Influenza decision-making method for megacities based on spatiotemporal flow of people with different economic attributes

Abstract

The present application relates to the technical field of influenza prevention and control, and more particularly to a large city influenza decision-making method based on spatiotemporal flow of people with different economic attributes. The steps are as follows: a meta-population flow network model based on individuals, groups and space is established, and a large city influenza evolution model is constructed by fusing a classic SEIRD model; a multi-objective optimization model is constructed, with the best influenza transmission control effect and the minimum economic loss as the target, and the regional blockade and the flow limitation of six types of people with different economic attributes between regions as the main decision. The large city influenza decision-making method based on spatiotemporal flow of people with different economic attributes provided by the present application, in view of the importance and complexity of influenza prevention and control in large cities, considers the economic attributes of different people in large cities, constructs an influenza evolution model based on spatiotemporal flow of people with different economic attributes, and carries out integrated decision-making of regional blockade and flow limitation of people with different economic attributes based on the double targets of influenza transmission control and economic impact.

Description

Technical Field

[0001] This invention relates to the field of influenza prevention and control technology, and in particular to an influenza decision-making method for megacities based on the spatiotemporal movement of populations with different economic attributes. Background Technology

[0002] In recent years, the global public health system has faced unprecedented challenges, and major public health emergencies have severely impacted global socio-economic development and public health security. Mega-cities, with their dense populations and frequent economic activities, have become high-risk areas for influenza transmission. Due to high population mobility and complex social structures, influenza transmission in megacities exhibits significant density, clustering, overlapping, and rapid spread, exacerbating the chain reaction, complexity, and multiplicity of influenza's impact.

[0003] The spread of influenza across populations and regions is a typical spatiotemporal evolutionary phenomenon and a process of human-environment interaction. Influenza prevention and control is essentially a spatiotemporal issue of the fight between humans and viruses. Population mobility in megacities, while driving economic prosperity and development, has also become a major route of influenza transmission. The economic development of megacities is highly dependent on the continuous flow of people across different industries. Therefore, interventions related to population mobility during influenza prevention and control inevitably have a significant impact on the economy. Thus, how to effectively control the spread of influenza while minimizing its negative economic impact is a crucial and pressing practical problem that needs to be addressed in influenza prevention and control management.

[0004] In summary, influenza evolution analysis based on infectious disease models and research on influenza prevention and intervention measures have become research hotspots in recent years. However, current research still has the following shortcomings in serving influenza prevention and control in megacities and in carrying out emergency management of major influenza outbreaks from the perspective of balancing influenza prevention and control with economic development.

[0005] (1) Most current influenza evolution models focus on the macro scale of a country or region. These models usually assume that the individuals involved in the evolution are homogeneous and equivalent. Although these studies provide important perspectives for understanding the macro spread of influenza, few studies have focused on the impact of the complexity of population mobility and social structure in megacities on influenza evolution, and thus cannot provide a scientific basis for precise influenza prevention and control in megacities.

[0006] (2) Most studies explore influenza control strategies from the perspectives of infection scale and economic loss. However, infection scale is directly related to social stability, while economic loss is related to economic development. These are two key and conflicting goals in influenza control. Currently, few studies attempt to comprehensively consider these two goals in order to seek a control strategy that balances influenza transmission control with minimizing economic impact.

[0007] Therefore, a decision-making method for influenza in megacities based on the spatiotemporal movement of populations with different economic attributes is designed to provide an alternative technical solution to the above-mentioned technical problems. Summary of the Invention

[0008] Therefore, it is necessary to provide a decision-making method for influenza in megacities based on the spatiotemporal movement of populations with different economic attributes, in order to solve the technical problems mentioned in the background.

[0009] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0010] A decision-making method for influenza in megacities based on the spatiotemporal movement of populations with different economic attributes, the steps of which are as follows:

[0011] Establish a meta-population mobility network model based on individuals, groups, and space, and integrate the classic SEIRD model to construct an influenza evolution model for megacities;

[0012] Construct a multi-objective optimization model with the goal of achieving the best influenza transmission control effect and minimizing economic losses, and with regional lockdown and flow restriction of six groups of people with different economic attributes between regions as the main decision-making factors;

[0013] The multi-objective optimization model was solved using the NSGA-II algorithm.

[0014] As a preferred embodiment of the influenza decision-making method for megacities based on the spatiotemporal flow of populations with different economic attributes provided by the present invention, the population flow network model comprises the following steps:

[0015] Through the directed spatiotemporal network graph G=(V,A) t ) represents a population mobility network model;

[0016] Where V represents the set of nodes, referring to the various regions that constitute the urban space, and A... t ={(i,j) t Let :i,j∈V} be the set of directed arcs at time t∈T, where T is the decision cycle duration. Any node is connected to each other through the various types of people flowing between nodes every day (t∈T). If there is a flow of people between any two nodes, then there is an arc connection between them.

[0017] Since there are six types of population flows along any arc, let R = {1,2,...,6} represent the sets of populations with different economic attributes, namely management and administrative support workers, retail workers, production and manufacturing workers, entertainment and leisure workers, service workers, and consumers. Then we have:

[0018] in, Let arc(i,j)∈A represent the time t∈T. t The number of people moving in the r∈R class.

[0019] As a preferred embodiment of the influenza decision-making method for megacities based on the spatiotemporal movement of populations with different economic attributes provided by the present invention, a megacity influenza evolution model is constructed by integrating the classic SEIRD model, and the steps are as follows:

[0020] Individuals at each node are categorized according to their health status into susceptible individuals, latent individuals, infected individuals, recovered individuals, and deceased individuals.

[0021] Susceptible individuals within each node are converted into latent individuals according to their infection probability. Latent individuals become infected individuals after the incubation period. Infected individuals recover and acquire immunity after the infection period. Some infected individuals die due to the mortality rate of influenza.

[0022] As a preferred embodiment of the influenza decision-making method for megacities based on the spatiotemporal flow of populations with different economic attributes provided by the present invention, a multi-objective optimization model is constructed with influenza transmission control and economic loss as objectives, and regional lockdown and flow restriction of six groups of populations with different economic attributes between regions as the main decision-making steps. The steps are as follows:

[0023] The target for influenza transmission control is the cumulative number of infections within the control period T, which is used as an indicator to measure the effectiveness of influenza transmission control. The expression is as follows:

[0024]

[0025] The economic loss target is calculated based on three parts: economic loss of health, economic loss of employment, and loss of consumption.

