A multi-level linkage command decision method and system based on digital twinning

Abstract

The application discloses a multi-level linkage commanding decision method and system based on digital twinning, which comprises the following steps: constructing a city-wide three-dimensional digital twinning sand table; multi-source data fusion dynamic situation awareness; resource scheduling demand prediction based on machine learning; multi-objective optimization cross-department resource matching and task intelligent assignment; four-level linkage instruction issuing and progress dynamic tracking; disposal closed-loop feedback and intelligent review optimization.The application has the beneficial effects that: relying on the digital twinning three-dimensional sand table, full-factor situation visualization is realized, the information limitation of traditional two-dimensional commanding is broken, an immersive, spatialized situation carrier is provided for commanding decision, and the whole process of city operation and disposal is intuitively presented; through a machine learning prediction model and a multi-objective optimization algorithm, intelligent decision of resource scheduling is realized, traditional artificial experience scheduling is replaced, resource matching accuracy is improved, response speed, resource utilization rate and scheduling cost are taken into account, and the cross-department resource collaboration problem is effectively solved.

Description

Technical Field

[0001] This invention relates to the field of urban governance and emergency command technology, and in particular to a multi-level linkage command and decision-making method and system based on digital twins. Background Technology

[0002] With the acceleration of urbanization, the demands for cross-departmental collaborative command in daily urban operation and management, emergency response, and major event support are constantly increasing. Traditional joint command systems suffer from the following core problems: First, data silos are prominent, with each commission and bureau's business systems operating independently, hindering efficient data sharing and resulting in a lack of comprehensive data support for command decisions. Second, the efficiency of command hierarchy transmission is low, with numerous steps involved in the flow of instructions between the city, districts, sub-districts / townships, and the field, leading to insufficient response speed to emergencies. Third, resource allocation relies on manual experience, lacking quantitative algorithm support, resulting in resource misallocation, unreasonable scheduling paths, and low collaborative efficiency. Fourth, the level of situational awareness visualization is insufficient; traditional two-dimensional maps cannot intuitively present the complex urban space and operational status, and command decisions lack an immersive, comprehensive situational awareness platform. Fifth, the closed-loop response is inadequate; event debriefing relies on manual processing, and experience cannot be effectively distilled into reusable scheduling rules.

[0003] Therefore, we need to develop a multi-level linkage command and decision-making method and system based on digital twins. Summary of the Invention

[0004] This invention provides a multi-level linkage command and decision-making method and system based on digital twins. It achieves full-domain situational awareness through a digital twin 3D sand table, realizes intelligent resource scheduling by relying on machine learning and multi-objective optimization algorithms, builds a four-level linkage architecture to improve collaborative efficiency, and finally forms a closed-loop command and decision-making system.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A multi-level coordinated command and decision-making method based on digital twins includes the following steps: S1. Construct a three-dimensional digital twin sand table for the entire city: Based on the city's high-precision geographic information, topography, building facilities and administrative division data, build a three-dimensional real-scene digital twin base covering the three-level administrative units of city-district-street and township, and map the spatial attributes and operational status of emergency resources, protection targets and infrastructure throughout the entire area; S2. Dynamic situational awareness through multi-source data fusion: Data from IoT sensing devices, government business systems, and manual reporting are accessed, cleaned, standardized, and fused with spatial coordinates, and then mapped to the digital twin sand table in real time to achieve dynamic updates of all elements of the city's daily operations, emergencies, and major events. S3. Resource scheduling demand prediction based on machine learning: Extract historical event datasets, real-time situation parameters and environmental feature factors, input them into the pre-trained resource demand prediction model, and output the prediction results of the type, quantity and optimal deployment location of resources required in the current scenario. S4. Multi-objective optimization of cross-departmental resource matching and intelligent task assignment: Using resource demand prediction results as input, a multi-objective optimization scheduling function is constructed. Combined with the list of responsibilities of each linked department and the real-time availability of resources, the function completes the accurate matching of cross-departmental resources and the intelligent assignment of tasks, generating the global optimal scheduling scheme. S5, Four-level linkage command issuance and dynamic progress tracking: Relying on the four-level linkage command architecture of city-district-street-township-site, dispatch instructions can be issued to the corresponding disposal entities with one click. The disposal progress, resource location and on-site feedback information are synchronized in real time through the digital twin sand table, realizing the visualized control of the entire disposal process; S6. Closed-loop feedback and intelligent review optimization: After the event is completed, the entire process data is automatically archived, a review report is generated based on a multi-dimensional performance evaluation model, and the resource demand prediction model and scheduling rules are iteratively optimized to form a closed-loop management of the entire process.

