Community old-age care facility planning diagnosis and feedback method based on multiple agents

Abstract

The invention discloses a community old-age care facility planning diagnosis and feedback method based on multiple agents. The method comprises five main steps of constructing a target community spatio-temporal data integration platform, constructing a target community elderly intelligent agent model, constructing a community old-age care facility planning scheme diagnosis intelligent agent, diagnosing a target community old-age care facility layout scheme, and performing decision feedback on the target community old-age care facility layout scheme. The invention aims to accurately analyze differentiated space-time behaviors of the old people in the community and excavate rules by applying a crowd digital portrait technology, and construct an old people intelligent agent in a target community; and determining key index composition of community pension facility planning scheme diagnosis and constructing a diagnosis agent, so that scheme diagnosis of facility layout can be performed and optimization decision feedback can be performed in combination with a real community physical environment and a pension facility planning scheme. According to the invention, layout diagnosis and optimization suggestion output of the target community old-age care facility planning scheme can be realized.

Description

Technical Field

[0001] The present invention relates to a multi-agent-based community elderly care facility planning diagnosis and feedback method, which belongs to the field of intelligent planning and decision support technology, and in particular to a method for diagnosing and optimizing community elderly care facility planning schemes using a multi-agent system. Background Art

[0002] With the aging of the population, the planning and layout of community elderly care facilities has become a focus of social attention. Because the needs of the elderly are often passive, multidimensional, and internally diverse, current elderly care facility planning faces supply-demand mismatches such as insufficient and delayed supply, and uneven facility distribution. Reasonable elderly care facility planning not only improves the quality of life of the elderly but also directly impacts the effective allocation of community resources and the efficiency of public services.

[0003] In practical applications, common planning methods include: one is to allocate elderly care facilities in proportion to the total regional population; the other is to optimize the allocation by constructing an evaluation index system to analyze the supply and demand relationship and the degree of coupling and coordination of urban life service facilities. Most of the above traditional planning methods rely on static indicators for demand analysis, lack dynamic adjustment of facility distribution, and fail to effectively identify and adapt to the personalized needs of different elderly groups. These problems are particularly prominent when faced with the rapid changes in the aging process of the community. Therefore, there is an urgent need for an intelligent planning tool that can comprehensively consider multiple factors, provide real-time feedback, and flexibly adjust to meet the challenges brought by aging. Summary of the Invention

[0004] Purpose of the Invention: This invention aims to provide a solution for dynamically adjusting and optimizing the layout of community elderly care facilities based on a multi-agent intelligent planning and decision-making feedback mechanism. By incorporating multi-agent technology, a multi-dimensional, real-time responsive elderly care facility planning and decision-making support system is constructed, tailored to specific community characteristics, elderly needs, and real-time data feedback. This system addresses the limitations and inadequacies of existing planning methods for community elderly care facility planning.

[0005] The technical solution adopted by the present invention is a multi-agent-based community elderly care facility planning diagnosis and feedback method, which includes:

[0006] Step S1: Construct a spatiotemporal data integration platform for the target community.

[0007] Obtain high-precision LBS data, land use data, POI data, road network data, and administrative division spatial data for the target community and perform data preprocessing; import the preprocessed data into the geographic information platform to construct a spatiotemporal data integration platform for the target community;

[0008] Step S2: Construction of the target community elderly agent model

[0009] Based on the spatiotemporal data integration platform described in step S1, people aged 60 and above are selected as the elderly group in the target community using the age field, and their stay points and behavior points in the community are identified. The activity behavior chain of the elderly in the target community is constructed, and the label type combination and community elderly portrait pedigree are constructed by combining their individual ontology dimension and spatiotemporal behavior dimension. Based on the above elderly portrait pedigree results, the target community elderly intelligent body is constructed.

[0010] Step S3: Construction of community elderly care facility planning scheme diagnosis agent

[0011] Determine the key indicators for diagnosing community elderly care facility planning schemes, combine the completeness, utilization rate, accessibility, and fairness of community elderly care facility layout planning as constraints, combine current planning standards and the minimum standard method to determine the thresholds of the above community elderly care facility layout diagnostic indicators, and construct a diagnostic intelligent agent agent model for community elderly care facility layout;

[0012] Step S4: Diagnosis of the current layout of elderly care facilities in the target community

[0013] Through the spatiotemporal data integration platform of the target community constructed in step S1, the spatial environment data of the target community and the layout plan of the community elderly care facilities are imported. Combined with the target community elderly agent described in step S2 and the community elderly care facility layout diagnosis agent described in step S3, a multi-agent model of the target community is constructed and run, and the layout diagnosis is performed by the diagnosis agent;

[0014] Step S5: Decision feedback on the layout plan of elderly care facilities in the target community

[0015] The operation and diagnosis results of the multi-agent model described in step S4 are optimized and evaluated to generate an optimization type set of the target community nursing facility planning scheme and corresponding optimization suggestions; then, the target community nursing facility layout plan, diagnosis report results and optimization suggestions are visualized and interactively displayed through urban three-dimensional digital holographic sandbox equipment and virtual reality equipment, and output through printing equipment.