[0026] Urban population flow is restricted by constraints on medical resource capacity, new infection numbers, population flow balance, and network connectivity.

[0027] As a preferred embodiment of the influenza decision-making method for megacities based on the spatiotemporal movement of populations with different economic attributes provided by the present invention, the economic loss target is calculated based on three parts: health economic loss, employment economic loss, and consumption loss. The steps are as follows:

[0028] Health economic loss refers to the economic burden and losses caused by influenza, expressed as follows:

[0029]

[0030] Among them, the economic losses directly caused by the infection at time t Let θ1 be the cost of rehabilitation treatment per unit case, θ2 be the cost of treatment for death per unit case, and θ3 be the opportunity cost of death. and These represent the number of infected individuals and the number of dead individuals belonging to node i at time t, respectively.

[0031] The economic loss in employment consists of the loss of employment compensation and the loss of production, expressed as follows:

[0032]

[0033] in, and w represents the number of people of the r-th group traveling from node i to node j at the initial time and time t of the influenza evolution. r For industry average salary r, p r For industry, r represents the per capita industry output; for time t, it represents the cumulative loss of returns among employed individuals. and industry cumulative production losses

[0034] At any given moment, employment and economic losses

[0035] Loss in consumption mainly refers to the potential loss in consumption caused by a reduction in supply due to a decrease in the employed population. Consumption that did not occur due to a direct decrease in the consumer base The expression is as follows:

[0036]

[0037] in, and c represents the number of people of the r-th group traveling from node i to node j at the initial time of influenza evolution and at time t, respectively. pc v represents per capita consumption. c p represents the economic value generated per unit of consumption. r The industry's per capita industry output value;

[0038] Consumption loss at time t

[0039] Minimize economic loss objective Z Economy The expression is as follows:

[0040]

[0041] in, Represents any arc at time t∈T The mobility proportion of the r-th group of people is l (l∈L). This represents the open and closed state of node i∈V at time t∈T; This indicates the node is open; otherwise, the node is closed.

[0042] The objective function can be expressed as:

[0043] Z = min(Z) Infect Z Economy ).

[0044] As a preferred embodiment of the influenza decision-making method for megacities based on the spatiotemporal movement of populations with different economic attributes provided by the present invention, the multi-objective optimization model is solved using the NSGA-II algorithm, and the steps are as follows:

[0045] An initial population is generated using a three-stage construction algorithm;

[0046] The selection, crossover, and mutation genetic operations are applied sequentially to continuously generate offspring populations;

[0047] By employing a dynamic elite strategy based on non-dominated ranking and crowding distance to select new populations in order to maintain population diversity and superiority, a Pareto optimal solution is obtained.

[0048] As a preferred embodiment of the influenza decision-making method for megacities based on the spatiotemporal movement of populations with different economic attributes provided by the present invention, an initial population is generated through a three-stage construction algorithm, the steps of which are as follows:

[0049] S1: Open all nodes and randomly assign flow ratios for the six types of people between any two nodes from the population flow ratio set L, thereby obtaining the seed solution S0;

[0050] S2: By adjusting the opening and closing states of nodes and the proportion of crowd flow, the seed solution S0 is disturbed to generate an expanded initial population with 2N solutions;

[0051] When constructing each solution, firstly, randomly select 0.1|V| nodes and change their state from open to closed; then, randomly select 0.5|V| crowd flow arcs and randomly perturb the crowd flow ratio on these arcs.

[0052] S3: Perform a fast non-dominated sort of individuals in the population according to their fitness values ​​and adopt an elite selection strategy to select the top N individuals to form the initial population.

[0053] As a preferred embodiment of the influenza decision-making method for megacities based on the spatiotemporal movement of populations with different economic attributes provided by the present invention, the following steps are taken to continuously generate offspring populations by sequentially applying selection, crossover, and mutation genetic operations:

[0054] Parental individuals were selected using a pairwise tournament selection method involving crossover and mutation operations;

[0055] For the two parent individuals undergoing crossover, single-point crossover is performed based on their two chromosome strings C and F respectively;

[0056] The values ​​of randomly selected gene loci are perturbed to create mutations, which are used to increase the diversity of the population.

[0057] It is clear without a doubt that the technical solution described above in this application can solve the technical problem that this application aims to address.

[0058] Meanwhile, through the above technical solutions, the present invention has at least the following beneficial effects:

[0059] 1. The influenza decision-making method for megacities based on the spatiotemporal flow of populations with different economic attributes provided by this invention addresses the importance and complexity of influenza prevention and control in megacities. It considers the economic attributes of different populations in megacities, constructs an influenza evolution model based on the spatiotemporal flow of populations with different economic attributes, and conducts integrated decision-making on regional lockdown and flow restriction of populations with different economic attributes under the dual objectives of influenza transmission control and economic impact.

[0060] 2. This invention addresses the impact of the spatiotemporal movement of different economic groups in megacities on influenza evolution and economic development. It establishes a meta-population flow network model based on individuals, groups, and space, and integrates the classic SEIRD model to construct a megacity influenza evolution model. Secondly, considering the conflict between the goals of influenza transmission and economic loss, a multi-objective optimization model is proposed to minimize both. This model combines regional lockdowns with flow restrictions for different population groups, and an improved NSGA-II algorithm is designed to solve the model based on the problem characteristics. Finally, using influenza prevention and control in the main urban area of ​​City A during an influenza outbreak as a case study, the effectiveness of the influenza evolution model and the prevention and control decision model is verified. Sensitivity analysis of key parameters yields several management insights beneficial to influenza prevention and control in megacities. Attached Figure Description

[0061] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a schematic diagram of influenza evolution based on population flow networks with different economic attributes, as presented in this invention.

[0063] Figure 2 This is the chromosome coding scheme of the present invention;

[0064] Figure 3 This is a schematic diagram illustrating the selection of a new population according to the present invention;

[0065] Figure 4 This is a schematic diagram of the cross-operation of the present invention;

[0066] Figure 5 This is a schematic diagram of the variation operation of the present invention;

[0067] Figure 6 This is a schematic diagram comparing simulated values ​​and statistical values ​​within the evolutionary cycle of this invention;

[0068] Figure 7 This is a heat map showing the distribution of infection cases in different regions according to the present invention;

[0069] Figure 8 This is a schematic diagram of the model's running results according to the present invention;

[0070] Figure 9 This is a schematic diagram illustrating the population flow network and the control measures at each node under different schemes of the present invention.