[0006] The construction process of the multi-objective optimization scheduling function in step S4 is as follows: taking the minimization of scheduling response time, the maximization of resource utilization, and the minimization of total scheduling cost as optimization objectives, a weighted multi-objective optimization objective function F(x) is constructed, and the calculation formula is as follows: ; In the formula The overall response time corresponding to scheduling scheme x is the sum of the time taken for all assigned task personnel / resources to arrive at the event site. The comprehensive utilization rate of resources is taken as the weighted average of the load rates of various resources participating in the scheduling. The total dispatch cost includes labor costs, material loss costs, and transportation costs. These are the weighting coefficients corresponding to response time, resource utilization, and scheduling cost, respectively, and satisfy the following conditions: The weighting coefficients are dynamically adjusted according to the event level and scenario type; the constraints of the multi-objective optimization function include: resource quantity constraints (the total amount of various resources called does not exceed the current available stock), timeliness constraints (the processing time of a single task does not exceed the preset level threshold), and spatial coverage constraints (the service radius of the scheduled resources completely covers the event points).

[0007] The resource demand prediction model described in step S3 uses a gradient boosting tree ensemble learning model, and the calculation formula for its prediction output is as follows: ; In the formula The output for resource demand forecasting includes prediction results across three dimensions: resource type, resource quantity, and deployment location coordinates; K represents the total number of regression decision trees in the ensemble model. Let be the output function corresponding to the k-th regression decision tree; is the function space formed by all regression decision trees; the input feature vector x of the model contains event type, event risk level, population density of the event area, surrounding resource stock, meteorological environmental parameters, terrain conditions and historical resource consumption statistics of similar events.

[0008] In step S6, the multi-dimensional effectiveness evaluation model adopts a weighted comprehensive scoring method. The formula for calculating the comprehensive score S of the disposal effectiveness is as follows: ; In the formula, R is the response timeliness score, calculated by normalizing the ratio of the time from the issuance of the instruction to the first response on site to the standard time; E is the handling efficiency score, calculated by normalizing the ratio of the actual total handling time of the event to the standard handling time of the same level event; C is the cross-departmental collaboration effect score, calculated by weighting the cross-departmental task completion rate and information synchronization timeliness rate; and D is the loss control effect score, calculated by normalizing the ratio of the actual loss caused by the event to the estimated loss. These are the weight coefficients for the four evaluation dimensions, and they satisfy... A higher overall score indicates better event handling efficiency.

[0009] The digital twin sand table configuration described in step S1 has a seamless switching mechanism between normal and emergency situations: in the normal operation state, it displays the operational status of all elements of urban management; when an emergency is triggered or a major event is launched, it can switch to the corresponding special command mode with one click, prioritizing the display of event locations, surrounding emergency resources and disposal links, and hiding irrelevant business layers to adapt to the command focus needs of different scenarios.

[0010] The multi-source data mentioned in step S2 specifically includes: IoT sensing data collected by video surveillance, environmental monitoring sensors, geological disaster monitoring equipment, and traffic flow collection equipment; business system data from public security, transportation, health, emergency management, and urban management departments; and on-site text, image, and audio / video data reported by mobile terminals. All data are standardized and then uniformly converted into the CGCS2000 national geodetic coordinate system to complete the spatial mapping with the digital twin sand table.

[0011] The four-level linkage command architecture described in step S5 adopts an instruction flow mode that combines hierarchical dispatching and direct dispatching to higher levels: routine events are dispatched and handled level by level from city to district to street / township; major and above-level emergency events support the city-level command center to directly dispatch orders to on-site personnel, while simultaneously synchronizing instruction information to the corresponding district and street / township management nodes to ensure that the instruction response delay for emergencies is ≤3 seconds.

[0012] Step S2, dynamic situational awareness, also includes intelligent early warning function: it automatically identifies abnormal events such as street vendors, pedestrian congestion, fire, and potential geological disasters through video AI recognition algorithms, automatically generates early warning work orders and pushes them to the command terminal, and marks the early warning location and risk level in the digital twin sand table with differentiated colors and flashing effects.

[0013] In step S5, the dynamic tracking of the handling progress is as follows: On-site personnel report the status of handling nodes, on-site audio and video materials, and new resource requirements in real time through mobile terminals. The system automatically updates the event status and resource location in the digital twin sandbox. The entire handling process is recorded and can be traced back to view the complete handling chain by timeline.

[0014] A multi-level collaborative command and decision-making system based on digital twins includes: The digital twin sand table module is used to build a three-dimensional real-scene digital twin base for the entire city, and to complete the digital mapping of all elements of geographic information, emergency resources, infrastructure and protection targets. The situation awareness module is used to access and integrate multi-source IoT sensing data, government business data and on-site reported data to achieve dynamic updates of the city's operational status and intelligent early warning of abnormal events. The intelligent prediction module has a built-in resource demand prediction model based on gradient boosting tree, which is used to output the prediction results of resource type, quantity and deployment location in the corresponding scenario according to the real-time situation. The intelligent scheduling module has a built-in multi-objective optimization scheduling model, which is used to accurately match cross-departmental resources and intelligently assign tasks, generating the globally optimal scheduling scheme. The four-level linkage command module covers command nodes at the city, district, street / township, and on-site levels, supporting one-click issuance of instructions from multiple channels, integrated communication dispatch, and real-time tracking of handling progress. The debriefing and optimization module is used to generate multi-dimensional handling effectiveness evaluation reports after the incident is closed, iteratively optimizing predictive model parameters and dispatch rules to continuously improve handling capabilities. Compared with the prior art, the present invention has the following advantages: 4. Relying on the digital twin 3D sand table, we can achieve full-element situation visualization, break through the information limitations of traditional two-dimensional command, provide an immersive and spatial situation carrier for command and decision-making, and intuitively present the entire process of urban operation and disposal.

