Urban public infrastructure intelligent performance governance method and system based on digital twinborn technology

Abstract

The invention discloses an urban public infrastructure intelligent performance governance method and system based on a digital twinborn technology. According to the method, firstly, sensor monitoring, expert evaluation and questionnaire data are fused, and performance evaluation, evolution deduction and operation and maintenance decision models are constructed; secondly, updating the model by utilizing machine learning and driving the digital twin, deducing performance degradation in a virtual environment, and optimizing and generating a maintenance strategy; and finally, writing the performance index output by the model and an evaluation result into a block chain, and automatically executing performance payment through an intelligent contract. The system correspondingly comprises a multivariate data fusion module, a model engine module, a digital twin decision-making module and a block chain contract module. According to the method, a treatment closed loop of data driving, model prediction, virtual decision making and automatic execution is formed, the problems of one-sided evaluation, uneconomical decision making and low execution efficiency of a traditional method are solved, and the accuracy, economical efficiency and automation level of infrastructure operation and maintenance are improved.

Description

Technical Field

[0001] This invention relates to the field of smart city and infrastructure operation and maintenance management technology, specifically to a smart performance governance method and system for urban public infrastructure based on digital twin, machine learning and blockchain technologies. Background Technology

[0002] Urban public infrastructure (such as bridges, tunnels, integrated utility tunnels, and water treatment facilities) is the lifeline for ensuring the operation of cities. With the acceleration of urbanization, the scale of infrastructure is growing rapidly, and its operation and maintenance management is transforming from the traditional "construction-heavy, maintenance-light" model to a "construction and management in tandem, refined governance throughout the entire life cycle" model. In this process, how to scientifically assess facility performance, predict its performance evolution, optimize operation and maintenance decisions, and achieve performance-based contract management has become a common challenge facing the industry.

[0003] Currently, common operation and maintenance performance management systems mainly rely on real-time monitoring using installed sensors and simple algorithms for evaluation and early warning. However, these methods have significant shortcomings:

[0004] The evaluation system is static and one-sided: existing evaluations rely on a single or a few static indicators and fail to build a comprehensive, dynamic and multi-attribute performance evaluation indicator system, making it difficult to accurately reflect the complex real state of facilities and future evolution trends.

[0005] The decision-making process lacks foresight and economy: Operation and maintenance decisions (such as maintenance timing and maintenance methods) are often based on experience or fixed cycles, lacking scientific deduction based on facility performance degradation models. This can easily lead to "over-maintenance" causing resource waste, or "under-maintenance" causing operational risks, making it impossible to achieve the optimal balance between total life cycle cost and facility reliability.

[0006] Data, models, and decision-making are disconnected: an effective closed loop has not yet been formed between monitoring data, evaluating models, simulations, and final decisions. The decision-making process rarely utilizes digital twin technology for "virtual-real" simulation and optimization, resulting in insufficient scientific rigor and reliability of decisions.

[0007] Performance management is disconnected from business processes: performance evaluation results are weakly linked to business execution processes such as contract payments, rewards and penalties, and rely on manual review and offline operations, which is inefficient and prone to disputes, failing to achieve automation and intelligence in "pay-for-performance".

[0008] Therefore, there is an urgent need for an innovative method and system that can integrate advanced information technology to achieve a closed-loop governance of the entire chain of "precise assessment - intelligent simulation - optimized decision-making - automatic execution" of urban public infrastructure performance. Summary of the Invention

[0009] The technical problem to be solved by this invention is to overcome the shortcomings of the existing technology, establish a dynamic, multi-dimensional, and scientific performance evaluation and evolution model, realize accurate profiling of facility status and reliable prediction of future trends, and provide a smart performance governance method and system for urban public infrastructure based on digital twin technology to solve the above problems.

[0010] The object of this invention is achieved in the following manner:

[0011] A smart performance governance method for urban public infrastructure based on digital twin technology includes the following steps: S1, Performance data collection and modeling stage: Real-time monitoring and acquisition of multi-source performance data of urban public infrastructure; construction of a performance evaluation and evolutionary deduction model based on the multi-source performance data, including a performance evaluation index system, a performance status evaluation model, a performance status evolution deduction model, and an operation and maintenance decision model, and generation of an initial performance database; S2, Dynamic update stage of digital twin model: Continuous training of the performance database using machine learning algorithms to update the parameters in the performance status evolution deduction model; based on the updated model parameters, synchronously updating the digital twin model of the urban public infrastructure to reflect the latest performance degradation trend of the physical facilities; S3, Intelligent operation and maintenance decision generation stage: Based on the updated digital twin model, running the performance status evolution deduction model to simulate the future evolution path of facility performance under different maintenance strategies; with the goal of minimizing total operation and maintenance costs, the operation and maintenance decision model uses a multi-objective optimization algorithm to jointly optimize maintenance time, maintenance method, and maintenance intensity to generate a recommended optimal operation and maintenance decision scheme; S4. Intelligent Execution Stage of Performance Contract: The key indicators of the optimal operation and maintenance decision scheme generated in step S3 and bound to the performance status, as well as the real-time performance evaluation results, are stored in the blockchain network; the smart contract deployed on the blockchain listens to the key indicators and evaluation results, and automatically executes the corresponding fee payment or quality reward and punishment clauses when the data meets the predefined performance contract triggering conditions.

[0012] In step S1, the specific process of constructing the performance evaluation indicator system includes: initially screening performance influencing factors through literature review and expert interviews; supplementing, eliminating and refining the factors using questionnaire surveys and factor analysis methods, and constructing a multi-level scenario element system that includes facility reliability, operating efficiency, maintenance costs and environmental impact.

[0013] The performance status assessment model is a multi-attribute fuzzy comprehensive evaluation model based on expert weights and factor weights. Its construction process includes: constructing an initial evaluation matrix based on expert scoring data; using the distance measure method based on ideal solutions to calculate the evaluation consistency weight of each expert and the discrimination weight of each scenario influencing factor; using the weights to weight and aggregate the initial matrix to obtain a comprehensive performance evaluation value; and calculating the membership degree of the comprehensive performance evaluation value to each performance level according to the preset performance level threshold using a triangular or trapezoidal membership function to complete the status assessment.