[0071] Figure 10 This is a schematic diagram comparing the Pareto solution sets under different virus infection rates according to the present invention;

[0072] Figure 11 This is a schematic diagram comparing the Pareto solution sets under different decision-making periods of the present invention;

[0073] Figure 12 This is a schematic diagram illustrating the changes in economic losses and the number of infections during the evolutionary cycle of this invention;

[0074] Figure 13 This is a schematic diagram illustrating the sensitivity analysis of different population types according to the present invention. Detailed Implementation

[0075] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0076] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0077] It should be noted that, unless otherwise specified, the embodiments and features and technical solutions in the present invention can be combined with each other.

[0078] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0079] Reference Figures 1-13 A decision-making method for influenza in megacities based on the spatiotemporal movement of populations with different economic attributes.

[0080] 1. Model Building

[0081] 1.1 Problem Solving

[0082] The main industries constituting urban economic activity are divided into five categories: management and administrative support (scientific research, public administration, etc.), retail (wholesale, retail, etc.), production and manufacturing (manufacturing, construction, etc.), entertainment and leisure (culture, sports and entertainment, etc.), and services (leasing, business, etc.). These five categories of productive economic activity groups, along with the consumer population, together constitute the six groups of people involved in urban economic activity. Different groups contribute differently to the economy, mainly reflected in industry output, wages, and consumption expenditure, thus possessing different economic attributes.

[0083] Based on administrative divisions and industrial characteristics, megacities are divided into a series of sub-regions (streets). Urban economic activities drive the movement of people with different economic attributes between different regions. In an influenza environment, individuals within each region with different health statuses (susceptible, infected, and latent, etc.) undergo changes in health status through contact behavior driven by economic activities within the current cycle, and return to their destination through a "origin-destination" link to participate in the next cycle's evolution. The influenza evolution based on population movement networks with different economic attributes is illustrated below. Figure 1 As shown.

[0084] In the process of influenza prevention and control, while regional lockdowns and restrictions on the movement of people between regions are used to control influenza, they inevitably lead to economic losses due to a decrease in the employed population, losses in consumption due to a decrease in the consumer population, and economic losses due to infection.

[0085] 1.2 Population Mobility Network Model

[0086] Based on the above analysis, a meta-population mobility network model based on individuals, groups, and space is first constructed to capture the dynamic changes of different groups under different decision-making states. The population mobility network model can be expressed as a directed spatiotemporal network graph G=(V,A) t Where V represents the set of nodes, referring to the various areas (streets, towns, or urban functional zones) that constitute the urban space; A t ={(i,j) t Let :i,j∈V} be the set of directed arcs at time t∈T, where T is the decision cycle duration. Any node is connected to each other through the various types of people flowing between nodes daily (t∈T). If there is a flow of people between any two nodes, then there is an arc connection between them. There is a flow of people. Since there are six types of population flows along any arc, let R = {1,2,...,6} represent the sets of populations with different economic attributes, namely management and administrative support workers, retail workers, production and manufacturing workers, entertainment and leisure workers, service workers, and consumers. Then we have: in, Let arc(i,j)∈A represent the time t∈T. t The number of people moving in the r-th class of R.

[0087] 1.3 Influenza Evolution Model

[0088] Based on the classic SEIRD model, the evolution of influenza at each node at time t∈T is carried out, and the infection response and spread are carried out through the population flow network at time t+1. This realizes the spatiotemporal dynamic evolution of influenza across the entire period, providing basic support for influenza control and economic loss calculation.

[0089] According to the SEIRD model, individuals at each node are classified into four groups based on their health status: Susceptible, Exposed, Infected, Recovered, and Deceased. Susceptible individuals within each node are converted to exposed individuals based on their infection probability; exposed individuals become infected after an incubation period; infected individuals recover and acquire immunity after an infectious period; some infected individuals die due to the mortality rate of influenza. Without loss of generality, the following assumptions are made regarding the evolutionary process:

[0090] (1) Once infected individuals are identified, they are promptly controlled and do not follow local management; only latent and susceptible individuals will move between regions and participate in the spread of the virus.

[0091] (2) Since the influenza evolution cycle is relatively short, the impact of newborns and deaths during the decision-making cycle on the influenza evolution is not considered.

[0092] (3) Influenza is transmitted only through direct contact, without considering other possible transmission routes.

[0093] Based on the above assumptions, for the influenza evolution within each node i∈V, the following symbols, parameters, and state variables are defined: The number of susceptible individuals belonging to node i who appear at node j at time t;

[0094] The number of lurkers belonging to node i who appear at node j at time t;

[0095] The number of infected individuals belonging to node i who appear at node j at time t;

[0096] The number of recovered patients belonging to node i who appear at node j at time t;

[0097] The number of people who died at time t belonging to node i;

[0098] The total number of people at node i at time t;

[0099] The total number of susceptible individuals at node i at time t;

[0100] The total number of lurking nodes at time t;

[0101] The total number of infected persons at node i at time t;

[0102] The total number of recovered patients at node i at time t;

[0103] The total number of deaths at node i at time t;

[0104] Total number of susceptible individuals migrating to node i at time t;

[0105] Time t: Total number of susceptible individuals migrating out of node i;

[0106] Total number of lurking individuals migrating to node i at time t;

[0107] Total number of lurking individuals who migrated out of node i at time t;

[0108] β i The probability of a susceptible individual at node i contracting an infected individual from contact with an infected individual;

[0109] δ i : The probability that a latent individual at node i will become a confirmed case;

[0110] γ i : The recovery rate of infected individuals at node i;

[0111] μ i : The mortality rate of infected individuals at node i.

[0112] Based on the above parameter definitions, the state transition equation of the SERID influenza evolution model for each region node i at time t+1 is as follows:

[0113]

[0114] Formulas (1) and (2) indicate that the susceptible and latent populations participating in the evolution of each node consist of those remaining locally and those migrating from other nodes; formulas (3) and (4) indicate that the susceptible and latent populations return to their respective nodes after the evolution of each node is completed; formula (5) indicates the change in the number of susceptible individuals. Formula (6) represents the number of infected individuals; formula (6) represents the change in the number of latent individuals, where... Formula (7) represents the number of confirmed infections; formula (8) represents the change in the number of infected persons, where and , respectively, represent the number of people who have transitioned from an infected state to a recovered state and to a death state in the wards; formulas (8) and (9) represent the changes in the number of people in the recovered state and the death state in the wards, respectively.