[0015] 5. By using machine learning prediction models and multi-objective optimization algorithms, intelligent decision-making for resource scheduling is achieved, replacing traditional manual experience-based scheduling, improving the accuracy of resource matching, and taking into account response speed, resource utilization, and scheduling costs, effectively solving the problem of cross-departmental resource collaboration.

[0016] 6. Establish a four-level linkage structure of city-district-street / township-on-site, and combine a tiered dispatching and direct dispatching mechanism to compress the instruction flow process, improve the response speed to emergencies, and break through hierarchical barriers.

[0017] 7. Construct a closed-loop system covering the entire process of "perception-prediction-scheduling-response-review". Through intelligent review, continuously iterate and optimize models and rules to achieve continuous improvement in command capabilities and adapt to the needs of multiple scenarios such as urban governance, emergency response, and major event support. Attached Figure Description

[0018] Figure 1 This is a flowchart of the multi-level linkage command and decision-making method of the present invention; Figure 2 This is a block diagram of the multi-level linkage command and decision-making system module of the present invention; Figure 3 This is a schematic diagram of the unattended collaborative command system of the present invention; Figure 4 This is a schematic representation of the duty calendar in the collaborative command business system of the present invention; Figure 5 This is a schematic diagram of the duty log of the collaborative command business system of the present invention; Figure 6 This is a schematic diagram of the template management for the collaborative command business system of the present invention; Figure 7 This is a schematic diagram of the event reporting system of the collaborative command business system of the present invention; Figure 8 This is a schematic diagram of the event flow in the collaborative command business system of the present invention; Figure 9 This is a schematic diagram illustrating the adaptation of the collaborative command business system solution of the present invention; Figure 10 This is a schematic diagram of the contingency plan handling of the collaborative command business system of the present invention; Figure 11 This is a schematic diagram of the task handling process of the collaborative command and control system of the present invention; Figure 12 This is a schematic diagram of the alarm list of the collaborative command business system of the present invention; Figure 13 This is a schematic diagram of the alarm handling system of the collaborative command business system of the present invention; Figure 14 This is a schematic diagram of a sand table exercise for the intelligent solution of the collaborative command business system of the present invention; Figure 15 This is a schematic diagram of the multi-department collaborative stress test estimation of the present invention; Figure 16 This is a schematic diagram illustrating the scheme demonstration and evaluation of the present invention; Figure 17 This is a schematic diagram of the flashing warning at emergency locations according to the present invention; Figure 18 This is a schematic diagram of the evaluation report for the present invention; Figure 19 This is a schematic diagram of the information aggregation of a single command and dispatch map according to the present invention; Figure 20 This is a schematic diagram of the command and dispatch map application of the present invention; Figure 21 This is a schematic diagram of the analytical function of the present invention; Figure 22 This is a schematic diagram of the command and dispatch of the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand and implement the technical solutions of the present invention, the present invention will be further described below in conjunction with specific embodiments and accompanying drawings. However, the embodiments described are not intended to limit the present invention.

[0020] like Figures 1 to 22 As shown, a multi-level collaborative command and decision-making method based on digital twins is described, and the method steps are as follows: Step S1: Construct a 3D digital twin sand table of the entire city Integrating high-precision GIS geographic data, DEM digital elevation models, 3D building models, CIM urban information model data, and administrative division data, a comprehensive 3D digital twin platform covering the city, district, and township levels is constructed. This platform maps the spatial location, attribute information, and real-time operational status of emergency resources (rescue teams, material warehouses, shelters, medical points, emergency vehicles), key protection targets, and infrastructure (roads, pipelines, monitoring equipment) across the entire region. It supports basic map operations such as layer visibility control, spatial measurement, and point retrieval. The sand table features a seamless switching mechanism between normal and emergency situations: in normal operation, it displays the overall urban management situation, including the city's appearance, traffic flow, and municipal facilities; when a sudden event is triggered or a major event is launched, it switches to the corresponding specialized command mode with one click, automatically focusing on the event / event area, prioritizing the display of event locations, surrounding emergency resources, and response links, while hiding irrelevant business layers, adapting to the command focus needs of different scenarios.

[0021] Step S2: Dynamic situational awareness through multi-source data fusion; Access three types of multi-source data through standardized interfaces: IoT sensing data: including data from front-end devices such as video surveillance, environmental monitoring sensors, geological disaster monitoring equipment, and traffic flow collection equipment, supporting access to standard protocols such as GB28181; Government business data: Connect with the business systems of public security, transportation, health, emergency management, urban management and other commissions and bureaus to obtain business information such as event data, resource data and police report data; On-site data reporting: Mobile terminals receive on-site information, including text, images, audio, and video, reported by frontline personnel. All incoming data undergoes data cleaning (duplicate removal, error correction, and completion), standardization (unified data encoding and format), and spatial coordinate transformation (converted to the CGCS2000 national geodetic coordinate system) before being synchronously mapped to the digital twin sand table in real time, enabling dynamic updates of the city's operational status. Simultaneously, an intelligent early warning function is configured: Video AI recognition algorithms automatically identify abnormal events such as street vendors, pedestrian congestion, fires, and potential geological hazards, automatically generating early warning work orders and pushing them to the command center. In the digital twin sand table, risk warnings are marked with differentiated colors (red, orange, yellow, and blue) and flashing effects, corresponding to levels one to four.