[0014] The performance state evolution inference model is a probabilistic graphical model based on dynamic Bayesian networks. Its construction process includes: abstracting the key elements in the multi-level scenario element system into network nodes; defining directed edges between nodes based on the causal relationship between elements to construct a cross-time slice network structure; and determining the conditional probability table of nodes based on historical operation and maintenance data or expert experience to describe the stochastic process of facility performance state evolution over time under no intervention or specific maintenance intervention.

[0015] The operation and maintenance decision model includes an equal-cycle model and a sequential-cycle model. The equal-cycle model uses a fixed maintenance interval as the decision variable, while the sequential-cycle model uses a sequence of non-equal-length maintenance intervals as the decision variable. The objective function F for both models is expressed as: F = Total Maintenance Cost (C_m) + Expected Risk Loss (C_r). The total maintenance cost C_m is related to the facility's construction quality, the number of maintenance operations, and the unit maintenance cost. The expected risk loss C_r is related to the facility's performance status and failure rate function within the current maintenance cycle, with the failure rate function increasing stepwise with the number of maintenance operations.

[0016] The multi-source performance data includes: operational status monitoring data collected in real time through a sensor network deployed in physical facilities; subjective evaluation data based on domain knowledge and experience obtained through an expert evaluation system; and satisfaction or evaluation data from operators, maintainers, or users obtained through a questionnaire survey system.

[0017] A smart performance governance system for urban public infrastructure based on digital twin technology, used to implement the method, includes: a construction service process performance monitoring module, used to access sensor data through an IoT interface and call the performance status evaluation model to calculate the real-time performance status; a full life cycle performance evolution simulation module, connected to the performance monitoring module, used to receive the real-time performance status and call the performance status evolution prediction model to perform multi-scenario simulation and prediction of the future performance of the facility in digital space, and output evolution trend data; a digital twin core module, bidirectionally connected to the performance monitoring module and the performance evolution simulation module, including: a twin model management unit, used to maintain the three-dimensional digital model and behavioral model corresponding to the physical facility; and a data-model driven unit, used to drive the real-time data of the performance monitoring module to the... The system includes: a digital model, which receives simulation data from the performance evolution pre-simulation module for verification and optimization; a decision generation unit, which runs the operation and maintenance decision model in the virtual environment provided by the digital model to generate operation and maintenance decision schemes; and a blockchain smart contract module, connected to the decision generation unit of the digital twin core module, which receives key performance commitment indicators from the operation and maintenance decision schemes. This module includes: a contract logic encapsulation unit, which encodes performance payment terms into automatically executable smart contract code; a performance data on-chain unit, which compares the consensus-verified actual performance data from the digital twin core module with the key performance commitment indicators and writes the comparison results to the blockchain; and an automatic execution and settlement unit, which automatically triggers payment transactions on the blockchain when the smart contract determines that performance targets have been met.

[0018] The digital twin core module is further connected to a performance status update module. The performance status update module is used to collect historical performance data and decision feedback results, and to use machine learning algorithms to train and fine-tune the parameters of the performance status assessment model, performance status evolution inference model and operation and maintenance decision model offline and online. The optimized model parameters are then pushed to the digital twin core module to form a closed-loop adaptive governance loop of "monitoring-assessment-decision-feedback-optimization".

[0019] In the blockchain smart contract module, the data written to the blockchain by the performance data on-chain unit includes at least: a timestamped comprehensive performance score output by the performance status evaluation model, and key performance indicator values ​​generated by the operation and maintenance decision model as the contract subject; the logic of the smart contract code is as follows: when the comprehensive performance score is consistently higher than the key performance indicator value within the agreed period, full payment is automatically executed; if it is lower than the indicator value, deduction or claim is automatically calculated and executed according to preset rules.

[0020] The beneficial effects of this invention are as follows: By constructing a dynamic evaluation model that integrates multi-attribute decision-making and fuzzy evaluation, this invention overcomes the one-sidedness of traditional single-indicator evaluation and achieves accurate quantification and level determination of the complex performance status of infrastructure. Based on an evolutionary inference model using dynamic Bayesian networks, it enables probabilistic prediction of future facility performance. Multi-objective optimization of operation and maintenance strategies within a digital twin environment transforms decision-making from "experience-driven" to "model and data-driven," effectively avoiding over-maintenance or under-maintenance and achieving optimal lifecycle costs. A complete closed loop of "physical perception - digital modeling - simulation inference - intelligent decision-making - entity execution - data feedback" is constructed. The digital twin, as a bridge between the virtual and real worlds, makes the entire governance process visible, verifiable, and optimizable. Attached Figure Description

[0021] Figure 1 This is the membership function for evaluating the performance status of urban public infrastructure in this embodiment of the invention.

[0022] Figure 2 This is a schematic diagram of a performance evolution scenario state network model based on dynamic Bayes.

[0023] Figure 3 This is a schematic diagram of the network structure of the performance state evolution inference model based on dynamic Bayesian network at a certain moment in an embodiment of the present invention.

[0024] Figure 4 This is a dynamic Bayesian network diagram of the performance evolution scenario in this embodiment of the invention.

[0025] Figure 5 This is a schematic diagram of the operation and maintenance decision-making of the service provider under the construction service mode in this embodiment of the invention.

[0026] Figure 6 This is an architectural block diagram of a smart performance governance system for urban public infrastructure based on digital twin technology, provided in an embodiment of the present invention. Detailed Implementation

[0027] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0028] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same technical meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0029] A smart performance governance method for urban public infrastructure based on digital twin technology includes the following steps: S1, Performance data collection and modeling stage: Real-time monitoring and acquisition of multi-source performance data of urban public infrastructure; construction of a performance evaluation and evolutionary deduction model based on the multi-source performance data, including a performance evaluation index system, a performance status evaluation model, a performance status evolution deduction model, and an operation and maintenance decision model, and generation of an initial performance database; S2, Dynamic update stage of digital twin model: Continuous training of the performance database using machine learning algorithms (extended Noisy-or Gate dynamic Bayesian model) to update the parameters in the performance status evolution deduction model; based on the updated model parameters, synchronously updating the digital twin model of the urban public infrastructure so that the digital model reflects the latest performance degradation trend of the physical facilities; S3. Intelligent Operation and Maintenance Decision Generation Stage: Based on the updated digital twin model, the performance state evolution simulation model is run to simulate the future evolution path of facility performance under different maintenance strategies. With the goal of minimizing total operation and maintenance costs, the operation and maintenance decision model uses a multi-objective optimization algorithm to jointly optimize maintenance time, maintenance method, and maintenance intensity to generate a recommended optimal operation and maintenance decision scheme. S4. Intelligent Execution Stage of Performance Contracts: The key indicators of the optimal operation and maintenance decision scheme generated in step S3 and bound to the performance state, as well as the real-time performance evaluation results, are stored in the blockchain network. The smart contract deployed on the blockchain monitors the key indicators and evaluation results. When the data meets the predefined performance contract triggering conditions, the corresponding fee payment or quality reward / penalty clauses are automatically executed.