[0016] Furthermore, the LBS data preprocessing method in step S1 is:

[0017] Step S1-11: Identify dwell points. Use a workstation equipped with a multi-core CPU and GPU to process and analyze the LBS data. Cluster the trajectory points in the acquired LBS data that have not moved within 10 minutes or have moved within a range of 100 meters. Identify their average centers as dwell points and generate dwell point information (including user ID, age, dwell point latitude and longitude, dwell duration, and dwell start and end times). The average center means that each dwell will return a point, which is used to represent the average center of the dwell in time and distance. The generated elements will have a time type interval.

[0018] Step S1-12: Data cleaning and effective user screening. An effective user is a user who has a stay point for a number of days greater than or equal to half of the total number of days in the observed data. The specific data cleaning and effective user screening method is as follows: aggregating the data tables of all stay points within the selected research time, summarizing and counting the number of stay days for each user by ID, screening out effective users, and associating all original stay point data by valid user ID. The data processed by step S1-11 is called effective user stay data. A distributed storage system is used to store the original data and calculation results.

[0019] The spatiotemporal data integration platform in step S1 includes basic spatial data and spatiotemporal trajectory data. The basic spatial data is obtained from the Open Street Map website, specifically including land use data, business type POI data, urban road network vector data, and administrative division spatial data; the user spatiotemporal trajectory data includes the valid user stay data obtained in step S1-12. The specific attributes of each type of data are described in the following table:

[0020]

[0021]

[0022] Drones equipped with high-precision cameras and LiDAR sensors capture aerial images and 3D modeling of specific areas, generating high-resolution spatial data to verify and supplement Open Street Map's land use and road network data. IoT sensors monitor pedestrian and vehicle traffic in real time, supplementing user spatiotemporal trajectory data.

[0023] Furthermore, the specific process of constructing the portrait pedigree of the elderly in the community in step S2 is as follows:

[0024] Step S2-11: Constructing the spatiotemporal behavior chain of the elderly. Based on the spatiotemporal data integration platform described in S1, extract the spatiotemporal data of the elderly population over 60 years old, spatially connect the location-based service (LBS) data of the elderly in the target community with the land use data and the 20M buffer area of ​​POI data, and identify their main activity areas and activity types, including residential areas, commercial areas, medical areas, cultural activity areas, and green areas. Based on time series analysis, construct the spatiotemporal behavior chain of the elderly to describe their daily behavior patterns.

[0025] Step S2-12: Label group combination; based on the ontological dimensions (gender, age) and spatiotemporal behavioral dimensions (activity type, frequency, duration) of the elderly, the K-means clustering algorithm is used to divide the elderly into groups and generate label groups; the specific formula is as follows:

[0026] K-means(X,k)

[0027] Where X is the feature vector of the elderly (including ontological dimension and spatiotemporal behavioral dimension), k is the number of clusters, and the optimal k value is determined by the elbow rule;

[0028] Step S2-13: Constructing the elderly portrait spectrum; Based on the label group results, construct the elderly portrait spectrum, taking the longest time spent in the activity area other than the residence during the day (7 am to 7 pm) as the portrait classification basis, use the decision tree algorithm to classify the elderly group, and generate elderly portrait labels, specifically including "healthy and active type", "home-dependent type", and "medical need type" (among which the portrait with only the residence as the activity area is classified as "home-dependent type"). The activity area and portrait classification comparison table is shown below:

[0029]

[0030]

[0031] The decision tree algorithm used is the CART algorithm, and the specific formula is as follows:

[0032] CART(D,A)

[0033] Among them, D is the behavioral activity dataset of the elderly, and A is the feature attribute set.

[0034] Furthermore, the specific process of constructing the target community elderly agent in step S2 is as follows:

[0035] Step S2-21: Initialize the elderly agent model; Based on the elderly portrait spectrum constructed above, extract the behavioral characteristics of each type of elderly people, including activity type, frequency, and duration as the initial attributes of the elderly agent; Use the Mesa framework in Python to create an agent instance and define the agent's attribute set Aa:

[0036] Aa={activity type, frequency, duration}

[0037] Step S2-22: Define the behavioral rules of the elderly agent. Based on the elderly's behavioral patterns, use the Markov decision process (MDP) to define the agent's behavioral rules, simulating the elderly's daily activity paths and preferences for using elderly care facilities. The specific formula is as follows:

[0038] MDP(S m ,A m ,P m ,R m )

[0039] Among them, S m is the state space, A m is the behavior space, P m is the state transition probability matrix, R m is the reward function.