[0115] By applying the above node-oriented influenza evolution model to the entire population flow network model within the evolution period T, the spatiotemporal evolution state of influenza can be obtained.

[0116] 1.4 Multi-objective optimization model

[0117] Based on the above analysis, and taking into account the main practical constraints of influenza prevention and control management and ensuring the normal operation of society under the influenza environment, a multi-objective optimization model is constructed with the goal of achieving the best influenza transmission control effect and minimizing economic losses, and with regional lockdown and flow restriction of six groups of people with different economic attributes between regions as the main decision-making factors.

[0118] 1.4.1 Decision Variables

[0119] There are two main types of decisions: regional lockdown and population flow restriction. Regional lockdown decisions refer to controlling the closed or open state of a region by selectively closing or opening certain nodes in a population flow network. Population flow restriction decisions refer to limiting the flow of six types of people between any two nodes in a population flow network. Based on the requirements for restricting population flow at different risk levels in real-world emergency management, flow restriction is reflected in the proportion of population flow at different levels, denoted as L. Therefore, the two types of decision variables are defined as follows:

[0120] The open and closed states of node i∈V at time t∈T. This indicates that the node is open; otherwise, the node is closed.

[0121] Any arc at time t∈T The mobility proportion of the r-th group of people is l (l∈L).

[0122] 1.4.2 Objective Function

[0123] The two objectives are to achieve the best possible control of influenza transmission and to minimize economic losses.

[0124] (1) Influenza transmission control objectives

[0125] The cumulative number of infections within the prevention and control period T is used as an indicator to measure the effectiveness of influenza transmission control, and the calculation formula is shown in formula (10).

[0126]

[0127] in, Let represent the total number of infected individuals at node ii at time t.

[0128] (2) Economic loss target

[0129] Drawing on the methods for quantifying economic losses in the context of influenza proposed in the literature, we calculate economic losses from three aspects: health economic losses, employment economic losses, and consumption losses. The calculation methods for each aspect are as follows.

[0130] ① Health economic loss. Health economic loss refers to the economic burden and losses caused by influenza, including not only the direct costs incurred in treating and managing the disease, but also the indirect economic losses resulting from illness or death. The health economic loss in this invention mainly includes the treatment costs of patients, the treatment costs of deceased patients, and the opportunity loss resulting from individual death. Since the costs of infection treatment and recovery differ from those of infection treatment and death, the cost of recovery treatment per unit case is defined as θ1, the cost of treatment for death per unit case as θ2, and the opportunity loss cost caused by death as θ3. Therefore, the direct economic loss caused by infection at time t is... The calculation formula is as follows:

[0131]

[0132] Among them, the economic losses directly caused by the infection at time t Let θ1 be the cost of rehabilitation treatment per unit case, θ2 be the cost of treatment for death per unit case, and θ3 be the opportunity cost of death. and These represent the number of infected and deceased individuals belonging to node i at time t, respectively.

[0133] ②Employment economic losses. Employment economic losses mainly include two aspects: first, the loss of employment compensation due to the reduction in the employed population caused by lockdown and movement restrictions; and second, the production losses due to the lack of personnel in work positions. Definition w represents the population size at the initial stage of influenza evolution. r For industry average salary r, p r Let r be the industry's per capita industry output. Then, at time t, the cumulative loss of returns for the employed population. and industry cumulative production losses The calculation formula is as follows:

[0134]

[0135] in, and w represents the number of people of the r-th group traveling from node i to node j at the initial time and time t of the influenza evolution. r For industry average salary r, p r For industry, r represents the per capita industry output; for time t, it represents the cumulative loss of returns among employed individuals. and industry cumulative production losses

[0136] Therefore, at any given moment, there is an economic loss of employment.

[0137] ③ Consumption loss. In this invention, consumption loss mainly refers to the potential consumption loss caused by the reduction in supply due to a decrease in the employed population. Consumption that did not occur due to a direct decrease in the consumer base definition Let c be the consumption at node i at time t. pc v represents per capita consumption. c This refers to the economic value generated per unit of consumption. and The calculation formula is as follows:

[0138]

[0139] in, and Let be the number of people of the r-th class traveling from node i to node j at the initial time of influenza evolution and at time t, respectively. Let c be the consumption at node i at time t. pc v represents per capita consumption. c p represents the economic value generated per unit of consumption. r The industry's per capita industry output value;

[0140] Therefore, the consumption loss at time t

[0141] Based on the above analysis, the objective of minimizing economic loss Z is... Economy It can be represented as:

[0142]

[0143] in, Represents any arc at time t∈T The mobility proportion of the r-th group of people is l (l∈L). This represents the open and closed state of node i∈V at time t∈T; This indicates that the node is open; otherwise, it indicates that the node is closed.

[0144] The objective function of the research problem can be expressed as:

[0145] Z = min(Z) Infect Z Economy (17)

[0146] 3. Constraints

[0147] (1) Constraints on medical resource capacity

[0148] In reality, influenza prevention and control involves limited medical resources and varying degrees of illness among infected individuals. Ensuring that patients' treatment needs are met is essential for maintaining social stability during influenza control. Therefore, the model considers the constraint of limited medical resource capacity, primarily reflected in the number of hospital beds at each node (region). and number of ICU patients limit.

[0149]

[0150] in, Let ξ represent the total number of infected individuals at node i at time t, and let ξ represent the proportion of severely ill patients among the infected individuals. Constraint formulas (18) and (19) respectively indicate that the number of mild and severe patients in each region at any time t∈T does not exceed the total number of hospital beds and ICU beds.

[0151] (2) New restrictions on infected persons

[0152] To effectively curb the spread of influenza, the number of new infections needs to be kept within a reasonable and manageable range. This means that the number of infected individuals decreases net within a certain period compared to the previous period. Therefore, the following constraint is defined:

[0153]

[0154] in, Let be the total number of infected individuals at node i at time t, and k be the evolution time period. It is worth noting that formula (20) does not constrain all nodes to achieve a simultaneous net decrease in the number of infected individuals; rather, it allows some nodes to maintain a certain infection level while the overall infection rate is declining. By allowing local areas to maintain a certain infection rate within the overall control framework, the relationship between influenza prevention and control and economic activities can be more flexibly balanced, achieving more comprehensive and sustainable influenza management.