[0030] In step S1, the specific process of constructing the performance evaluation indicator system includes: initially screening performance influencing factors through literature review and expert interviews; supplementing, eliminating, and refining the factors using questionnaire surveys and factor analysis methods to construct a multi-level scenario element system that includes facility reliability, operational efficiency, maintenance costs, and environmental impact. Specifically: First, by searching relevant literature, industry standards, and performance appraisal management methods, research results from scholars on performance management and performance influencing factors of urban public infrastructure projects are collected and organized to initially screen the influencing factors of urban public infrastructure performance evolution scenarios; second, the influencing factors of the initially screened urban public infrastructure performance evolution scenarios are supplemented, eliminated, and refined using expert interviews, questionnaire surveys, and other methods, and an influencing factor screening model is constructed to finally determine the performance evolution scenario elements.

[0031] The performance status assessment model is a multi-attribute fuzzy comprehensive evaluation model based on expert weights and factor weights. Its construction process includes: constructing an initial evaluation matrix based on expert scoring data; using the distance measure method based on ideal solutions to calculate the evaluation consistency weight of each expert and the discrimination weight of each scenario influencing factor; using the weights to weight and aggregate the initial matrix to obtain a comprehensive performance evaluation value; and calculating the membership degree of the comprehensive performance evaluation value to each performance level according to the preset performance level threshold using a triangular or trapezoidal membership function to complete the status assessment.

[0032] The performance state evolution inference model is a probabilistic graphical model based on dynamic Bayesian networks. Its construction process includes: abstracting the key elements in the multi-level scenario element system into network nodes; defining directed edges between nodes based on the causal relationship between elements to construct a cross-time slice network structure; and determining the conditional probability table of nodes based on historical operation and maintenance data or expert experience to describe the stochastic process of facility performance state evolution over time under no intervention or specific maintenance intervention.

[0033] The operation and maintenance decision model includes an equal-cycle model and a sequential-cycle model. The equal-cycle model uses a fixed maintenance interval as the decision variable, while the sequential-cycle model uses a sequence of non-equal-length maintenance intervals as the decision variable. The objective function F for both models is expressed as: F = Total Maintenance Cost (C_m) + Expected Risk Loss (C_r). The total maintenance cost C_m is related to the facility's construction quality, the number of maintenance operations, and the unit maintenance cost. The expected risk loss C_r is related to the facility's performance status and failure rate function within the current maintenance cycle, with the failure rate function increasing stepwise with the number of maintenance operations.

[0034] The multi-source performance data includes: operational status monitoring data collected in real time through a sensor network deployed in physical facilities; subjective evaluation data based on domain knowledge and experience obtained through an expert evaluation system; and satisfaction or evaluation data from operators, maintainers, or users obtained through a questionnaire survey system.

[0035] The method for intelligent governance of urban public infrastructure performance based on operation and maintenance decision-making further includes: storing all performance data information in a database, constructing a smart contract platform based on blockchain technology, and combining the performance data of the digital twin performance governance platform. If the performance data (performance score) is >90 points, payment and contract management will be carried out through smart contracts (contract information entry). If the performance data is below 90 points, the payment ratio will be calculated based on the actual data, and corresponding maintenance requirements and the next acceptance node will be given. If the next acceptance is qualified, the difference will continue to be paid, thereby realizing performance payment and contract management.

[0036] A smart performance governance system for urban public infrastructure based on digital twin technology includes: a construction service process performance monitoring module, used to access sensor data via an IoT interface and call the performance status evaluation model to calculate the real-time performance status; a full life cycle performance evolution simulation module, connected to the performance monitoring module, used to receive the real-time performance status and call the performance status evolution prediction model to perform multi-scenario simulation and prediction of the future performance of the facility in digital space, and output evolution trend data; and a digital twin core module, bidirectionally connected to both the performance monitoring module and the performance evolution simulation module, including: a twin model management unit, used to maintain the three-dimensional digital model and behavioral model corresponding to the physical facility; and a data-model driving unit, used to drive the digital model with the real-time data from the performance monitoring module. Simultaneously, simulation data from the performance evolution pre-simulation module is injected into the digital model for verification and optimization; a decision generation unit is used to run the operation and maintenance decision model in the virtual environment provided by the digital model to generate an operation and maintenance decision plan; a blockchain smart contract module is connected to the decision generation unit of the digital twin core module and is used to receive key performance commitment indicators in the operation and maintenance decision plan; this module includes: a contract logic encapsulation unit, used to encode performance payment terms into automatically executable smart contract code; a performance data on-chain unit, used to compare the consensus-verified actual performance data from the digital twin core module with the key performance commitment indicators and write the comparison results into the blockchain; and an automatic execution and settlement unit, used to automatically trigger payment transactions on the blockchain when the smart contract determines that the performance has met the standards.

[0037] The digital twin core module is further connected to a performance status update module. The performance status update module is used to collect historical performance data and decision feedback results, and to use machine learning algorithms to train and fine-tune the parameters of the performance status assessment model, performance status evolution inference model and operation and maintenance decision model offline and online. The optimized model parameters are then pushed to the digital twin core module to form a closed-loop adaptive governance loop of "monitoring-assessment-decision-feedback-optimization".

[0038] In the blockchain smart contract module, the data written to the blockchain by the performance data on-chain unit includes at least: a comprehensive performance score with a timestamp output by the performance status evaluation model, and key performance indicator values ​​generated by the operation and maintenance decision model as the contract subject.

[0039] The logic of the smart contract code is as follows: when the comprehensive performance score is consistently higher than the key performance indicator value within the agreed period, full payment is automatically executed; if it is lower than the indicator value, deduction or claim is automatically calculated and executed according to preset rules.