[0040] Furthermore, the key indicators for diagnosing the layout plan of community elderly care facilities in step S3 are specifically composed as follows:

[0041]

[0042]

[0043] Furthermore, the specific construction process of the diagnostic agent agent model in step S3 is as follows:

[0044] Step S3-21: Determine the benchmark value of the intelligent agent diagnostic index; the benchmark value of the diagnostic index of the community elderly care facility layout plan in step S3 is determined according to the current planning standards and the minimum standard method, specifically:

[0045]

[0046]

[0047] Step S3-22: Construction of diagnostic agent indicators; Combined with the agent diagnostic indicator benchmark value described in step S3-21, it is used as a constraint condition for the construction of the community elderly care facility planning diagnostic agent, and the community elderly care facility planning scheme that meets the diagnostic indicator benchmark value is defined as passing the diagnosis, otherwise it is failing the diagnosis, and outputs the specific diagnostic results, including the specific calculation results of the above-mentioned key indicators; among them, the specific rule set of the community elderly care facility planning diagnostic agent is: Completeness rule: If C1=1, then the facility types are complete; otherwise, the facility types are incomplete; Utilization rate rule: If U≥60%, then the facility utilization rate meets the standard; otherwise, the utilization rate is insufficient; Coverage rate rule: If C≥90%, then the coverage rate meets the standard; otherwise, the coverage rate is insufficient; Accessibility rule: If A≤15min, then the accessibility meets the standard; otherwise, the accessibility is insufficient; Fairness rule: If F≤0.3, then the fairness meets the standard; otherwise, the fairness is insufficient.

[0048] Furthermore, the specific construction and operation process of the target community multi-agent model in step S4 is as follows:

[0049] Step S4-1: Construct a multi-agent model framework for the target community. Based on the target community spatiotemporal data integration platform constructed in step S1, import the spatial environment data of the corresponding community, including land use data, business type POI data, urban road network vector data, and community elderly care facility layout plan data. Use a workstation equipped with a multi-core CPU and GPU to run the Mesa framework in Python to build a multi-agent model. Define the spatial environment framework of the model based on the aforementioned spatial data. Initialize the attribute sets of the corresponding agents based on the elderly agent model constructed in step S2 and the diagnostic agent model constructed in step S3.

[0050] Step S4-2: The multi-agent model of the target community is run; the elderly population distribution data of the target community is imported, and the initial residence and specific activity preferences (including home-dependent, healthy and active, and medical needs) are assigned to the elderly agents; within each time step, the daily activity paths and elderly care facility usage behaviors of the elderly agents are simulated according to the Markov decision process based on the behavioral rules of the corresponding elderly agents, and the activity data of the elderly agents are monitored in real time through the diagnostic agents and the actual values ​​of various diagnostic indicators of the elderly care facility plan are calculated; the diagnostic agent determines whether the elderly care facility layout plan of the target community has passed the diagnosis by comparing the actual value with the benchmark value, generates the diagnostic results and outputs the diagnostic report.

[0051] Furthermore, in the optimization suggestion of the target community elderly care facility planning scheme in step S5, if the diagnostic indicator passes, "no optimization required" is output; if the diagnostic indicator fails, the optimization type determination suggestion is specifically:

[0052]

[0053] Furthermore, in step S5, the target community elderly care facility layout plan, diagnosis report results and optimization suggestions are visualized and interactively displayed through urban three-dimensional digital holographic sandbox equipment and virtual reality equipment. Among them, healthy people use panoramic virtual glasses equipped with binocular resolution of 3664x1920 pixels and 98° field of view sensor and display for interaction, and the elderly use an intelligent guide robot with a 3664x1920 pixel camera, text recognition to speech, more than 90% accurate speech recognition, second-level facial recognition technology, meter-level positioning accuracy and high-fidelity voice output equipment to perform real-life interaction of images and text to speech; then, the target community elderly care facility layout plan and the corresponding diagnosis results and optimization suggestions are printed through high-precision 3D printing equipment.

[0054] Beneficial effects:

[0055] Compared with the prior art, the present invention has the following advantages:

[0056] 1. This invention acquires high-precision location-based services (LBS) data, land use data, point of interest (POI) data, road network data, and administrative spatial data from target communities, performs data preprocessing and integration, and builds a unified spatiotemporal data integration platform. This platform then accurately screens seniors aged 60 and over, identifies their stopover points and behavior points, and generates a portrait spectrum of the elderly by combining individual ontology dimensions with spatiotemporal and behavioral dimensions. It also constructs an intelligent model of the elderly. This process enables precise characterization and behavioral analysis of the elderly population, providing a scientific and reliable data foundation and decision-making support for community elderly care facility planning, significantly improving the accuracy and relevance of planning solutions.