[0155] (3) Population flow balance constraint

[0156] At any time t, the population flow of a node should satisfy the inflow-outflow balance. Based on assumption (1), that is, once an infected person is diagnosed, they are isolated locally, the population flow of a node should satisfy the following constraint: the number of susceptible and latent individuals initially flowing in should be equal to the number of susceptible and latent individuals initially flowing out, plus the total number of confirmed cases remaining in the node. Therefore, the following balance constraint is defined:

[0157]

[0158] Among them, I ji Let represent the number of infected individuals who migrated from node j to node i.

[0159] (4) Network connectivity constraints

[0160] Because lockdowns and crowd control measures disrupt normal city operations and can cause social instability due to public resistance, influenza prevention and control efforts must consider both social stability and the maintenance of core urban functions. Therefore, network connectivity constraints are introduced to support the basic operation of critical urban services.

[0161] First, define the population mobility network G = (V, A) t The connected matrix at time t is M. t Its elements It is expressed as follows:

[0162]

[0163] when When two nodes are connected, it means that the two nodes are connected. Otherwise, it is not connected. Define |M t | represents the node size of the largest connected subgraph of network G at time t, and S represents the node set of the connected subgraph. The specific calculation formula is as follows:

[0164]

[0165] In the process of influenza prevention and control management, the population flow network at any time t must have a certain degree of fluidity to support the basic operation of the city, subject to the following constraints:

[0166] |M t |≥ε|V| (24)

[0167] Where ε represents the minimum connectivity required to maintain the operation of the city, and |V| represents the number of nodes in the population flow network.

[0168] 2. Algorithm Solution Design

[0169] 2.1 Algorithm Framework

[0170] Because there is a clear conflict between the objectives of controlling influenza transmission and minimizing economic losses, the proposed optimization model is a typical multi-objective optimization problem. (Deb et al.) [Deb K,Pratap A,Agarwal S,et al.A fast and elitist multiobjective genetic algorithm:NSGA-II[J].IEEE transactions on evolutionary computation,2002,6(2):182-197] Proposed in 2002, NSGA-II (Nondominated Sorting Genetic Algorithm II) is an effective algorithm for solving multi-objective optimization problems. With its fast nondominated sorting method and crowding calculation strategy, this algorithm performs exceptionally well in handling multi-objective optimization problems, attracting widespread attention and application. Therefore, this invention designs an improved Nondominated Sorting Genetic Algorithm (NSGA-II) based on the characteristics of the research problem to solve the model.

[0171] The overall algorithm process is as follows: An initial population is generated through a designed three-stage construction algorithm. Genetic operations such as selection, crossover, and mutation are then applied sequentially to continuously generate offspring populations. A dynamic elitist strategy based on non-dominated sorting and crowding distance is used to select new populations to maintain population diversity and superiority, thereby gradually approaching the Pareto optimal solution. During the algorithm's evolution, to improve optimization efficiency, a local search procedure based on perturbation and adaptation of the solution space after each crossover and mutation operation is designed to enhance the quality of offspring solutions.

[0172] 2.2 Encoding Scheme

[0173] Encoding the solution is the first and crucial step in genetic algorithms; the effectiveness of the encoding scheme directly affects the algorithm's solution complexity and efficiency. Based on the characteristics of the problem, an encoding scheme consisting of two chromosome strings is designed.

[0174] Chromosome string C is a two-level integer list of length |V|, representing node control decisions. The first level stores region indices, and each gene value in the second level represents the control status of the corresponding node: 1 indicates the node is open, and 0 indicates the node is controlled. Chromosome string F is a list of floating-point numbers of length |R|×|V'|×|V'|, where each gene position stores the flow ratio of corresponding population categories between any two nodes, representing the flow restriction decisions for various population categories between regions. Here, V'∈V represents the set of open nodes, obtained from chromosome string C. Figure 1 The corresponding chromosome expression of the scheme is as follows Figure 2 As shown, regions 1-6 are open, with a control state of 1 in chromosome string C, while regions 7-9 are closed, with a corresponding control state of 0. Since regions 1-6 are open, for any node, there is a |R| type of population flow to 6 nodes (including the node itself). That is, any node corresponds to |R|×6 elements to store the flow limiting decision. Therefore, the length of chromosome string F is |R|×6×6.

[0175] 2.3 Initialize the population

[0176] Considering the impact of the quality of the initial population on the convergence and global search capability of the genetic algorithm, this invention designs a three-stage initial population construction algorithm: "first constructing a seed solution—then constructing and expanding the initial population—finally selecting the population," to construct an initial population of size N. The specific steps are as follows:

[0177] Phase 1: Constructing the seed solution. Open all nodes and randomly assign flow proportions for the six types of people between any two nodes from the population flow proportion set L, thereby obtaining the seed solution S0.

[0178] Phase 2: Constructing an expanded initial population. An expanded initial population with 2N solutions is generated by perturbing the seed solution S0 by adjusting the open / closed states of nodes and the flow ratio of the crowd. When constructing each solution, firstly, 0.1|V| nodes are randomly selected and their states are changed from open to closed; then, 0.5|V| flow arcs of the crowd are randomly selected and the flow ratio of the crowd on these arcs is randomly perturbed.

[0179] Phase 3: Population Selection. Individuals within the population are rapidly sorted according to their fitness values ​​using a non-dominated ranking method, and an elite selection strategy is adopted to select the top N individuals to form the initial population.

[0180] 2.4 Fitness Assessment

[0181] Since both objectives of the decision-making model are minimization functions, the objective function is used as the fitness function to evaluate the performance of each individual. According to the multi-objective optimization model, the objective function value Z for each individual is... Infect and Z Economy As an individual fitness value.

[0182] 2.5 Quick Non-Dominated Sort

[0183] For the combined population of parent and offspring during algorithm evolution, based on the individual fitness value Z... Infect and Z Economy The individuals in the population are ranked according to their non-dominant order. The specific steps are as follows:

[0184] Step 1: Calculate dominance relationships. For any pair of individuals (p...) within the population... i ,p j ), compare fitness values. If for all targets, f(p) i )≤f(p j And at least on one target f(p) i )<f(p j ), then p i Dominate p j Noted as p i <p j And p j Add to pi The dominating set S i Increase p j The dominant number n j .

[0185] Step 2: Divide the frontier. Starting from the first frontier set F1, progressively divide the subsequent frontier sets. For each individual p in the first frontier... i Each individual p in its dominion set j , will n j Reduce by 1. For each n... j Individuals whose count drops to 0 are placed in the next frontier.

[0186] Step 3: Repeat the above steps until all individuals have been classified into different levels.