[0040] Example:

[0041] The performance evolution scenario element system was determined using methods such as literature review, interviews, and expert evaluation. This system was developed as follows: First, by searching relevant literature, industry standards, and performance appraisal management methods, research findings on performance management and influencing factors of urban public infrastructure projects were collected and organized to initially screen the influencing factors of urban public infrastructure performance evolution scenarios. Second, expert interviews and questionnaires were used to supplement, eliminate, and refine the initially screened influencing factors, and an influencing factor screening model was constructed to ultimately determine the performance evolution scenario elements.

[0042] The performance status evaluation model is characterized by including expert weights, context-influencing factor weights, and a multi-attribute evaluation model. This represents the set of all influencing factors being evaluated. This indicates the gathering of all invited experts. Let represent the set of weights for all invited experts, where ; Indicates different time periods, This represents the weights for different time periods, where ; Indicates the time period The initial evaluation matrix within, Indicates the time period Inner An expert Regarding the first The performance evolution scenario status evaluation values ​​of each influencing factor are determined. By aggregating the status assessment results of all experts, a comprehensive evaluation result of the performance evolution scenario status of urban public infrastructure can be obtained.

[0043] The element weights and expert weights are defined as follows: For the time period Inner The evaluation expert's opinion on the first The positive ideal evaluation value of the scenario state of the performance evolution of each influencing factor. For the time period Inner The evaluation expert's opinion on the first The negative ideal evaluation value of the scenario state of the performance evolution of each influencing factor. For the time period Inner The evaluation expert's opinion on the first Evaluation values ​​of the scenario state of the performance evolution of each influencing factor To the ideal evaluation value Euclidean distance, For the time period Inner The evaluation expert's opinion on the first Evaluation values ​​of the scenario state of the performance evolution of each influencing factor To negative ideal evaluation value The Euclidean distance.

[0044] First, the determination of expert weights is based on the principle of consistency of evaluation information among evaluation experts. That is, for a given facility's performance evolution scenario, if an expert's evaluation information is highly consistent with the evaluation values ​​of other evaluation experts for that facility, then that expert is given a higher weight. Based on this, the... Evaluation experts and others The evaluation expert's opinion on the first The degree of consistency in the evaluation opinions of the performance evolution scenario states of each influencing factor is defined as follows: , , , (1); furthermore, the first The weight of each evaluation expert is: , (2).

[0045] Secondly, in order to determine the element weights, the definitions of the comprehensive performance evolution scenario state level measure and its distance measure are proposed as follows: Definition 1 Let... For those invited The evaluation expert's opinion on the first The comprehensive performance evolution scenario state level measure of each influencing factor is then: (3), where, and These are sets of influencing factors that are either benefit-oriented or cost-oriented. For the first The weight of each evaluation expert, , Obviously, and The larger the value, the more likely it is that the first... The higher the overall performance evolution scenario state level of each influencing factor, the lower the level of the scenario state.

[0046] Definition 2 Let For the first The performance evolution scenario state evaluation values ​​of each influencing factor and other factors The distance measure of the comprehensive level of the performance evolution scenario state evaluation values ​​of each influencing factor is then: , , (4).

[0047] As can be seen from equation (4), and The larger the value, the better the performance evolution scenario of the evaluated influencing factor compared to others. The greater the gap between performance evolution scenarios, the better.

[0048] Based on Definitions 1 and 2, the weights of each influencing factor are: (5).

[0049] The performance status assessment model: In the process of evaluating the performance evolution scenario status of urban public infrastructure, the performance evolution scenario status assessment matrix is ​​obtained based on the project performance assessment score. Based on this, the expert weight and element weight are calculated, and the initial assessment matrix is ​​weighted to obtain the comprehensive performance evolution scenario status assessment result. Finally, the membership degree of the urban public infrastructure performance evolution scenario status is determined by the constructed membership degree function. The main calculation steps are as follows: (1) Construct the performance evolution scenario status assessment matrix.

[0050] Based on the influencing factors of the constructed performance evolution scenario, an initial performance evolution scenario state assessment matrix can be obtained by collecting assessment data from different experts: (6), where, Indicates a certain moment Performance evolution scenario state assessment matrix Indicates time, Indicates the first One expert, Indicates the first One influencing factor, express Time Expert For the Performance evolution scenario state scores for each influencing factor.

[0051] (2) Comprehensive performance evolution scenario state assessment matrix: Combining the constructed expert weight and element weight model, the weighted comprehensive performance evolution scenario state assessment matrix can be obtained: (7).

[0052] (3) Comprehensive performance evolution scenario status assessment results: Based on formula (7), the expert weights and influencing factor weights are solved using formulas (1)-(5). By gathering the evaluation information among different experts, the performance evolution scenario status assessment results of different influencing factors can be obtained. Furthermore, by gathering the evaluation information among different factors, the overall performance evolution scenario status assessment results of the facility can be obtained.

[0053] (4) Membership Function Construction: The core of the fuzzy comprehensive evaluation of the performance evolution scenario state of urban public infrastructure is the construction of the membership function. The construction of the membership function should be based on the corresponding level standards of the performance evolution scenario state. Membership is fuzzy information and can effectively overcome the differences in the level standards of various influencing factors. It can be widely used for comparison of different states at different times. According to the actual project performance appraisal management method, the project performance status level is usually divided into five levels: [0,60), [60,70), [70,80), [80,90), and [90,100]. Service providers charge project performance appraisal fees based on the project performance appraisal level they obtain. Accordingly, the fuzzy comprehensive evaluation using the membership function gives the result of the performance evolution scenario state evaluation. The membership function of the performance evolution scenario state of urban public infrastructure is as follows: Figure 1 As shown.

[0054] according to Figure 1 The membership function of the scenario state in the performance evolution of urban public infrastructure is in the following form: ; ; ; ; (8).

[0055] Therefore, the membership matrix of different levels to which the performance evolution scenarios of urban public infrastructure belong can be obtained, as follows: (9), among which, Indicates at time Inner Membership degree of each influencing factor's performance evolution scenario state. Indicates time; Indicates different influencing factors, , The membership function represents different levels.