[0057] 2. Based on the portrait spectrum of the elderly, the present invention uses the Markov decision process (MDP) to define the behavioral rules of the elderly intelligent agent and simulate their daily activity paths and preferences for the use of elderly care facilities. Combined with the key indicators of the layout of community elderly care facilities, a diagnostic intelligent agent agent model is constructed to achieve intelligent diagnosis and optimization of the layout of elderly care facilities. This process solves the problems of unreasonable layout and low resource utilization in traditional planning. It not only improves the scientificity and rationality of the planning scheme, but also significantly improves the utilization rate and service fairness of elderly care facilities through multi-agent collaborative optimization.

[0058] 3. The present invention constructs a diagnostic intelligent agent model, clarifies the benchmark values ​​of the integrity, utilization rate, coverage rate, accessibility and fairness of community elderly care facilities based on current planning standards and minimum standard methods, and uses them as constraints for the diagnostic intelligent agent. Through a set of rules (such as integrity rules, utilization rate rules, coverage rate rules, accessibility rules, and fairness rules), the planning scheme is intelligently diagnosed and specific diagnostic results are output. This method solves the problems of lax standard implementation and unbalanced resource allocation in traditional planning, and ensures the scientificity, standardization and operability of elderly care facility planning.

[0059] 4. This patent uses a diagnostic agent to accurately assess various indicators (completeness, utilization, coverage, accessibility, and fairness) of community elderly care facility layout plans and outputs targeted optimization recommendations based on the diagnostic results. This solves the problems of ambiguous optimization directions and irrational resource allocation in traditional planning, providing a scientific basis for precise adjustments to planning schemes.

[0060] 5. This patent utilizes a three-dimensional urban digital holographic sandbox, virtual reality equipment, and intelligent navigation robots to visualize and interactively display the target community's elderly care facility layout plan, diagnostic report results, and optimization recommendations in a highly precise, multimodal manner, further enhancing the intuitiveness and operability of planning outcomes. This approach addresses the challenges of traditional planning, such as difficulty in understanding and low public participation, significantly improving the transparency and public engagement of planning proposals and providing an efficient and intuitive communication and feedback channel between decision makers and community residents. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 This is a flow chart of the multi-agent-based community elderly care facility planning, diagnosis and feedback method of the present invention. DETAILED DESCRIPTION

[0062] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0063] The technical solution of the present invention is explained in detail using the diagnosis and decision-making of elderly care facility planning in a community in Nanjing, Jiangsu Province:

[0064] like Figure 1 As shown, a multi-agent-based community elderly care facility planning diagnosis and feedback method includes the following steps:

[0065] Step S1: Construct a spatiotemporal data integration platform for the target community.

[0066] Obtain high-precision LBS data, land use data, POI data, road network data, and administrative division spatial data for the target community and perform data preprocessing; import the preprocessed data into the geographic information platform to build a spatiotemporal data integration platform for the target community. This includes the following steps:

[0067] S1-11 Dwell point identification. Using a workstation equipped with a multi-core CPU and GPU, we process and analyze LBS data. We cluster trajectory points in the acquired LBS data that have not moved within 10 minutes or have moved within a 100-meter range. We identify their mean centers as dwell points and generate dwell point information (including user ID, age, latitude and longitude of the dwell point, dwell duration, and dwell start and end times). The mean center returns a point for each dwell, representing the mean center of the dwell time and distance. The generated features have a time interval type.

[0068] S1-12 Data cleaning and effective user screening. Effective users are defined as those with a stay point for at least half the total number of days observed. The specific data cleaning and effective user screening method is as follows: A data table of all stay points within the selected study period is compiled, and the number of days each user stays is summarized by ID. Effective users are screened out, and all original stay point data is linked by valid user ID. The data processed in step S1-11 is referred to as effective user stay data. The distributed storage system Hadoop HDFS is used to store the original data and calculation results.

[0069] S1-21 builds a spatiotemporal data integration platform. The spatiotemporal data integration platform includes basic spatial data and spatiotemporal trajectory data. Basic spatial data is obtained from the Open Street Map website and specifically includes land use data, business type POI data, urban road network vector data, and administrative division spatial data. User spatiotemporal trajectory data includes the valid user stay data obtained in step S1-11. The specific attributes of each type of data are described in the following table:

[0070]

[0071] Drones equipped with high-precision cameras and LiDAR sensors capture aerial images and 3D modeling of specific areas, generating high-resolution spatial data to verify and supplement Open Street Map's land use and road network data. IoT sensors monitor pedestrian and vehicle traffic in real time, supplementing user spatiotemporal trajectory data.