[0022] Step S3: Resource scheduling demand prediction based on machine learning A gradient boosting tree (GBDT) ensemble learning model for predicting resource demand is constructed. The model is pre-trained using a dataset of historical events from the past five years, with labels including event type, risk level, regional characteristics, environmental parameters, and resource consumption. Upon receiving a warning or report, the model extracts the event type, risk level, population density of the affected area, surrounding resource reserves, meteorological parameters, terrain conditions, and historical resource consumption statistics for similar events as input feature vectors. These vectors are then fed into the pre-trained model, which outputs three prediction results: the type and quantity of resources required in the current scenario, and the coordinates of the optimal deployment point. This provides a quantitative basis for subsequent scheduling.

[0023] Step S4: Multi-objective optimization of cross-departmental resource matching and intelligent task assignment Using resource demand forecasts as input, and combining the lists of responsibilities of various departments, real-time availability of resources, and spatial location, a weighted multi-objective optimization scheduling function is constructed to solve for the globally optimal scheduling scheme.

[0024] The objective function is ,in For the overall response time, To improve the comprehensive utilization rate of resources, Total scheduling cost; weight The alert level is dynamically adjusted based on the severity of the event, such as in a Level 1 emergency. The value has been increased to 0.6 to prioritize response speed.

[0025] The constraints include: Resource quantity constraint: The total amount of all types of resources used shall not exceed the current available stock. Time constraint: The processing time for a single task shall not exceed the preset threshold for the corresponding event level; Spatial coverage constraint: The service radius of the scheduled resources fully covers the event locations. By solving the above multi-objective optimization problem, the system achieves accurate matching of cross-departmental resources and intelligent assignment of tasks, generating a complete scheduling plan that includes the responsible department, personnel list, resource list, travel path, and processing time limit.

[0026] Step S5: Issuance of commands and dynamic tracking of progress for four-level linkage Based on a four-tiered command structure (city-district-street / township-on-site), a combined tiered dispatch and direct dispatch model is adopted: routine incidents are dispatched level by level from city to district to street / township; major and above-level emergencies support direct dispatch from the city-level command center to on-site personnel, while simultaneously transmitting instructions to the corresponding district and street / township management nodes, ensuring a response delay of ≤3 seconds. Instructions can be simultaneously issued through multiple channels including voice, video, telephone, SMS, and system messages. On-site personnel receive instructions via mobile terminals, reporting the status of handling nodes, on-site audio and video materials, and new resource requests in real time. The system automatically updates the event status and resource location in the digital twin sandbox, leaving a complete record of the entire handling process. Commanders can visually view the handling progress through the sandbox and review the complete handling chain via a timeline.

[0027] Step 6: Handling closed-loop feedback and intelligent review and optimization After an incident is resolved, the system automatically archives all incident data, including report information, dispatch plan, instruction flow records, handling process data, resource allocation records, and on-site documentation. A comprehensive handling efficiency score is calculated based on a multi-dimensional performance evaluation model; the calculation formula is as follows: The system quantifies response effectiveness across four dimensions: response timeliness, handling efficiency, collaborative effectiveness, and loss control. It automatically generates intelligent debriefing reports that include highlights, problem analysis, and optimization suggestions. The debriefing data is simultaneously used to iteratively optimize the parameters and scheduling rules of the resource demand forecasting model, forming a continuous improvement loop of "handling-debriefing-optimization" to continuously enhance the scientific nature and accuracy of command and decision-making.

[0028] This embodiment also provides a multi-level linkage command and decision-making system based on digital twins to implement the above method, specifically including: Digital Twin Sand Table Module: Used to build a three-dimensional real-scene digital twin base for the entire city, complete the digital mapping of all elements of geographic information, emergency resources, infrastructure and protection targets, and support the switching between normal and emergency scenarios and spatial visualization operations.

[0029] Situational awareness module: Used to access and integrate multi-source IoT sensing data, government business data and on-site reported data, complete data cleaning, standardization and spatial mapping, and realize dynamic updates of urban operation status and intelligent early warning of abnormal events.

[0030] Intelligent prediction module: It has a built-in resource demand prediction model based on gradient boosting tree, which is used to output the prediction results of resource type, quantity and deployment location in the corresponding scenario according to real-time event situation and environmental characteristics.

[0031] Intelligent scheduling module: It has a built-in multi-objective optimization scheduling model that combines departmental responsibilities and resource status to complete the accurate matching of cross-departmental resource disposal and intelligent assignment of disposal tasks, generating the global optimal scheduling plan.