[0056] The performance state evolution model states that urban public infrastructure, during its long-term service life, will deteriorate or experience cumulative wear and tear, leading to gradual performance degradation until failure. In this process, facility performance is primarily characterized by performance evolution scenario states, which are dynamically changing. If we let... This represents the state function of the performance evolution scenario. For urban public infrastructure at all times The performance evolution scenario state, if its initial performance evolution scenario state is The time-varying parameters are The invariant parameter is Then the city's public infrastructure is in constant motion The situation state can be represented as: (10)

[0057] In the process of modeling the performance evolution scenarios of urban public infrastructure, the performance state will be... Treating these as random variables, and combining them with a dynamic Bayesian network model, a scenario-state network model for the performance evolution of urban public infrastructure based on an extended Noisy-or Gate is constructed. A discrete-time Markov process is established using conditional probability tables of the random variables to represent the scenario-state of urban public infrastructure performance evolution without performance governance measures. In time The evolutionary process within.

[0058] Furthermore, assuming the performance evolution scenario of urban public infrastructure is in time... The corresponding state sequence is generated internally. The probability distribution of the state sequence is given by a vector. Characterization, in which and Based on the memoryless property of Markov processes, we can conclude that: (11).

[0059] For ease of calculation, the time of a certain facility is... Performance evolution scenario In other words, assuming it starts from time... At the time The state probabilities are unaffected by time. Considering the factors influencing the performance evolution scenarios of urban public infrastructure, time-varying and invariant parameters are introduced to construct a performance evolution scenario state network model based on dynamic Bayesianism, such as... Figure 2 As shown.

[0060] According to the relevant principles of dynamic Bayes, the main steps in constructing a performance evolution scenario state network model based on dynamic Bayes are: (1) Determine the node variables on the network. The first step in constructing a dynamic Bayes network model is to determine the network node variables. The essence of constructing a dynamic Bayes network for performance evolution scenario states is to theoretically model urban public infrastructure in the real world. Through theoretical modeling, the scenario elements of urban public infrastructure performance evolution are abstracted and expressed, and these abstracted key scenario elements become the node variables that make up the network.

[0061] (2) Determine the relationships between node variables. Determining the relationships between node variables within a dynamic Bayesian network involves clarifying the causal relationships of key elements and drawing them out using directed edges to form a complete scenario evolution process. Based on the division of key elements in the scenario, the three levels of key influencing elements in the urban public infrastructure performance evolution scenario are as follows: Level 3 Second level First level In this model, the final performance evolution scenario of the facility is influenced by factors at each level. Therefore, this invention will primarily analyze the impact of changes in third-level influencing factors on the scenario states of second-level and first-level influencing factors. The Bayesian network describing the relationships between network node variables at a given moment is as follows: Figure 3 As shown.

[0062] Will Figure 3 A Bayesian network at a certain moment is embedded in a continuous and dynamic situational state evolution process. If the entire performance evolution situational process is divided into... Each moment is denoted as _____. ,in If this is the initial moment of the entire performance evolution scenario process, then the entire performance evolution scenario process can be used... Figure 4 express.

[0063] for Performance evolution scenario at the initial moment In terms of input node variables and Under its influence, it enters the performance evolution scenario state of the next moment. The initial state of the situation and In the moment and It became The input node variables are then used, and so on. It can be seen that, from the perspective of input variables, the entire performance evolution scenario dynamic Bayesian network can be divided into two parts: one part is the initial performance evolution scenario network, where the performance evolution scenario state is... One part consists of only two types of input variables at the initial time point; the other part is the developmental scenario network, which evolves from performance-based scenario states. Initially, there will be three types of input node variables, which are the second-level influencing factors at that moment. First-level influencing factors and the performance evolution scenario state at the previous moment. ,in .

[0064] (3) Determination of prior probability and conditional probability. In the model constructed in this invention, the node variables that need to be assigned probabilities are divided into two categories: First, for node variables without parent nodes, their prior probabilities need to be determined based on professional knowledge or experience; Second, for node variables with parent nodes, their conditional probabilities need to be clarified. Using the extended Noisy-orGate model constructed in claim 1, combined with historical experience or historical statistical data, the conditional probability table of the dynamic Bayesian scenario network model of the performance evolution of the entire city's public infrastructure can be obtained by solving the problem.

[0065] The operation and maintenance decision model includes an equal-period operation and maintenance decision model and a sequential-period model. The equal-period operation and maintenance decision model is as follows:

[0066] Taking into account factors such as facility construction quality, construction cost, service provider's maintenance intervals, maintenance frequency, maintenance behavior, and risk losses due to natural deterioration, this study constructs an objective function to minimize the service provider's total cost during the facility construction and operation and maintenance phases. It investigates facility operation and maintenance decisions based on performance evolution from both cyclical and sequential perspectives. Specifically, the objective function consists of two parts: the first part represents the costs incurred by the service provider during the construction phase and the maintenance costs incurred during the operation and maintenance phase; the second part represents the risk losses related to the service provider's maintenance methods and their effectiveness during the operation and maintenance phase. It is assumed that the service provider's preventative maintenance interval for the facility is... ,like This is the case of periodic operation and maintenance decision-making, such as... Figure 5 As shown in (a); if This is the case of sequential cycle operation and maintenance decision-making, such as... Figure 5 As shown in (b).

[0067] Since the total cost of facility operation and maintenance includes both repair costs and risk loss costs, and these two are negatively correlated, reducing facility risk loss requires increasing repair costs, and vice versa. Therefore, to achieve a balance between the two, this section establishes an optimal maintenance strategy model with the objective function of minimizing the sum of the total cost of facility maintenance and the facility's risk loss: (12), where, It is the objective function. and This refers to the total cost and risk of service providers repairing facilities during the construction and operation of the project.

[0068] (1) Determining the failure rate: It is advisable to set the time point for service providers to conduct preventive maintenance on the facilities as follows: That is, in Preventive maintenance should be performed regularly, during which replacements or repairs can be carried out. If it is replaced frequently, the facility will be in the [location / stage]. Repair cycle Internal failure rate function It can be represented as (13), where, Adding a factor to the failure rate indicates that after the first failure... The rate of increase in facility failure rate after preventive maintenance. The service age regression factor indicates the number of years of service after which service age has decreased. The rate of reduction in the effective service life of a facility after a preventive maintenance.

[0069] According to the recursive algorithm, the facility can be obtained at the [number]th [time]. The failure rate function within a maintenance cycle is: (14), then it is in the first place. Repair cycle The number of repairs within the period is .

[0070] Introducing maintenance method identification factors When performing minor repairs on the facilities When carrying out major repairs on facilities Under different maintenance strategies, the failure rate increase factor... and the factor of service age regression It is expressed as follows: (15), (16), where, and These represent the failure rate increment factors for repair and replacement, respectively, and have... , and These represent the service life reduction factors for maintenance and replacement, respectively, and have... .