[0072] Step S2: Construction of the target community elderly agent model

[0073] Based on the spatiotemporal data integration platform described in step S1, people aged 60 and above are selected as the elderly group in the target community using the age field. The elderly's stay points and behavior points in the community are identified, and the activity behavior chain of the elderly in the target community is constructed. In addition, the label type combination and community elderly portrait pedigree are constructed based on their individual ontology dimension and spatiotemporal behavior dimension. Based on the above elderly portrait pedigree results, the target community elderly intelligent body is constructed. Specifically, the following steps are included:

[0074] S2-11: Constructing the spatiotemporal behavior chain for the elderly. Based on the spatiotemporal data integration platform described in step S1, extract spatiotemporal data for the elderly population aged 60 and over. Spatial connection is performed between the location-based services (LBS) data of the elderly in the target community and the 20M buffer zone of the land use data and point of interest (POI) data. This identifies their primary activity areas and types, specifically residential, commercial, medical, cultural, and green spaces. Based on time series analysis, construct a spatiotemporal behavior chain for the elderly to describe their daily activity patterns.

[0075] S2-12 Label group combination. Based on the ontological dimensions (gender, age) and spatiotemporal behavioral dimensions (activity type, frequency, duration) of the elderly, the K-means clustering algorithm is used to divide the elderly into groups and generate label groups. The specific formula is as follows:

[0076] K-means(X,k)

[0077] Where X is the feature vector of the elderly (including ontological dimension and spatiotemporal behavior dimension), k is the number of clusters, and the optimal k value is determined by the elbow rule.

[0078] S2-13 Construction of elderly portrait spectrum. Based on the results of the label group, the elderly portrait spectrum was constructed. The longest time spent in the activity area outside the residence during the day (7 am to 7 pm) was used as the portrait classification basis. The decision tree algorithm was used to classify the elderly group and generate elderly portrait labels, including "healthy and active type", "home-dependent type", and "medical need type" (among which the portrait with only the residence as the activity location is classified as "home-dependent type"). The activity location and portrait classification comparison table is shown below:

[0079]

[0080] The decision tree algorithm used is the CART algorithm, and the specific formula is as follows:

[0081] CART(D,A)

[0082] Among them, D is the behavioral activity dataset of the elderly, and A is the feature attribute set.

[0083] S2-21: Initialize the elderly agent model. Based on the elderly portrait spectrum constructed in S2-1, extract the behavioral characteristics of each type of elderly person, including activity type, frequency, and duration, as the initial attributes of the elderly agent. Use the Mesa framework in Python to create an agent instance and define the agent's attribute set Aa:

[0084] Aa={activity type, frequency, duration}

[0085] S2-22: Definition of behavioral rules for the elderly agent. Based on the behavior patterns of the elderly, the Markov decision process (MDP) is used to define the behavior rules of the agent, simulating the elderly's daily activity paths and preferences for using elderly care facilities. The specific formula is as follows:

[0086] MDP(S m ,A m ,P m ,R m )

[0087] Among them, S m is the state space, A m is the behavior space, P m is the state transition probability matrix, R m is the reward function.

[0088] Step S3: Construction of community elderly care facility planning scheme diagnosis agent

[0089] Determine the key indicators for diagnosing community elderly care facility planning schemes, combine the completeness, utilization rate, accessibility, and fairness of community elderly care facility layout planning as constraints, combine current planning standards and the minimum standard method to determine the thresholds of the above community elderly care facility layout diagnostic indicators, and construct a diagnostic intelligent agent agent model for community elderly care facility layout. The specific steps include:

[0090] S3-11: Set key indicators for diagnosing community elderly care facility layout plans. The details are as follows:

[0091]

[0092]

[0093] S3-21: Determination of the benchmark values ​​of the diagnostic indicators of the intelligent body. The benchmark values ​​of the diagnostic indicators of the community elderly care facility layout plan in step S3 are determined according to the current planning standards and the minimum standard method, specifically:

[0094]

[0095]

[0096] S3-22: Construction of diagnostic agent indicators. Combined with the diagnostic indicator benchmark values ​​of the agent described in step S3-21, they are used as constraints for the construction of the community elderly care facility planning diagnostic agent. The community elderly care facility planning scheme that meets the diagnostic indicator benchmark values ​​is defined as passing the diagnosis, otherwise it is not passing the diagnosis, and the specific diagnostic results are output, including the specific calculation results of the above key indicators. Among them, the specific rule set of the community elderly care facility planning diagnostic agent is: completeness rule: if C1=1, then the facility types are complete; otherwise, the facility types are incomplete; utilization rate rule: if U≥60%, then the facility utilization rate meets the standard; otherwise, the utilization rate is insufficient; coverage rate rule: if C≥90%, then the coverage rate meets the standard; otherwise, the coverage rate is insufficient; accessibility rule: if A≤15min, then the accessibility meets the standard; otherwise, the accessibility is insufficient; fairness rule: if F≤0.3, then the fairness meets the standard; otherwise, the fairness is insufficient.