[0187] 2.6 Crowding Calculation

[0188] After obtaining the non-dominant levels of individuals in the population, the non-dominant levels F are further calculated. j Internal crowding First, individuals within each level are classified according to their fitness value Z. Infect and Z Economy Sort in ascending order, then accumulate the congestion level for each objective. The specific calculation formula is as follows:

[0189]

[0190] In the formula, f k i+1 and f k i-1 Representing F respectively j Within the non-dominant hierarchy, the target values ​​f of the adjacent individuals of individual i after sorting according to the k-th target. k max and f k min F j The maximum and minimum values ​​within a non-dominated hierarchy, |F j | represents the number of individuals within this non-dominated level. For the boundary solution, the crowding degree is set to a maximum value M.

[0191] 2.7 Dynamic Elite Population Selection

[0192] Considering that suboptimal solutions help expand the algorithm's search space, and to better balance the concentration and diversity of the algorithm's search, a dynamic elite population selection strategy based on Pareto fronts is designed to effectively manage the population and improve the algorithm's search efficiency. Specifically, for a combined population consisting of parent and offspring populations, when the selection conditions for the new population are met, all individuals in the population are first sorted using a non-dominated ordering to obtain the Pareto front set F = {F1, F2, ..., F...}. m}, and then for any Pareto front set F i Individuals within F are ranked by crowding, and finally, an elite strategy is adopted within this frontier set to dynamically select the top n. i Individuals constitute a new population, among which Select the diagram as shown below. Figure 3 As shown.

[0193] Initially, an equal number of individuals are selected from each Pareto front set to enter the new population. As the number of algorithm iterations increases, the number of individuals selected from the first Pareto front is continuously increased within a certain range according to the step size τ, while the number of individuals selected from other fronts is decreased by an equal amount to ensure the consistency of the new population size. The formula for calculating the number of individuals selected within each Pareto front is as follows:

[0194]

[0195] Where iter is the iteration number. This represents the maximum number of individuals selected from the Pareto first frontier.

[0196] As can be seen, unlike traditional elitist strategies that prioritize selecting elite solutions from the leading Pareto frontiers, the dynamic elitist strategy proposed in this invention incorporates individuals from different Pareto frontiers during the search process, enriching the diversity of solutions. Simultaneously, as the evolutionary search deepens, better individuals have a higher selection probability, resulting in better concentration of the algorithm in the later stages. Thus, this strategy achieves a balance between algorithmic diversity and concentration.

[0197] 2.8 Selection, Crossover, and Mutation Operations

[0198] (1) Select operation

[0199] Parent individuals were selected using a pairwise tournament selection method with crossover and mutation operations. Individuals in the tournament group were compared based on their non-dominated ranking and crowding distance. Individuals with different non-dominated rankings were retained; individuals with the larger crowding distance were retained when the non-dominated rankings were the same.

[0200] (2) Cross operation

[0201] For the two parent individuals undergoing crossover, single-point crossover is performed based on their two chromosome strings C and F. The crossover point is randomly selected during the operation. Offspring 1 inherits the first half of parent 1 and the second half of parent 2, while offspring 2 inherits the opposite, thus forming a new gene combination. Since chromosome strings C and F correspond to node control and population flow ratio decisions, respectively, there is a significant dependency between the two decisions. Therefore, during crossover, chromosome string C is crossed first, and the nodes in chromosome string F with a population flow limit value of 0 are updated based on the crossover result. Then, the crossover operation on chromosome string F is performed. The crossover operation is illustrated below. Figure 4 As shown;

[0202] (3) Mutation operation

[0203] Mutation operations increase population diversity by perturbing the values ​​of randomly selected gene loci. During the operation, mutation operations are performed sequentially on chromosome strings C and F of the individuals participating in the mutation. Specifically, for chromosome string C, 1-3 gene loci are randomly selected, and their values ​​are swapped between 0 and 1 (i.e., changing from 0 to 1 or from 1 to 0). Then, for chromosome string F, 1-3 gene loci are randomly selected, and a new population flow level is selected from the population flow level set L. The mutation operation is illustrated below. Figure 5 As shown;

[0204] 2.9 Local Search Algorithm

[0205] To improve the algorithm's optimization performance, a local search algorithm is used for post-optimization of each new solution generated by crossover and mutation, thereby improving the algorithm's search efficiency and solution quality. The core idea of ​​the local search algorithm is to perform fine-grained exploration of the solution space perturbed by crossover and mutation, considering changes in population flow levels, thus improving the local quality of new individuals. This reduces the randomness of crossover and mutation, thereby improving the algorithm's running efficiency. The algorithm's pseudocode is as follows:

[0206]

[0207]

[0208] 3. Experimental Simulation

[0209] To verify the effectiveness of the proposed model, this invention uses influenza prevention and control in 156 townships and subdistricts across 9 districts and counties in the main urban area of ​​City A as a case study. Simulation analysis was conducted with November 1, 2022, as the initial running date and December 10, 2022, as the ending date. The experimental environment was as follows: the model was implemented using Python 3.11.2, the computer processor was a 12th-generation Intel(R) Core(TM) i7-12700 with a CPU frequency of 2.10GHz, 16.0 GB of RAM, and Windows 10 Professional.

[0210] 3.1 Parameter Settings

[0211] The relevant parameters involved in the experimental simulation mainly include three parts: influenza evolution model parameters, optimization model parameters, and algorithm parameters.

[0212] Regarding the parameters of the influenza evolution model, based on relevant research [周林,黄鹏,代应,等.区域互救与外部救助耦合的突发性疫情初期应急物资协同调度研究[J].中国管理科学,2024,32(08):297-307] The settings are as follows: β = 0.41, δ = 0.4, γ = 0.01, μ = 0.004, ξ = 0.1, k = 10, ε = 0.9.

[0213] Based on the 2021 Economic Statistical Bulletin and the 2022 Statistical Yearbook of City A, the average monthly salaries and industry output values ​​of five different groups are categorized according to the parameters of the decision-making model. Specific figures are as follows: The average monthly salaries per person for management and administrative support, retail, manufacturing, entertainment and leisure, and services are RMB 11,142.25, RMB 6,249.66, RMB 8,022.42, RMB 8,913.8, and RMB 6,685.35, respectively; the corresponding industry output values ​​are RMB 16,713.38, RMB 9,374.49, RMB 12,033.63, RMB 13,370.7, and RMB 10,028.03, respectively; and the average monthly consumption economic value per person is RMB 2,487.5. The treatment costs and opportunity loss costs for each infected and deceased person are set as follows: θ1 = RMB 20,016, θ2 = RMB 300,000, and θ3 = RMB 438,599.1. Considering the strictness of the current lockdown and flow restrictions, the population flow level is set as L={0,0.25,0.5,0.75,1}.