[0032] The four-level linkage command module covers four levels of command nodes: city, district, street / township, and on-site. It supports one-click issuance of instructions from multiple channels, integrated communication scheduling, and real-time tracking of the handling progress, and supports cross-level and cross-departmental collaborative command.

[0033] The review and optimization module is used to generate a multi-dimensional performance evaluation report and review suggestions after an incident is resolved, and to iteratively optimize the parameters of the prediction model and the scheduling rules to achieve continuous improvement in handling capabilities.

[0034] The four-level linkage command module includes a collaborative command business system, a unified command and dispatch map, and mobile command and dispatch.

[0035] (1) Collaborative Command System As the "business brain" of the platform, it focuses on cross-level and cross-departmental business collaboration and event lifecycle management. It is adapted to the four-level management architecture of Enshi Prefecture, namely "prefecture-district-street-township-community". It achieves scenario-based function coverage for the three major positioning. The main functions include unattended operation, event management, alarm management, resource management, solution drills, knowledge base, notification management, summary and evaluation, etc.

[0036] (2) Command and dispatch map Based on a high-precision GIS map of the main urban area of ​​Enshi Prefecture, the system integrates multi-dimensional data to achieve "one map for viewing the whole area and one map for managing the overall situation" in three positioning scenarios. It intuitively presents the city's operational status, the progress of emergency response, and the security of major events. The functional modules include information aggregation, map application, analysis and judgment, event response, command and dispatch, and notification issuance.

[0037] (3) Mobile command and dispatch system To address the "last mile" problem of linking "street and township frontline, event site" with the operation center, this system supports mobile terminal use and has the following main functions: First, it allows users to view resource distribution and event dynamics through the Enzheng Tong app and conduct remote command and dispatch; second, it integrates with the Enzheng Tong address book system; and third, it supports online remote command and dispatch of frontline situations (through voice, video, telephone, SMS, and other methods).

[0038] The collaborative command system includes unattended operation, event management, alarm management, resource management, scenario drills, knowledge base, notification management, summary and evaluation, and other functions.

[0039] Unattended Operation: This function provides intelligent duty scheduling management, supports the generation of duty rosters, and ensures that the command center's 24 / 7 duty arrangements are scientific, reasonable, and that responsibilities are assigned to specific individuals. This comprehensively improves the standardization of duty management and the efficiency of human resource allocation. Figure 3 As shown.

[0040] The duty calendar includes a duty panel display (containing duty officer information, scrolling notifications, etc.), attendance status (attended, pending, absent), duty date, and today's duty personnel information (a list of duty personnel divided by administrative region, including name, contact information, and duty position). The interface displays the current duty calendar and today's duty personnel, supporting viewing the daily duty status for each administrative region. It allows for quick communication with today's duty personnel, enabling operations such as calling, SMS, and instant messaging. Figure 4 As shown.

[0041] The duty log manages the duty content of duty personnel. Using preset duty log templates, duty personnel can quickly edit the day's duty information to create a daily duty record. The log also supports exporting duty logs. Figure 5 As shown.

[0042] Template management provides flexible configuration capabilities for duty record templates, supporting custom template fields, format specifications, and filling rules. It allows for the creation of exclusive templates for different duty positions and business scenarios, achieving standardized and personalized unified management of duty information collection, improving recording efficiency and data standardization. Figure 6 As shown.

[0043] Incident reporting primarily utilizes a unified online system for incident push notifications, supplemented by multiple methods (such as telephone, mobile app, and system quick entry) for receiving and entering incident reporting information. Figure 7 As shown.

[0044] 1. Create new events, supporting intelligent creation of events based on the content of telephone calls (structured event data is automatically generated from the call content). 2. Supports integration with third-party event flow systems for event data.

[0045] The event workflow involves reviewing received event information to determine whether to initiate the event-linked handling process. Multi-level event review processes can be configured based on the user's actual event business workflow, enabling online event review and handling. All event handling process data is archived in the system for easy review and summarization. Figure 8 As shown.

[0046] The system adapts to different scenarios, recommending appropriate solutions based on factors such as event type and location. Once a solution is determined, it automatically assigns tasks and notifies relevant personnel. Figure 9 As shown.

[0047] Based on the event type and its development (information such as number of casualties, affected area, and economic losses), the system can recommend and match relevant contingency plans. After matching a plan, it can send time-related notifications (via email, SMS, voice notifications, etc.) according to the configured process nodes. It can automatically assign tasks according to the plan's configuration and quickly organize meetings based on the command structure configured in the plan. Figure 10 As shown.

[0048] Task handling displays event information requiring processing in the event center in a list format. Event handling provides various processing operations such as event acceptance, processing updates, task assignment, multi-person calls, multi-person consultations, application notifications, and event closure. Based on the initial assessment of the event, event tasks can be assigned for simple processing, such as... Figure 11 As shown.

[0049] The alarm list supports displaying alarms from various third-party monitoring devices, asset monitoring, and business alarms. Alarm content includes information such as alarm name, alarm type, time, and alarm location. Figure 12 As shown.

