a time series graph convolution network to generate a resource conflict heat map; Comparing the resource conflict heat map with a conflict threshold to identify a high-risk conflict area and deduce a substitute scheme for the dynamic evolution spatio-temporal knowledge graph to output an executable path adjustment suggestion; Based on the executable path adjustment suggestion, performing alternative adjustment and digital instruction conversion on the initial office process execution chain through a process reconstruction engine to obtain a reconstructed office process execution chain; 2. The business information management method for digital office according to claim 1, wherein: Performing multi-dimensional efficiency analysis on the reconstructed office process execution chain to obtain a high-efficiency business information management scheme. The multi-source office data comprises conference room reservation log data, equipment usage record data, employee schedule table data and cross-department collaboration communication data; 3. The business information management method for digital office according to claim 2, wherein: The preprocessing comprises data cleaning, timestamp standardization, field alignment and missing value interpolation. The dynamic evolution spatio-temporal knowledge graph is constructed through the following steps: Extracting entity relationships from the preprocessed multi-source office data to obtain a time-series entity relationship set and performing spatio-temporal dimension enhancement to generate a spatio-temporal enhanced relationship set; Defining a spatio-temporal fusion mode for the spatio-temporal enhanced relationship set through a graph mode tool to obtain a spatio-temporal knowledge graph mode definition file; Combining the spatio-temporal knowledge graph mode definition file with the spatio-temporal enhanced relationship set to construct an initial dynamic evolution spatio-temporal knowledge graph; 4. The business information management method for digitized offices according to claim 3, characterized by: Performing incremental updating and integrity checking on the initial dynamic evolution spatio-temporal knowledge graph to obtain a dynamic evolution spatio-temporal knowledge graph. The resource conflict prediction and conflict risk level labeling on the dynamic evolution spatio-temporal knowledge graph through the time series graph convolution network to generate the resource conflict heat map are performed through the following steps: Based on the historical dynamic evolution spatio-temporal knowledge graph, the time series graph convolution network is trained through a back propagation algorithm for multiple rounds of iteration to obtain a trained time series graph convolution network; The dynamic evolution spatio-temporal knowledge graph is input into the trained time series graph convolution network for spatio-temporal feature extraction and conflict mode identification to obtain a conflict risk probability distribution; 5. The business information management method for digitized offices according to claim 4, characterized by: The conflict risk probability distribution is divided into probability intervals and risk level mapping to obtain a conflict risk level labeling result, which is projected and color-coded to generate a resource conflict heat map. The resource conflict heat map is compared with the conflict threshold to identify a high-risk conflict area, and a substitute scheme is deduced for the dynamic evolution spatio-temporal knowledge graph to output an executable path adjustment suggestion through the following steps: The spatio-temporal distribution characteristics of the historical resource conflict heat map are extracted through a moving window statistical analysis method to obtain a conflict threshold, which is compared with the resource conflict heat map region by region to identify a high-risk conflict area; Based on the high-risk conflict area, the resource scheduling path of the dynamic evolution spatio-temporal knowledge graph is simulated and deduced through a counterfactual reasoning algorithm to output an alternative optimized resource scheduling path; The alternative optimized resource scheduling path is evaluated in terms of cost and benefit, and the alternative optimized path with the highest comprehensive score is selected to generate an executable path adjustment suggestion.

6. The business information management method for digitized offices according to claim 5, characterized by: The executable path adjustment suggestion is used to perform alternative adjustment and digital instruction conversion on the initial office process execution chain through a process reconstruction engine to obtain a reconstructed office process execution chain, and the specific steps are as follows, The resource scheduling path sequence of the dynamic evolution space-time knowledge graph is extracted to generate an initial office process execution chain. The resource availability of the executable path adjustment suggestion is verified based on the dynamic evolution space-time knowledge graph to obtain a verification adjustment instruction set. Based on the verification adjustment instruction set, the initial office process execution chain is adjusted and converted into digital instructions through the process reconstruction engine to output the reconstructed office process execution chain.

7. The business information management method for digitized offices according to claim 6, characterized by: The reconstructed office process execution chain is subjected to multi-dimensional efficiency analysis to obtain a high-efficiency business information management scheme, and the specific steps are as follows, The reconstructed office process execution chain is subjected to timestamp data extraction, calculation of the execution time of each process approval node and end-to-end delay, and generation of a time efficiency analysis report. The reconstructed office process execution chain is compared with the initial office process execution chain in terms of resource utilization rate to generate a resource utilization rate comparison table, and the number and density changes of high-risk conflict areas are quantified to output a conflict improvement analysis chart. The areas requiring optimization in the conflict improvement analysis chart are subjected to human work hour and loss accounting to obtain a cost-benefit assessment list. The time efficiency analysis report, the resource utilization rate comparison table, the conflict improvement analysis chart and the cost-benefit assessment list are subjected to multi-dimensional data fusion and cross-validation to output a high-efficiency business information management scheme.

8. A business information management system for digital office based on the business information management method for digital office according to any one of claims 1 to 7, characterized by: It includes, The acquisition module is used to acquire and preprocess multi-source office data and construct a dynamic evolution space-time knowledge graph; The prediction module is used to perform resource conflict prediction and conflict risk level labeling on the dynamic evolution space-time knowledge graph through a time series graph convolution network to generate a resource conflict heat map; The reasoning module is used to compare the resource conflict heat map with a conflict threshold, identify high-risk conflict areas, and perform alternative scheme reasoning on the dynamic evolution space-time knowledge graph to output an executable path adjustment suggestion; The conversion module is used to perform alternative adjustment and digital instruction conversion on the initial office process execution chain based on the executable path adjustment suggestion through a process reconstruction engine to obtain a reconstructed office process execution chain; The evaluation module is used to perform multi-dimensional efficiency analysis on the reconstructed office process execution chain to obtain a high-efficiency business information management scheme. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to realize the steps of the business information management method for digital office in any one of claims 1-7.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the business information management method for digital office in any one of claims 1-7.

Images