A public service facility elderly accessibility prediction method

Abstract

The application provides a public service facility old people accessibility prediction method, which predicts the old people population distribution according to the mobile phone user age structure information and the regional population natural growth law, and on the basis, combines the public facility resource supply and the spatial distribution condition, and uses the Gaussian two-step moving search method to predict the urban old people public service facility accessibility. The application of the technical scheme can predict the old people population spatial distribution condition in the future target time, and predict the old people population to evaluate the demand condition of the public service facility.

Description

Technical Field

[0001] This invention relates to the field of geographic information service technology, and in particular to a method for predicting the accessibility of public service facilities for the elderly. Background Technology

[0002] my country is rapidly entering an aging society. Understanding the future needs of the elderly is crucial for supporting decisions on optimizing the social security and service system, and for more effectively meeting their aspirations for a better life. Public service facilities are a basic necessity for the daily activities of the elderly, and the accessibility and convenience of these resources directly impact their quality of life. Predicting the future needs of the elderly for public service facilities is of great significance for guiding the formulation of age-friendly policies and effectively meeting the needs of an aging society.

[0003] The shortcomings of existing technologies are: (1) In terms of population prediction methods, traditional methods can only obtain the total number of elderly people in a macro region, lack detailed population distribution information, and are difficult to effectively support the measurement of the accessibility of public service facilities for the elderly; while the population distribution prediction method based on mobile phone data is currently mainly used to predict the dynamic population distribution in a short period of time, such as each hour of the day, and due to the general lack of age information, it has not yet been used to predict the spatial distribution of the elderly population over a long period of time; (2) Due to the lack of effective spatial distribution information of the elderly, existing accessibility measurement methods have not yet been used to support the prediction of the accessibility of public service facilities for the elderly. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a method for predicting the accessibility of public service facilities for the elderly, which can predict the spatial distribution of the elderly population in a future target time and assess the demand for public service facilities based on the predicted elderly population.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for predicting the accessibility of public service facilities for the elderly, comprising the following steps:

[0006] The first step is to clean the mobile phone location data, including the following aspects: (1) removing duplicate values; (2) deleting null values; (3) deleting typical error values, which refers to records with abnormal latitude and longitude, anonymous user ID, and time.

[0007] The second step is to identify the user's home location. This is done by identifying the areas where the user spends the most time at night. Specifically, the location where the user stays for the longest time is identified based on the length of time the user spends in each place. If the location identified by the user exceeds a certain duration threshold or is the location where the user spends the most time, then that location is identified as the user's home location. Otherwise, the location closest to the centroid of these locations is selected as the user's home location.

[0008] The third step is to identify the elderly and potential elderly population; (1) Identify the existing elderly population, that is, among the users whose home locations are identified, users aged 60 or above are identified as elderly and their home locations are obtained; (2) Identify the potential elderly population, determine the screening conditions according to the predicted time period, and obtain the home locations of potential elderly people.

[0009] The fourth step is to predict the elderly population. The number of elderly people in the corresponding area this year is calculated by subtracting the number of elderly people who died from the base number of elderly people in the previous year and adding the net increase in the number of elderly people. (1) The study area is divided into regular grids as spatial analysis units. The elderly people whose residences fall within the grids are summarized to obtain the number of elderly people in the grids, which is used as the base number for elderly prediction. (2) The number of elderly people in each grid corresponding to the target time is predicted according to the elderly population prediction model.

[0010] The fifth step is to acquire data on public service facilities. This includes a collection of various types of public service facilities, or specific types of public service facilities such as tertiary hospitals and parks and green spaces. The relevant data includes the location and service capacity information of the public service facilities. The service capacity reflects the scale of services that the corresponding public facilities can provide to the public. The acquisition method is determined according to actual needs, including direct acquisition in cooperation with relevant competent authorities or self-collection from authoritative public service platforms.

[0011] Step 6: Time cost calculation; obtain the time required for people to travel from their permanent residence to public service facilities, mainly by assessing the path distance and mode of travel between the two; calculate the optimal distance between the two using road network data, and assess the required time information based on the capacity of common transportation tools; or obtain it directly from the route planning interface provided by an authoritative platform.

[0012] Step 7: Accessibility prediction of public service facilities for the elderly; using the Gaussian two-step moving search method, with the predicted number of elderly people as the demand scale, and based on the service scale of public service facilities, calculate the accessibility value of public service facilities for the elderly.