[0214] Regarding the algorithm parameters, through extensive preliminary experiments, the parameter values ​​are as follows: population size N = 100, crossover rate and mutation rate are 0.8 and 0.05 respectively, Iter_LS = 1.5|V|, τ = 0.05, n1 f =0.85N, the algorithm terminates when the best solution does not increase the number of iterations Iter_noimp = 20.

[0215] 3.2 Simulation of Influenza Evolution

[0216] To verify the applicability and effectiveness of the proposed influenza evolution model based on population mobility networks, a simulation analysis of influenza evolution in the nine main urban districts of City A was first conducted under a scenario without lockdown or flow restriction decisions. The comparison between the simulated and actual cumulative infection numbers in the nine main urban districts of City A during the evolution period is shown below. Figure 6 As shown.

[0217] from Figure 6It can be seen that, within the evolutionary cycle, the simulated values ​​obtained by the evolutionary model generally exhibit the same growth trend as the statistical values. In terms of cumulative infections, the evolutionary model yielded 140,130 infections, while the official confirmed cases numbered 132,587, resulting in an error of 5.6%. Considering the circumstances under which influenza was prevalent at the time, including some missed detections and the possibility that some mild or asymptomatic infections went undetected due to technical reasons, the actual number of infections would likely be higher than the statistically confirmed cases. Therefore, this demonstrates that the influenza evolutionary model proposed in this invention can effectively simulate the overall development of influenza.

[0218] Furthermore, from Figure 7 The heatmap showing the distribution of infection cases reveals a good agreement between the simulated and statistical values ​​for the number of infections in most areas, indicating that the proposed influenza evolution model has good accuracy in the evolution of influenza in various urban areas. Among the nine areas, the most severely affected by influenza are Yuzhong District and Yubei District (simulated values ​​of 22,334 and 23,414 cases, respectively, and statistical values ​​of 22,271 and 23,093 cases, respectively). These two areas are densely populated and have frequent traffic, exhibiting high population density and frequent spatiotemporal population movement. This suggests that high population density and population mobility play a significant role in the spread of influenza.

[0219] 3.3 Results of Multi-Target Influenza Control

[0220] Based on the effectiveness of the evolutionary model, the NSGA-II algorithm is run to solve the multi-objective optimization model, and the resulting Pareto front is as follows: Figure 8 As shown, the Pareto front illustrates the relationship between two sets of objectives. It reveals a conflict between the goals of increasing infection numbers and economic losses; a lower cumulative number of infections corresponds to greater economic losses, and vice versa. Economic losses range from 44 billion to 65 billion yuan, while the number of infections varies between 94,800 and 99,000 cases. Overall, the trend shows a shift from high infection numbers and low economic losses to low infection numbers and high economic losses.

[0221] 3.4 Comparative Analysis of Different Prevention and Control Strategies

[0222] To analyze the effectiveness of the multi-objective optimization model on influenza control, three schemes (denoted as Pareto schemes 1, 2, and 3) were selected from the Pareto optimal solution set and compared with the baseline scenario (no control scheme) and schemes under two single-objective optimization strategies (the economic loss minimization scheme and the infection minimization scheme). Table 1 shows the comparison of the above schemes on five key indicators: economic loss, infection number, reduction in the number of workers, reduction in the number of consumers, and number of lockdown nodes. To visually demonstrate the lockdown and flow restriction situations under different schemes, the lockdown intensity is defined as the ratio of the restricted number of people traveling from one node to other nodes to the normal state. Figure 9 The paper presents the urban population flow network and the control measures at each node under Pareto scheme 1, the scheme with the minimum economic loss, and the scheme with the minimum number of infections.

[0223] Table 1 Comparison of Schemes under Different Control Strategies

[0224]

[0225] Table 1 shows that Pareto scheme 1 significantly outperforms the no-control scheme in both key indicators of economic loss and number of infections, indicating that multi-objective control strategies can achieve a better overall decision-making solution across both objectives. Furthermore, compared to single-objective schemes, the three Pareto schemes achieve a more balanced trade-off between the two objectives, compared to the extremely high number of infections resulting from minimizing economic loss, and the dramatic increase in economic loss resulting from minimizing the number of infections. Simultaneously, because the Pareto frontier schemes have a certain non-dominated range for the two objective values, they provide decision-makers with a flexible decision-making space, facilitating the selection of the most suitable decision-making scheme based on actual circumstances and preferences.

[0226] 3.5 Parameter Sensitivity Analysis

[0227] (1) Sensitivity analysis of infection rate

[0228] To reflect the impact of viral infection rate on influenza prevention and control, sensitivity analyses were performed with β = 0.41, 0.51, and 0.61. Simulation results of the multi-objective control model under the three infection rates were compared with those of the uncontrolled scenario. Figure 10 As shown.

[0229] Depend on Figure 10 It is evident that the scale of infection and economic losses increase significantly with the increase in viral infection rate, and the effectiveness of influenza control varies significantly under different infection rates. As the viral infection rate increases, the control effect of the multi-objective optimization model is more pronounced than in the uncontrolled scenario. This indicates that the more severe the influenza outbreak, the more significant the effect of the multi-objective optimization model.

[0230] (2) Sensitivity analysis of decision-making cycle

[0231] To investigate the impact of different lockdown and flow restriction decision-making cycles on influenza control, with other parameters remaining constant, the decision-making cycles for lockdown and flow restriction were set to 7 days, 14 days, 28 days, and 40 days, respectively. The results of the multi-objective optimization model under the four decision-making cycles are as follows: Figure 11 As shown.

[0232] A comparison of influenza control effects under different decision-making cycles reveals that shorter decision-making cycles significantly reduce economic losses and the number of infections, indicating that refined lockdown and flow restriction decisions can respond promptly to changes in influenza patterns, leading to better influenza prevention and control outcomes. Furthermore, to further visualize the changes in the dual objectives under different decision-making cycles, Figure 12 The study presents a comparison of the changes in the dual objectives within the evolutionary period under two scenarios: a decision-making period of 7 days and 40 days. It can be seen that, compared to a 40-day decision-making period, the effect is more significant after day 20 when the decision-making period is 7 days.