[0050] Alarm handling, through verification and judgment of alarm information, supports actions such as ignoring alarms, transferring them to task handling, transferring them to event handling, and handling them as related events. Figure 13 As shown.

[0051] Resource management, based on the map data of the Enshi Prefecture Urban Operation Management Center's Digital Public Base Urban Information Modeling Platform (CIM), marks various emergency support resources required for urban operation.

[0052] The intelligent scenario simulation exercise automatically links emergency response plans to a knowledge base, generating a multi-dimensional flowchart of the response process. Commanders can manually adjust the order of nodes and compare the simulation results. Figure 14 As shown.

[0053] Multi-department collaborative stress testing was conducted by constructing a virtual command network to simulate scenarios where multiple departments, such as public security, fire protection, and medical services, simultaneously access the system. The test examined command response latency and resource conflicts. Figure 15 As shown.

[0054] The assessment of the plan drills primarily uses charts such as pie charts and bar charts to present statistical analysis of plan data from multiple dimensions, including comprehensive plans, specific plans, on-site response plans, plan levels, and plan classifications. It supports exporting statistical chart images, such as... Figure 16 As shown.

[0055] The system provides flashing alerts for emergency incidents. For alarm information received by the command center, the system supports configuring various parameters such as alarm level, type, threshold, time, and frequency. Highlighted locations on the command and dispatch map flash, notifying relevant personnel to receive the alarm notification and take timely action. Figure 17 As shown.

[0056] Based on on-site investigations of the accident's impact and the extent of damage, and in conjunction with relevant assessment indicators, an overall assessment is conducted, supporting the generation of a summary assessment report, which can be downloaded and printed. Scoring is assigned according to a scoring rule, and a summary assessment report can be output, such as... Figure 18 As shown.

[0057] The "Command and Dispatch Unified Map" is the core visualization hub of the coordinated command platform. Based on the Digital Enshi Public Infrastructure Platform, it integrates multi-dimensional data resources across the entire region, mapping the core resource elements of Enshi Prefecture onto a unified base map. This constructs a command mode that is "globally visible, comprehensively overviewed, precisely dispatched, and closed-loop controlled." It enables the querying, classification, display, statistics, overlay, and emergency command of basic emergency information such as emergencies, protected targets, rescue teams, emergency supplies and equipment, shelters, and experts on a single map. This provides intuitive and efficient decision support for cross-departmental and cross-level coordinated response, completely breaking down information silos and dispatch barriers.

[0058] Information aggregation and command and dispatch on a single map: The information aggregation function serves as the "situational hub" of the mobile command platform. By integrating fragmented data from across the entire domain, it presents core command elements centrally on the mobile map interface through "comprehensive information aggregation + data visualization," providing commanders with efficient support for "knowing the entire domain from a single map and using data to aid decision-making." This achieves a transformation from "dispersed information" to "data fusion," such as... Figure 19 As shown.

[0059] The comprehensive information aggregation module focuses on "scenario-based and practical" approaches. It structures and aggregates key information that is of high concern during command and dispatch, avoiding the need for commanders to switch between multiple modules to query data and making core data "within reach". It mainly covers five sub-modules: today's duty, event statistics, event broadcasting, resource support, and handling plans.

[0060] Obtain duty calendar data to display the names, phone numbers, and WeChat contact information of the duty officer and staff for the day. Support one-click voice dialing for quick contact and improve emergency response efficiency.

[0061] Event statistics are categorized by event type, showing the cumulative number of events this week, this month, and this year. The efficiency trends of handling events are analyzed through charts and visualizations, providing data support for decision optimization.

[0062] Event broadcasts are integrated with the event management module in real time, displaying the latest event updates in a scrolling manner. Users can click to view details and locate the event location on a map for quick understanding of the situation.

[0063] Resource support, linked to the resource support module, classifies and statistically analyzes resource data such as materials, personnel, and equipment, supports viewing detailed information and spatial distribution heat maps / distribution maps, and optimizes resource scheduling.

[0064] The handling plan is integrated with the plan management module, which classifies and statistically analyzes handling plans according to event level and type, supports quick retrieval of suitable plans, and improves the standardization of handling.

[0065] Data visualization, based on mobile maps, uses visualization methods such as icons and heat maps to transform data from the comprehensive information summary module into intuitive graphics, reducing the cost of data interpretation for commanders and enabling "graphical data viewing and situation assessment".

[0066] Map applications are the foundational support for a comprehensive overview. Through flexible layer management and convenient map operations, they achieve deep integration of geospatial data and operational data, providing commanders with a customized and highly efficient map experience. This ensures adequate geographic information support for scenarios such as situational awareness, regional control, and precise dispatching. Figure 20 As shown.

[0067] The analysis and judgment function serves as the "intelligent brain" for coordinated command and decision-making. Focusing on the entire process of "data integration - information mining - risk prediction - correlation analysis," it transforms fragmented data into valuable decision-making evidence through four core capabilities: basic event information query, resource matching information verification, risk assessment and early warning, and event correlation mining. This helps commanders quickly clarify the event context, assess risk levels, match resources for response, and identify potential correlations, providing support for precise dispatch and scientific handling. It is particularly suitable for decision-making needs in complex events, concurrent events, and major emergency scenarios, such as... Figure 21 As shown.