[0097] Step S4: Diagnosis of the current layout of elderly care facilities in the target community

[0098] Through the spatiotemporal data integration platform of the target community constructed in step S1, the spatial environment data of the target community and the layout plan of the community elderly care facilities are imported. Combined with the target community elderly agent described in step S2 and the community elderly care facility layout diagnosis agent described in step S3, a multi-agent model of the target community is constructed and run, and the layout diagnosis is performed by the diagnosis agent. Specifically, the following steps are included:

[0099] S4-1: Constructing the multi-agent model framework for the target community. Based on the target community spatiotemporal data integration platform constructed in step S1, import the corresponding community's spatial environment data, including land use data, point of interest (POI) data, road network data, and community elderly care facility layout plan data. Use a workstation equipped with a multi-core CPU and GPU to run the Mesa framework in Python to build the multi-agent model. Define the spatial environment framework of the model based on the aforementioned spatial data. Initialize the attribute sets of the corresponding agents based on the elderly agent model constructed in step S2 and the diagnostic agent model constructed in step S3.

[0100] S4-2: Target community multi-agent model operation. Import the elderly population distribution data of the target community and assign the elderly agents an initial residence and specific activity preferences (including home-dependent, healthy and active, and medical needs). At each time step, the Markov decision process is used to simulate the elderly agent's daily activity path and elderly care facility usage behavior according to the corresponding elderly agent's behavioral rules. The diagnostic agent monitors the elderly agent's activity data in real time and calculates the actual values ​​of various diagnostic indicators of the elderly care facility plan. The diagnostic agent compares the actual values ​​with the baseline values ​​to determine whether the target community's elderly care facility layout plan has been diagnosed, generates diagnostic results, and outputs a diagnostic report.

[0101] Step S5: Decision feedback on the layout plan of elderly care facilities in the target community

[0102] The operation and diagnosis results of the multi-agent model described in step S4 are optimized and evaluated to generate an optimization type set of the target community elderly care facility planning scheme and corresponding optimization suggestions; furthermore, the target community elderly care facility layout plan, diagnosis report results and optimization suggestions are visualized and interactively displayed through urban three-dimensional digital holographic sandbox equipment and virtual reality equipment, and output through a printing device. Specifically, the following steps are included:

[0103] S5-1: Generate optimization suggestions for the target community elderly care facility planning scheme. If the diagnostic indicators are passed, the output is "no optimization required"; if the diagnostic indicators are not passed, the optimization type determination suggestions are as follows:

[0104]

[0105] S5-2: Visualization and Interactive Presentation of the Plan. The target community's elderly care facility layout plan, diagnostic report results, and optimization recommendations are visualized and interactively presented using urban 3D digital holographic sandbox equipment and virtual reality devices. Healthy individuals interact with the plan using panoramic virtual glasses equipped with binocular sensors and displays with a resolution of 3664x1920 pixels and a 98° field of view. Seniors interact with the plan using an intelligent guide robot equipped with a 3664x1920 pixel camera, text-to-speech recognition, over 90% accurate speech recognition, sub-second facial recognition technology, an intelligent navigation system with meter-level positioning accuracy, and a high-fidelity voice output device. The target community's elderly care facility layout plan, along with the corresponding diagnostic results and optimization recommendations, is then printed using high-precision 3D printing equipment.

[0106] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention, and such changes and modifications fall within the scope of the invention as claimed.

Claims

1. A multi-agent-based community elderly care facility planning diagnosis and feedback method, characterized by: The method comprises the following steps: Step S1: Constructing a spatiotemporal data integration platform for the target community Obtain high-precision LBS data, land use data, POI data, road network data, and administrative division spatial data for the target community and perform data preprocessing; import the preprocessed data into the geographic information platform to construct a spatiotemporal data integration platform for the target community; Step S2: Construction of the target community elderly agent model Based on the spatiotemporal data integration platform described in step S1, people aged 60 and above are selected as the elderly group in the target community using the age field, and their stay points and behavior points in the community are identified. The activity behavior chain of the elderly in the target community is constructed, and the label type combination and community elderly portrait pedigree are constructed by combining their individual ontology dimension and spatiotemporal behavior dimension. Based on the above elderly portrait pedigree results, the target community elderly intelligent body is constructed. Step S3: Construction of community elderly care facility planning scheme diagnosis agent Determine the key indicators for diagnosing community elderly care facility planning schemes, combine the completeness, utilization rate, accessibility, and fairness of community elderly care facility layout planning as constraints, combine current planning standards and the minimum standard method to determine the thresholds of the above community elderly care facility layout diagnostic indicators, and construct a diagnostic intelligent agent agent model for community elderly care facility layout; Step S4: Diagnosis of the current layout of elderly care facilities in the target community Through the spatiotemporal data integration platform of the target community constructed in step S1, the spatial environment data of the target community and the layout plan of the community elderly care facilities are imported. Combined with the target community elderly agent described in step S2 and the community elderly care facility layout diagnosis agent described in step S3, a multi-agent model of the target community is constructed and run, and the layout diagnosis is performed by the diagnosis agent; Step S5: Decision feedback on the layout plan of elderly care facilities in the target community The operation and diagnosis results of the multi-agent model described in step S4 are optimized and evaluated to generate an optimization type set of the target community nursing facility planning scheme and corresponding optimization suggestions; then, the target community nursing facility layout plan, diagnosis report results and optimization suggestions are visualized and interactively displayed through urban three-dimensional digital holographic sandbox equipment and virtual reality equipment, and output through printing equipment.