[0071] (2) Determination of maintenance costs: For a certain facility, let's assume that the maintenance cost is calculated on the 1st... The unit cost of repairs per repair cycle is The replacement cost is The unit cost of preventive maintenance is ,set up For construction period costs, Operation and maintenance costs, thus the total cost of the facility during the construction and operation and maintenance phases, are: (17), where, and These are the construction period cost coefficients, Unit maintenance cost for the facility during the construction phase (initial performance status). for ,in and These are the maintenance cost coefficients during the operation and maintenance period. For the facility in The operational and maintenance effectiveness (operational and maintenance performance status) of each maintenance cycle.

[0072] (3) Determination of risks and losses due to natural deterioration of facilities: In reality, many facilities experience gradual performance degradation due to natural deterioration and cumulative wear. Furthermore, the failure rate of a facility is related not only to its own usage time but also to the number of repairs and the effectiveness of repairs within adjacent repair periods. Generally, the more repairs performed, the higher the failure rate of the facility. If the repair effect of the previous repair was poor, the probability of facility failure is also higher. The failure rate of a facility affects the cycle of preventative maintenance or replacement, thereby affecting the cost of operation and maintenance.

[0073] For a specific facility, if using Indicates that it is in the first The deterioration and failure loss in the maintenance cycle, then in the first maintenance cycle Risk loss cost per maintenance cycle Defined as degradation failure rate With failure loss The product of, i.e. (18), where the degradation failure rate is... With facilities in The operational performance is related to the maintenance cycle, among which For the facility in The performance status of each maintenance cycle, and the deterioration and failure rate of the facility during the first maintenance period. Related to the construction quality during the construction period, Losses due to facility deterioration and failure. , , and These are the procurement cost, residual value rate, average service life, and depreciation value of the facility components after experiencing a failure. and All of these are undetermined coefficients, and these quantities can be obtained from historical statistical data of similar facilities.

[0074] The facility operation and maintenance decision-making model based on performance evolution under the construction service model can be described as follows: (19) ,in, The total service life of the facility components includes the construction phase and the service life of the facility components. The performance status of each maintenance cycle meets the requirements. .

[0075] The sequential cycle operation and maintenance decision model, under the construction service model, considers the optimal life cycle cost and tries to avoid problems such as over-maintenance or under-maintenance that may be caused by the determination of maintenance intervals and maintenance strategies. Based on the equal cycle facility operation and maintenance decision model, this section considers the case of unequal maintenance intervals and studies the sequential cycle facility operation and maintenance decision model.

[0076] (1) Determination of failure rate: The time point for the service provider to perform preventive maintenance on the facility is set as follows: Facilities in Preventive maintenance should be carried out regularly, during which replacement or repair can be performed. The facility should be in [a certain condition / location]. Failure rate function within a maintenance cycle It can be represented as: (20), where, Adding a factor to the failure rate indicates that after the first failure... The rate of increase in facility failure rate after preventive maintenance; The service age regression factor indicates the number of years of service after which service age has decreased. The rate of reduction in the effective service life of a facility after a preventive maintenance.

[0077] According to the recursive algorithm, the facility can be obtained at the [number]th [time]. The failure rate function within a maintenance cycle is: (21), then it is in the first place. Repair cycle The number of repairs within the period is .

[0078] (2) Maintenance strategy selection factor: Introduce maintenance strategy selection factor When performing minor repairs on the facilities When carrying out major repairs on facilities Under different maintenance strategies, the failure rate increase factor... and the factor of service age regression It is expressed as follows: (twenty two), (23), where, and These represent the failure rate increment factors for minor and major repairs, respectively, and have... , and These represent the service life reduction factors for maintenance and replacement, respectively, and have... .

[0079] (3) Determination of risk loss due to natural deterioration of facilities: For a specific facility, if... Indicates the project number The deterioration and failure loss in the first maintenance cycle, then the facility in the first maintenance cycle Risk loss cost per maintenance cycle Defined as facility degradation failure rate Rather than deterioration and failure loss The product of, i.e. (24), where the degradation failure rate is _____. With facilities in The operational performance is related to the maintenance cycle, among which For the facility in The performance status of each maintenance cycle, and the deterioration and failure rate of the facility during the first maintenance period. Related to the construction quality during the construction period, Losses due to facility deterioration and failure. , , and These are the facility's procurement cost, residual value rate, average service life, and depreciation value after experiencing a failure. and All of these are undetermined coefficients, and these quantities can be obtained from historical statistical data of similar facilities.

[0080] (3) Determination of total operation and maintenance cost: Assuming the facility is in the first stage of operation and maintenance... Repairs are performed every repair cycle, and there are two repair methods: minor repair and major repair. The unit costs for minor repair and major repair are respectively... and Its replacement cost is The unit cost of preventive maintenance is The total maintenance cost of the facility is: (25), where, and These are the construction period cost coefficients, The unit cost of minor repairs and major maintenance for the facility during the construction phase (initial performance status). and They are respectively and ,in For the facility in Performance status for each maintenance cycle. This serves as a maintenance method identification factor when a facility undergoes major repairs. When performing minor repairs on the facilities .

[0081] In summary, the facility operation and maintenance decision-making model based on performance evolution under the construction service can be described as follows: (26) ,in, The total service life of the facility components refers to the period during which the facility components are constructed and during the service life of the facility components. The performance status of each maintenance cycle meets the requirements. .

[0082] A smart performance governance system for urban public infrastructure based on digital twin technology includes: a construction service process performance monitoring module, which constructs a multi-level data monitoring system under digital twin technology and uses a constructed evaluation model to assess performance status. Based on the performance monitoring objectives, it makes decisions on construction service strategies; a full lifecycle performance evolution simulation module, used for performance evolution and visualization simulation of the design, construction, operation, and maintenance processes, while feeding back the results of the performance evolution simulation to the design and construction processes, and transmitting the data to a performance evolution simulation database; a performance status update module, including data analysis and mining, training and optimizing models using machine learning and artificial intelligence algorithms, and proposing corresponding performance governance strategies and methods; and a digital twin module, which constructs digital models and digital projections of the physical space of urban public infrastructure, and combines the practical data from the construction service process performance monitoring module to control and make decisions regarding the construction service process. Real data is stored in a database; a digital model is constructed through multi-dimensional mapping, and combined with virtual data from the full lifecycle performance evolution simulation module for verification and simulation, with the simulated data stored in the database; further, digital twin technology is used to monitor and predict the performance of urban public infrastructure according to relevant rules and logic, and dynamic adjustments and decisions are made based on experience and knowledge, which is then fed back to the physical space of the facilities; the blockchain smart contract module defines the role of managers based on the performance monitoring data in the digital twin module, and creates corresponding contract logic based on actual business contract relationships, combined with performance contract terms, to realize the digitalization of the quality / payment process, complete acceptance, and make payment, etc.