[0233] (3) Sensitivity analysis of population type proportion

[0234] To investigate the impact of population changes across different economic sectors on influenza prevention and control and economic losses, the populations of five working groups were adjusted sequentially in increments of 2% within the range of [0, 14%]. The reduction in the working group size was then proportionally applied to the remaining four groups. Since the total number of infected individuals remained constant while population activity patterns remained unchanged, the economic losses resulting from population changes across different groups are as follows: Figure 13 As shown.

[0235] It can be seen that the impact of personnel reduction on the economy varies significantly across different industries. The impact is strongly positively correlated with the per capita output of each industry; the higher the per capita output, the greater the economic loss.

[0236] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A decision-making method for influenza in megacities based on the spatiotemporal movement of populations with different economic attributes, characterized in that: The steps are as follows: Establish a meta-population mobility network model based on individuals, groups, and space, and integrate the classic SEIRD model to construct an influenza evolution model for megacities; Construct a multi-objective optimization model with the goal of achieving the best influenza transmission control effect and minimizing economic losses, and with regional lockdown and flow restriction of six groups of people with different economic attributes between regions as the main decision-making factors; The multi-objective optimization model was solved using the NSGA-II algorithm. The population flow network model comprises the following steps: Through directed spatiotemporal network diagrams Represents a population mobility network model; in, This represents a set of nodes, referring to the various areas that constitute the urban space. for The directed arc set at time, For the decision-making cycle duration, any node uses daily ( Various groups of people flowing between nodes are interconnected, and if there is a flow of people between any two nodes, then there is an arc connection between them. Since there are six types of people moving along any arc, using Let the groups of people with different economic attributes be categorized as management and administrative support workers, retail workers, production and manufacturing workers, entertainment and leisure workers, service workers, and consumers. ; in, express Time arc Upper The number of people moving around; A multi-objective optimization model is constructed with the goals of controlling influenza transmission and minimizing economic losses, and with regional lockdowns and restrictions on the movement of six groups of people with different economic attributes between regions as the main decision-making steps. The steps are as follows: The goal of influenza transmission control is to control the spread of influenza within a certain cycle. The cumulative number of infections within a given period is used as an indicator to measure the effectiveness of influenza transmission control, expressed as follows: The economic loss target is calculated based on three parts: economic loss of health, economic loss of employment, and loss of consumption. Urban population flow is restricted by constraints on medical resource capacity, new infection numbers, population flow balance, and network connectivity.

2. The influenza decision-making method for megacities based on the spatiotemporal movement of populations with different economic attributes as described in claim 1, characterized in that, The following steps are taken to construct an influenza evolution model for megacities by integrating the classic SEIRD model: Individuals at each node are categorized according to their health status into susceptible individuals, latent individuals, infected individuals, recovered individuals, and deceased individuals. Susceptible individuals within each node are converted into latent individuals according to their infection probability. Latent individuals become infected individuals after the incubation period. Infected individuals recover and acquire immunity after the infection period. Some infected individuals die due to the mortality rate of influenza.

3. The influenza decision-making method for megacities based on the spatiotemporal movement of populations with different economic attributes as described in claim 1, characterized in that, The economic loss target is calculated based on three parts: economic loss of health, economic loss of employment, and economic loss of consumption. The steps are as follows: Health economic loss refers to the economic burden and losses caused by influenza, expressed as follows: ； Among them, time Economic losses directly caused by infection The cost of rehabilitation treatment per unit case is defined as The treatment cost per case death is And the opportunity cost of death is , and They belong to the nodes respectively In time The number of infected and deceased; The economic loss in employment consists of the loss of employment compensation and the loss of production, expressed as follows: ； ； in, and These are the initial and final moments of influenza evolution, respectively. by node Go to node The Number of people in each category For the industry Average salary For the industry Per capita industry output; time Cumulative Loss of Compensation for Employed Population and industry cumulative production losses ; time Employment and economic losses ; Loss in consumption mainly refers to the potential loss in consumption caused by a reduction in supply due to a decrease in the employed population. Consumption that did not occur due to a direct decrease in the consumer base The expression is as follows: ； ； in, and These are the initial and final moments of influenza evolution, respectively. by node Go to node The Number of people in each category For a moment node Consumption volume Per capita consumption The economic value generated per unit of consumption. For the industry Per capita industry output value; time Loss of consumption ; Minimize economic losses The expression is as follows: ； in, Indicates time Arbitrary arc Upper The proportion of this type of population mobility is , ; Indicates time node The status of opening and lockdown; This indicates the node is open; otherwise, the node is closed. The objective function is expressed as: 。 4. The influenza decision-making method for megacities based on the spatiotemporal movement of populations with different economic attributes as described in claim 1, characterized in that, The multi-objective optimization model is solved using the NSGA-II algorithm, and the steps are as follows: An initial population is generated using a three-stage construction algorithm; The selection, crossover, and mutation genetic operations are applied sequentially to continuously generate offspring populations; By employing a dynamic elite strategy based on non-dominated ranking and crowding distance to select new populations in order to maintain population diversity and superiority, a Pareto optimal solution is obtained.

5. The influenza decision-making method for megacities based on the spatiotemporal movement of populations with different economic attributes as described in claim 4, characterized in that, The initial population is generated using a three-stage construction algorithm, with the following steps: S1: Open all nodes, representing the set of population flow ratios for six types of people between any two nodes. The flow ratio is randomly assigned, thereby obtaining the seed solution. ; S2: Perturb the seed solution by adjusting the node's open / closed state and the proportion of crowd flow. Generate with Expanding the initial population for quantitative solutions; When constructing each solution, first randomly select a number of... The node is changed from open to closed; then, a random number of nodes are selected. The flow arc of the crowd is defined and the proportion of the crowd flow on that flow arc is randomly disturbed. S3: Perform a rapid non-dominated ranking of individuals within the population based on their fitness values ​​and adopt an elite selection strategy, prior to selection... Individuals form the initial population.

6. The influenza decision-making method for megacities based on the spatiotemporal movement of populations with different economic attributes as described in claim 4, is characterized in that, The selection, crossover, and mutation genetic operations are applied sequentially to continuously generate offspring populations, as follows: Parental individuals were selected using a pairwise tournament selection method involving crossover and mutation operations; For the two parent individuals undergoing crossover, based on their two chromosome segments... and Perform single-point intersections separately; The values ​​of randomly selected gene loci are perturbed to create mutations, which are used to increase the diversity of the population.

Images