2. The multi-agent-based community elderly care facility planning diagnosis and feedback method according to claim 1 is characterized in that: The pre-processing method of LBS data in step S1 is: Step S1-11: Identify dwell points. Use a workstation equipped with a multi-core CPU and GPU to process and analyze the LBS data. Cluster the trajectory points in the acquired LBS data that have not moved within 10 minutes or have moved within a range of 100 meters. Identify their average centers as dwell points and generate dwell point information. The average center means that each dwell will return a point to represent the average center of the dwell in time and distance. The generated features will have a time type interval. Step S1-12: data cleaning and effective user screening; the effective user is a user who has a stay point for more than or equal to half of the total number of days of observation data; the specific data cleaning and effective user screening method is: summarize the data table of all stay points within the selected research time, summarize and count the number of stay days of each user by ID, screen out effective users, and associate all original stay point data by effective user ID, and the data processed by step S1-11 is called effective user stay data; use a distributed storage system to store the original data and calculation results.

3. The multi-agent-based community elderly care facility planning diagnosis and feedback method according to claim 2 is characterized in that: The spatiotemporal data integration platform in step S1 includes basic spatial data and spatiotemporal trajectory data. The basic spatial data is obtained from the Open Street Map website, specifically including land use data, business type POI data, urban road network vector data, and administrative division spatial data; the user spatiotemporal trajectory data includes the valid user stay data obtained in step S1-12. The specific attributes of each type of data are described in the following table: Using drones equipped with high-precision cameras and lidar to conduct aerial photography and 3D modeling of specific areas, we can obtain high-resolution spatial data to verify and supplement the land use and road network data in Open Street Map. Use IoT sensors to monitor pedestrian and vehicle flows in real time and supplement user spatiotemporal trajectory data.

4. The multi-agent-based community elderly care facility planning diagnosis and feedback method according to claim 3 is characterized in that: The specific process of constructing the portrait pedigree of the elderly in the community in step S2 is as follows: Step S2-11: Constructing the spatiotemporal behavior chain of the elderly. Based on the spatiotemporal data integration platform described in S1, extract the spatiotemporal data of the elderly population over 60 years old, spatially connect the location-based service (LBS) data of the elderly in the target community with the land use data and the 20M buffer area of ​​POI data, and identify their main activity areas and activity types, including residential areas, commercial areas, medical areas, cultural activity areas, and green areas. Based on time series analysis, construct the spatiotemporal behavior chain of the elderly to describe their daily behavior patterns. Step S2-12: Label group combination; based on the ontological dimension and spatiotemporal behavior dimension of the elderly, the K-means clustering algorithm is used to divide the elderly into groups and generate label groups; the specific formula is as follows: K-means(X,k) Where X is the feature vector of the elderly, k is the number of clusters, and the optimal k value is determined by the elbow rule; Step S2-13: Constructing a portrait spectrum of the elderly. Based on the label group results, construct a portrait spectrum of the elderly. Use the longest time spent in the activity area outside the residence during the day as the basis for portrait classification. Use the decision tree algorithm to classify the elderly group and generate elderly portrait labels, including "healthy and active type", "home-dependent type", and "medical need type". The comparison table of activity area and portrait classification is as follows: The decision tree algorithm used is the CART algorithm, and the specific formula is as follows: CART(D,A) Among them, D is the behavioral activity dataset of the elderly, and A is the feature attribute set.