[0083] The core of the digital twin module described in this invention lies in constructing and maintaining a virtual digital twin that is fully synchronized with the physical urban public infrastructure (such as bridges, pipelines, and tunnels) throughout its entire lifecycle and with high fidelity. This twin is not merely a geometric model, but a computable, simulable, and predictable intelligent entity integrating physical attributes, behavioral rules, and real-time data. Its construction and operation specifically include the following layers:

[0084] 1. Construction of multi-dimensional and multi-scale digital twins: Geometric model layer: Based on BIM, GIS or high-precision 3D scanning data, construct a high-precision 3D geometric model of the facility to reflect its spatial structure and form.

[0085] Physical property layer: Assigns material properties (such as elastic modulus and fatigue strength), mechanical properties, and hydraulic properties (such as the roughness coefficient of the pipe network) to the geometric model.

[0086] Behavioral rules layer: Defines the behavioral rules of facilities under load and environmental influence, such as mechanical response models based on the finite element method, or hydraulic models of pipe networks based on hydrodynamic principles.

[0087] Data perception layer: Establishes real-time data interfaces with sensors (strain gauges, flow meters, video surveillance, etc.) on physical facilities to form a twin "sensory system".

[0088] 2. Real-time data-driven and model update mechanism: Data synchronization channel: Establish a low-latency, high-reliability data flow from physical sensors to digital twins through IoT platforms or industrial internet protocols.

[0089] Online model parameter calibration: By using real-time monitoring data and through data assimilation technology or machine learning algorithms, key parameters in the digital twin model are dynamically calibrated (e.g., correcting material degradation parameters), so that the behavior of the virtual model closely approximates the real state of the physical entity.

[0090] 3. Hybrid Simulation and Inference Engine: Physics-based Simulation: Run the aforementioned physical models (such as structural analysis and fluid simulation) in a digital twin to predict the performance of the facility under future loads or extreme conditions in a virtual environment.

[0091] Data-driven prediction: Integrating the performance state evolution projection model described in this invention. For example, using historical performance data accumulated from digital twins, an LSTM-based prediction model is trained to predict future performance trends.

[0092] "What-If" scenario simulation: Allows users to set different "what if" scenarios (such as: traffic increases by 20%, a magnitude 7 earthquake occurs, or a certain maintenance is carried out) on the digital twin, quickly simulating the consequences and providing a basis for decision-making.

[0093] 4. Virtual-Real Closed-Loop Decision-Making and Feedback: Decision Generation: The operation and maintenance decision model runs in the digital twin to evaluate the cost, benefits, and risks of different operation and maintenance strategies in the virtual world; Instruction Issuance: The optimized decision (such as "perform preventive maintenance on pump station A at 2:00 AM next Wednesday") is converted into an executable instruction; Feedback Verification: After the instruction is executed in the physical world, the result data is fed back to the digital twin to verify the effect of the decision and to be used for model iterative optimization, forming a closed loop of "perception-analysis-decision-execution-feedback".

[0094] The relationship between the digital twin module and the performance monitoring module: The digital twin is the central hub and visualization carrier for performance data.

[0095] The relationship between the digital twin module and the evolutionary pre-simulation module: The twin provides the high-fidelity virtual environment and real-time initial state required for the pre-simulation.

[0096] The relationship between the digital twin module and the operation and maintenance decision module: The twin is a simulation sandbox and effect test field for decision-making schemes.

[0097] The relationship between the digital twin module and the blockchain smart contract module: The digital twin is an objective and reliable data source for the triggering conditions of the smart contract.

[0098] Compared with the prior art, the present invention has the following significant advantages:

[0099] Scientific and precise assessment: By constructing a dynamic assessment model that integrates multi-attribute decision-making and fuzzy evaluation, the one-sidedness of traditional single-indicator assessment is overcome, and the complex performance status of infrastructure is accurately quantified and graded.

[0100] Proactive Decision Optimization: An evolutionary model based on dynamic Bayesian networks enables probabilistic prediction of future facility performance. Multi-objective optimization of operation and maintenance strategies within a digital twin environment transforms decision-making from "experience-driven" to "model and data-driven," effectively avoiding over-maintenance or under-maintenance and achieving optimal lifecycle costs.

[0101] Intelligent closed-loop process: A complete closed loop has been constructed, encompassing "physical perception - digital modeling - simulation and deduction - intelligent decision-making - entity execution - data feedback." Digital twins, acting as a bridge between the virtual and physical worlds, make the entire governance process visible, verifiable, and optimizable.

[0102] Transparent and trustworthy execution: This innovative approach combines key performance data with smart contracts. Leveraging the distributed ledger and immutability of blockchain, the trustworthiness of performance data is ensured. Automated payment execution through coded contracts significantly improves contract execution efficiency, reduces human disputes, and truly realizes a "pay-for-performance" business model innovation.

[0103] System self-evolution: By introducing a machine learning-driven model update module, the system can continuously learn from historical data and automatically optimize internal model parameters, making it more and more accurate over time and with the accumulation of data, thus possessing adaptive growth capabilities.

[0104] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several changes and improvements without departing from the overall concept of the present invention, and these should also be considered within the scope of protection of the present invention.