[0068] Command and dispatch functions are the "core of execution" in coordinated command. Focusing on the goals of "full-domain situational awareness, direct multi-channel instruction delivery, and efficient cross-departmental collaboration," it integrates event detail display, event broadcasting and early warning, and multi-dimensional dispatch capabilities via voice / video / telephone / SMS. This constructs a closed-loop command system encompassing "situational awareness - instruction issuance - execution feedback," adaptable to various scenarios such as large-scale event support, emergency response, and daily collaborative management. It ensures that command instructions are accurately, quickly, and comprehensively transmitted to all relevant stakeholders. Figure 22 As shown.

[0069] Event details, serving as a prerequisite for command and dispatch, this function structurally integrates and maps the core information of large-scale events (Enshi Prefecture Anniversary, Marathon, Selenium Expo, etc.) onto a map, providing commanders with a basis for "grasping the overall situation of the event from one map" and ensuring that dispatch decisions are aligned with the actual event scenario.

[0070] The mobile command and dispatch system utilizes mobile devices such as smartphones, professional law enforcement terminals, and tablets. Leveraging technologies like 4G / 5G and BeiDou / GPS positioning, it focuses on four core modules: resource query, event dynamics, address book, and command and dispatch. This constructs a mobile work system that is "operable anytime, anywhere, with comprehensive information control, and closed-loop command flow," breaking the spatial limitations of fixed command centers. It allows commanders and frontline personnel to complete the entire process of "situational monitoring, command transmission and reception, resource dispatch, and on-site feedback" in mobile scenarios, becoming a "mobile hub" for cross-departmental collaborative operations and providing efficient support for emergency response, daily management, and cross-departmental collaboration.

[0071] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if these modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations.

Claims

1. A multi-level coordinated command and decision-making method based on digital twins, characterized in that: Includes the following steps: S1. Construct a three-dimensional digital twin sand table for the entire city: Based on the city's high-precision geographic information, topography, building facilities and administrative division data, build a three-dimensional real-scene digital twin base covering the three-level administrative units of city-district-street and township, and map the spatial attributes and operational status of emergency resources, protection targets and infrastructure throughout the entire area; S2. Dynamic situational awareness through multi-source data fusion: Data from IoT sensing devices, government business systems, and manual reporting are accessed, cleaned, standardized, and fused with spatial coordinates, and then mapped to the digital twin sand table in real time to achieve dynamic updates of all elements of the city's daily operations, emergencies, and major events. S3. Resource scheduling demand prediction based on machine learning: Extract historical event datasets, real-time situation parameters and environmental feature factors, input them into the pre-trained resource demand prediction model, and output the prediction results of the type, quantity and optimal deployment location of resources required in the current scenario. S4. Multi-objective optimization of cross-departmental resource matching and intelligent task assignment: Using resource demand prediction results as input, a multi-objective optimization scheduling function is constructed. Combined with the list of responsibilities of each linked department and the real-time availability of resources, the function completes the accurate matching of cross-departmental resources and the intelligent assignment of tasks, generating the global optimal scheduling scheme. S5, Four-level linkage command issuance and dynamic progress tracking: Relying on the four-level linkage command architecture of city-district-street-township-site, dispatch instructions can be issued to the corresponding disposal entities with one click. The disposal progress, resource location and on-site feedback information are synchronized in real time through the digital twin sand table, realizing the visualized control of the entire disposal process; S6. Closed-loop feedback and intelligent review optimization: After the event is completed, the entire process data is automatically archived, a review report is generated based on a multi-dimensional performance evaluation model, and the resource demand prediction model and scheduling rules are iteratively optimized to form a closed-loop management of the entire process.

2. The multi-level linkage command and decision-making method based on digital twins according to claim 1, characterized in that: The construction process of the multi-objective optimization scheduling function in step S4 is as follows: taking the minimization of scheduling response time, the maximization of resource utilization, and the minimization of total scheduling cost as optimization objectives, a weighted multi-objective optimization objective function F(x) is constructed, and the calculation formula is as follows: ； In the formula The overall response time corresponding to scheduling scheme x is the sum of the time taken for all assigned task personnel / resources to arrive at the event site. The comprehensive utilization rate of resources is taken as the weighted average of the load rates of various resources participating in the scheduling. The total dispatch cost includes labor costs, material loss costs, and transportation costs. These are the weighting coefficients corresponding to response time, resource utilization, and scheduling cost, respectively, and satisfy the following conditions: The weighting coefficients are dynamically adjusted based on the event level and scenario type. The constraints of the multi-objective optimization function include: resource quantity constraints, timeliness constraints, and spatial coverage constraints.