5. The multi-agent-based community elderly care facility planning diagnosis and feedback method according to claim 4 is characterized in that: The specific process of constructing the target community elderly agent in step S2 is as follows: Step S2-21: Initialize the elderly agent model; Based on the elderly portrait spectrum constructed above, extract the behavioral characteristics of each type of elderly people, including activity type, frequency, and duration as the initial attributes of the elderly agent; Use the Mesa framework in Python to create an agent instance and define the agent's attribute set Aa: Aa={activity type, frequency, duration} Step S2-22: Define the behavioral rules of the elderly agent. Based on the elderly's behavioral patterns, use the Markov decision process to define the agent's behavioral rules, simulating the elderly's daily activity paths and preferences for using elderly care facilities. The specific formula is as follows: MDP(S m ,A m ,P m ,R m ) Among them, S m is the state space, A m is the behavior space, P m is the state transition probability matrix, R m is the reward function.

6. The multi-agent-based community elderly care facility planning diagnosis and feedback method according to claim 5 is characterized in that: The key indicators for diagnosing the layout plan of community elderly care facilities in step S3 are as follows:

7. The multi-agent-based community elderly care facility planning diagnosis and feedback method according to claim 1 is characterized in that: The specific construction process of the diagnostic agent agent model in step S3 is as follows: Step S3-21: Determine the benchmark value of the intelligent agent diagnostic index; the benchmark value of the diagnostic index of the community elderly care facility layout plan in step S3 is determined according to the current planning standards and the minimum standard method, specifically: Step S3-22: Construction of diagnostic agent indicators; Combined with the agent diagnostic indicator benchmark value described in step S3-21, it is used as a constraint condition for the construction of the community elderly care facility planning diagnostic agent, and the community elderly care facility planning scheme that meets the diagnostic indicator benchmark value is defined as passing the diagnosis, otherwise it is failing the diagnosis, and outputs the specific diagnostic results, including the specific calculation results of the above-mentioned key indicators; among them, the specific rule set of the community elderly care facility planning diagnostic agent is: Completeness rule: If C1=1, then the facility types are complete; otherwise, the facility types are incomplete; Utilization rate rule: If U≥60%, then the facility utilization rate meets the standard; otherwise, the utilization rate is insufficient; Coverage rate rule: If C≥90%, then the coverage rate meets the standard; otherwise, the coverage rate is insufficient; Accessibility rule: If A≤15min, then the accessibility meets the standard; otherwise, the accessibility is insufficient; Fairness rule: If F≤0.3, then the fairness meets the standard; otherwise, the fairness is insufficient.

8. The multi-agent-based community elderly care facility planning, diagnosis and feedback method according to claim 7 is characterized in that: The specific construction and operation process of the target community multi-agent model in step S4 is as follows: Step S4-1: Construct a multi-agent model framework for the target community. Based on the target community spatiotemporal data integration platform constructed in step S1, import the spatial environment data of the corresponding community, including land use data, business type POI data, urban road network vector data, and community elderly care facility layout plan data. Use a workstation equipped with a multi-core CPU and GPU to run the Mesa framework in Python to build a multi-agent model. Define the spatial environment framework of the model based on the aforementioned spatial data. Initialize the attribute sets of the corresponding agents based on the elderly agent model constructed in step S2 and the diagnostic agent model constructed in step S3. Step S4-2: The multi-agent model of the target community is run; the elderly population distribution data of the target community is imported, and the initial residence and specific activity preferences are assigned to the elderly agents; within each time step, the daily activity paths and elderly care facility usage behaviors of the elderly agents are simulated according to the Markov decision process based on the behavioral rules of the corresponding elderly agents, and the activity data of the elderly agents are monitored in real time through the diagnostic agents and the actual values ​​of various diagnostic indicators of the elderly care facility plan are calculated; the diagnostic agents judge whether the elderly care facility layout plan of the target community has passed the diagnosis by comparing the actual values ​​with the benchmark values, generate diagnostic results and output a diagnostic report.

9. The multi-agent-based community elderly care facility planning, diagnosis and feedback method according to claim 8, characterized in that: In step S5, if the diagnostic indicators are passed, the optimization suggestion for the target community elderly care facility planning scheme is output as "no optimization required"; if the diagnostic indicators are not passed, the optimization type determination suggestion is specifically as follows:

10. The multi-agent-based community elderly care facility planning, diagnosis and feedback method according to claim 9, characterized in that: In step S5, the target community elderly care facility layout plan, diagnosis report results and optimization suggestions are visualized and interactively displayed through urban three-dimensional digital holographic sandbox equipment and virtual reality equipment. Among them, healthy people use panoramic virtual glasses equipped with binocular resolution of 3664x1920 pixels and 98° field of view sensor and display for interaction, and the elderly use an intelligent guide robot with a 3664x1920 pixel camera, text recognition to speech, more than 90% accurate speech recognition, second-level facial recognition technology, meter-level positioning accuracy and high-fidelity voice output equipment to perform real-life interaction of images and text to speech; then, the target community elderly care facility layout plan and the corresponding diagnosis results and optimization suggestions are printed through high-precision 3D printing equipment.

Images