Claims

1. A smart performance governance method for urban public infrastructure based on digital twin technology, characterized in that, Includes the following steps: S1. Performance Data Collection and Modeling Stage: Real-time monitoring and acquisition of multi-source performance data of urban public infrastructure; construction of a performance evaluation and evolution model based on the multi-source performance data, including a performance evaluation indicator system, a performance status evaluation model, a performance status evolution deduction model, and an operation and maintenance decision model; and generation of an initial performance database. S2, Dynamic Update Phase of Digital Twin Model: The performance database is continuously trained using machine learning algorithms to update the parameters in the performance state evolution inference model; Based on the updated model parameters, the digital twin model of the urban public infrastructure is updated synchronously so that the digital model reflects the latest performance degradation trend of the physical facilities; S3, Intelligent Operation and Maintenance Decision Generation Phase: Based on the updated digital twin model, the performance state evolution inference model is run to simulate the future evolution path of facility performance under different maintenance strategies; With the goal of minimizing total operation and maintenance costs, the operation and maintenance decision model uses a multi-objective optimization algorithm to jointly optimize maintenance time, maintenance method and maintenance intensity, and generate a recommended optimal operation and maintenance decision scheme. S4. Intelligent Execution Phase of Performance Contracts: The key indicators of the optimal operation and maintenance decision scheme generated in step S3 and bound to the performance status, as well as the real-time performance evaluation results, are stored in the blockchain network. The smart contract deployed on the blockchain monitors the key indicators and evaluation results. When the data meets the predefined performance contract trigger conditions, it automatically executes the corresponding fee payment or quality reward / penalty clauses.

2. The smart performance governance method for urban public infrastructure based on digital twin technology according to claim 1, characterized in that: In step S1, the specific process of constructing the performance evaluation indicator system includes: Preliminary screening of performance influencing factors was conducted through literature review and expert interviews. By using questionnaire surveys and factor analysis, the elements were supplemented, eliminated, and refined to construct a multi-level scenario element system that includes facility reliability, operational efficiency, maintenance costs, and environmental impact.

3. The smart performance governance method for urban public infrastructure based on digital twin technology according to claim 1, characterized in that: The performance status evaluation model is a multi-attribute fuzzy comprehensive evaluation model based on expert weights and factor weights. Its construction process includes: An initial evaluation matrix was constructed based on expert scoring data; The distance metric method based on the ideal solution is used to calculate the consistency weight of each expert's evaluation and the discrimination weight of each scenario's influencing factor. The initial matrix is ​​weighted and aggregated using the aforementioned weights to obtain a comprehensive performance evaluation value. Based on the preset performance level thresholds, the membership degree of the comprehensive performance evaluation value to each performance level is calculated using a triangular or trapezoidal membership function, thus completing the status assessment.

4. The smart performance governance method for urban public infrastructure based on digital twin technology according to claim 2, characterized in that: The performance state evolution inference model is a probabilistic graphical model based on a dynamic Bayesian network; Its construction process includes: The key elements in the multi-level scenario element system are abstracted into network nodes; Define directed edges between nodes based on the causal relationships between elements, and construct a network structure spanning time slices; Based on historical operation and maintenance data or expert experience, conditional probability tables for nodes are determined to describe the stochastic process of facility performance status evolving over time under no intervention or specific maintenance intervention.

5. The smart performance governance method for urban public infrastructure based on digital twin technology according to claim 1, characterized in that: The operation and maintenance decision model includes an equal-cycle model and a sequential-cycle model. The equal-cycle model uses a fixed maintenance interval as the decision variable, while the sequential-cycle model uses a sequence of non-equal-length maintenance intervals as the decision variable. The objective function F for both models is expressed as: F = Total Maintenance Cost (C_m) + Expected Risk Loss (C_r). The total maintenance cost C_m is related to the facility's construction quality, the number of maintenance operations, and the unit maintenance cost. The expected risk loss C_r is related to the facility's performance status and failure rate function within the current maintenance cycle, with the failure rate function increasing stepwise with the number of maintenance operations.

6. The smart performance governance method for urban public infrastructure based on digital twin technology according to claim 1, characterized in that: The multi-source performance data includes: Real-time operational status monitoring data collected through a sensor network deployed in physical facilities; Subjective evaluation data obtained through an expert evaluation system, based on domain knowledge and experience; Satisfaction or evaluation data obtained through a questionnaire survey system from operators, maintainers, or users.

7. A smart performance governance system for urban public infrastructure based on digital twin technology, used to implement the method of any one of claims 1-6, characterized in that: include: A construction service process performance monitoring module is used to access sensor data through an IoT interface and call the performance status evaluation model to calculate the real-time performance status. The full life cycle performance evolution simulation module is connected to the performance monitoring module for receiving real-time performance status and calling the performance status evolution inference model to perform multi-scenario simulation and inference of the future performance of the facility in the digital space, and output evolution trend data. The digital twin core module is bidirectionally connected to both the performance monitoring module and the performance evolution simulation module, and includes: The twin model management unit is used to maintain the three-dimensional digital model and behavioral model corresponding to the physical facility; The data-model driven unit is used to drive the digital model with real-time data from the performance monitoring module, and simultaneously inject simulation data from the performance evolution simulation module into the digital model for verification and optimization. The decision generation unit is used to run the operation and maintenance decision model in the virtual environment provided by the digital model and generate operation and maintenance decision schemes. A blockchain smart contract module, connected to the decision generation unit of the digital twin core module, is used to receive key performance indicators from the operation and maintenance decision scheme; this module includes: Contract logic encapsulation unit, used to encode performance payment terms into automatically executable smart contract code; The performance data on-chain unit is used to compare the consensus-verified actual performance data from the core module of the digital twin with the key performance commitment indicators, and write the comparison results into the blockchain; The automatic execution and settlement unit is used to automatically trigger payment transactions on the blockchain when the smart contract determines that the performance targets have been met.

8. The system according to claim 6, characterized in that: The core module of the digital twin is further connected to a performance status update module; The performance status update module is used to collect historical performance data and decision feedback results, and to use machine learning algorithms to train and fine-tune the parameters of the performance status assessment model, performance status evolution inference model and operation and maintenance decision model offline and online. The optimized model parameters are then pushed to the digital twin core module to form a closed-loop adaptive governance loop of "monitoring-assessment-decision-feedback-optimization".

9. The system according to claim 6, characterized in that: In the blockchain smart contract module, the data written to the blockchain by the performance data on-chain unit includes at least: a comprehensive performance score with a timestamp output by the performance status evaluation model, and key performance indicator values ​​generated by the operation and maintenance decision model as the contract subject. The logic of the smart contract code is as follows: when the comprehensive performance score is consistently higher than the key performance indicator value within the agreed period, full payment is automatically executed; if it is lower than the indicator value, deduction or claim is automatically calculated and executed according to preset rules.

Images