3. The multi-level linkage command and decision-making method based on digital twins according to claim 1, characterized in that: The resource demand prediction model described in step S3 uses a gradient boosting tree ensemble learning model, and the calculation formula for its prediction output is as follows: ； In the formula The output for resource demand forecasting includes prediction results across three dimensions: resource type, resource quantity, and deployment location coordinates; K represents the total number of regression decision trees in the ensemble model. Let be the output function corresponding to the k-th regression decision tree; is the function space formed by all regression decision trees; the input feature vector x of the model contains event type, event risk level, population density of the event area, surrounding resource stock, meteorological environmental parameters, terrain conditions and historical resource consumption statistics of similar events.

4. The multi-level linkage command and decision-making method based on digital twins according to claim 1, characterized in that: In step S6, the multi-dimensional effectiveness evaluation model adopts a weighted comprehensive scoring method. The formula for calculating the comprehensive score S of the disposal effectiveness is as follows: ； In the formula, R is the response timeliness score, calculated by normalizing the ratio of the time from the issuance of the instruction to the first response on site to the standard time; E is the handling efficiency score, calculated by normalizing the ratio of the actual total handling time of the event to the standard handling time of the same level event; C is the cross-departmental collaboration effect score, calculated by weighting the cross-departmental task completion rate and information synchronization timeliness rate; and D is the loss control effect score, calculated by normalizing the ratio of the actual loss caused by the event to the estimated loss. These are the weight coefficients for the four evaluation dimensions, and they satisfy... A higher overall score indicates better event handling efficiency.

5. The multi-level linkage command and decision-making method based on digital twins according to claim 1, characterized in that: The digital twin sand table configuration described in step S1 has a seamless switching mechanism between normal and emergency situations: in the normal operation state, it displays the operational status of all elements of urban management; when an emergency is triggered or a major event is launched, it can switch to the corresponding special command mode with one click, prioritizing the display of event locations, surrounding emergency resources and disposal links, and hiding irrelevant business layers to adapt to the command focus needs of different scenarios.

6. The multi-level linkage command and decision-making method based on digital twins according to claim 1, characterized in that: The multi-source data mentioned in step S2 specifically includes: IoT sensing data collected by video surveillance, environmental monitoring sensors, geological disaster monitoring equipment, and traffic flow collection equipment; business system data from public security, transportation, health, emergency management, and urban management departments; and on-site text, image, and audio / video data reported by mobile terminals. All data are standardized and then uniformly converted into the CGCS2000 national geodetic coordinate system to complete the spatial mapping with the digital twin sand table.

7. The multi-level linkage command and decision-making method based on digital twins according to claim 1, characterized in that: The four-level linkage command architecture described in step S5 adopts an instruction flow mode that combines hierarchical dispatching and direct dispatching to higher levels: routine events are dispatched and handled level by level from city to district to street / township; major and above-level emergency events support the city-level command center to directly dispatch orders to on-site personnel, while simultaneously synchronizing instruction information to the corresponding district and street / township management nodes to ensure that the instruction response delay for emergencies is ≤3 seconds.

8. The multi-level linkage command and decision-making method based on digital twins according to claim 1, characterized in that: Step S2, dynamic situational awareness, also includes intelligent early warning function: it automatically identifies abnormal events such as street vendors, pedestrian congestion, fire, and potential geological disasters through video AI recognition algorithms, automatically generates early warning work orders and pushes them to the command terminal, and marks the early warning location and risk level in the digital twin sand table with differentiated colors and flashing effects.

9. A multi-level linkage command and decision-making method based on digital twins according to claim 1, characterized in that: In step S5, the dynamic tracking of the handling progress is as follows: On-site personnel report the status of handling nodes, on-site audio and video materials, and new resource requirements in real time through mobile terminals. The system automatically updates the event status and resource location in the digital twin sandbox. The entire handling process is recorded and can be traced back to view the complete handling chain by timeline.

10. A multi-level linkage command and decision-making system based on digital twins, characterized in that: include: The digital twin sand table module is used to build a three-dimensional real-scene digital twin base for the entire city, and to complete the digital mapping of all elements of geographic information, emergency resources, infrastructure and protection targets. The situation awareness module is used to access and integrate multi-source IoT sensing data, government business data and on-site reported data to achieve dynamic updates of the city's operational status and intelligent early warning of abnormal events. The intelligent prediction module has a built-in resource demand prediction model based on gradient boosting tree, which is used to output the prediction results of resource type, quantity and deployment location in the corresponding scenario according to the real-time situation. The intelligent scheduling module has a built-in multi-objective optimization scheduling model, which is used to accurately match cross-departmental resources and intelligently assign tasks, generating the globally optimal scheduling scheme. The four-level linkage command module covers four levels of command nodes: city, district, street / township, and on-site. It supports one-click issuance of instructions from multiple channels, integrated communication scheduling, and real-time tracking of the progress of the handling. The review and optimization module is used to generate a multi-dimensional performance evaluation report after an event is resolved, and to iteratively optimize the parameters of the prediction model and the scheduling rules to achieve continuous improvement in handling capabilities. The digital twin sand table module, situational awareness module, intelligent prediction module, intelligent scheduling module, and debriefing and optimization module are all connected to the four-level linkage command module; The four-level linkage command module includes a collaborative command business system, a unified command and dispatch map, and mobile command and dispatch.

Images