[0013] In a preferred embodiment, the elderly population prediction specifically refers to: the number P of the elderly population in a local area in the next year. m+1 Equal to the number of elderly people P in year m m Subtract its number of deaths η m In addition to the net increase in the number of people μ m As shown in Formula 1:

[0014] P m+1 =P m -η m +μ m (1)

[0015] Number of deaths η m Based on the base population P of the elderly population in that year m The mortality rate was calculated, and considering that the mortality rate of the elderly was significantly higher than that of the general population, a correction factor α was used to adjust the mortality rate of the general population in that year. m The mortality rate of the elderly is obtained by making corrections; as shown in Formula 2:

[0016] η m =P m ×s m ×α (2)

[0017] Net new number of people μ m The mortality rate of the potential elderly population is calculated by subtracting the number of deaths from the potential elderly population number Δ in that year. Considering the difference between the mortality rate of the potential elderly population and the mortality rate of the total population, a correction factor β is used to correct for the mortality rate sm of the total population in that year, thus obtaining the mortality rate of the potential elderly population; as shown in Formula 3:

[0018] μ m =Δ-Δ×s m ×β (3)

[0019] Annual total population mortality rate s m The mortality rate is estimated based on historical mortality data obtained from the corresponding regional population bulletins or statistical yearbooks. Taking a univariate linear regression model as an example, this illustrates how a mortality prediction model is constructed to determine the corresponding annual mortality rate sm based on historical mortality rates; specifically, as shown in Formula 4:

[0020] s m = a×m+b (4)

[0021] Where a and b are the coefficients to be evaluated, which can be assessed using the least squares principle based on historical mortality rates; while the mortality correction coefficients for the elderly and potential elderly in Formulas 2 and 3 are calculated based on the number of deaths in different age groups and the total number of deaths in the population published by authoritative departments, as shown in Formulas 5 and 6:

[0022]

[0023]

[0024] In the formula, POP0 is the total population, POP1 is the total number of deaths, p0 is the elderly population, p1 is the elderly population (i.e., the number of deaths in the age group of 60 years or older), p'0 is the potential elderly population (specifically, the population in the age group of 50 years or older but less than 60 years old), and p'1 is the number of deaths in the potential elderly population age group.

[0025] In a preferred embodiment, the accessibility calculation of public service facilities specifically includes:

[0026] The first step is to calculate the supply-demand ratio. First, the location of the public service facility is obtained as the supply point j. Then, based on the facility's service standards, the time service threshold d0 of supply point j is determined. Next, all demand points (grid centroids i) within the time threshold d0 of all supply points are searched, and the supply-demand ratio R for each supply point j is calculated. j ,Right now

[0027]

[0028] In the formula, d ij S represents the travel time between demand point i and supply point j; j Let D be the service scale corresponding to point j; i G(d) represents the number of elderly people at demand point i; ij d0) is a Gaussian equation function, and its calculation formula is:

[0029]

[0030] Step 2: Calculate accessibility; for each demand point i, search for all supply points j within the time threshold d0 of i, and calculate the supply-demand ratio R for each supply point falling within the spatial domain. j Weights are assigned using a Gaussian function, and these weighted supply-demand ratios R are then... j By summing the results, the reachability A of demand point i can be calculated. i The calculation formula is as follows:

[0031]

[0032] In the formula, R j Within the search area centered on demand point i, d ij ≤d0, the supply-demand ratio of public service facility j, A i The larger the value, the better the accessibility.

[0033] Compared with the prior art, the present invention has the following beneficial effects: it can predict the spatial distribution of the elderly population in the future target time period, and assess the demand for public service facilities based on the prediction of the elderly population. Attached Figure Description

[0034] Figure 1 This is a schematic diagram of a method for predicting the accessibility of public service facilities for the elderly, according to a preferred embodiment of the present invention. Detailed Implementation

[0035] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0036] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0037] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations according to this application; as used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise; furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0038] A method for predicting the accessibility of public service facilities for the elderly, with reference to Figure 1 It includes the following steps:

[0039] The first step is to clean the mobile phone location data, including the following aspects: (1) removing duplicate values; (2) deleting null values; (3) deleting typical error values, which refers to records with abnormal latitude and longitude, anonymous user ID, and time.

[0040] The second step is to identify the user's home location. This is done by identifying the areas where the user spends the most time at night. Specifically, the location where the user stays for the longest time is identified based on the length of time the user spends in each place. If the location identified by the user exceeds a certain duration threshold or is the location where the user spends the most time, then that location is identified as the user's home location. Otherwise, the location closest to the centroid of these locations is selected as the user's home location.

[0041] The third step is to identify the elderly and potential elderly population; (1) Identify the existing elderly population, that is, among the users whose home locations are identified, users aged 60 or above are identified as elderly and their home locations are obtained; (2) Identify the potential elderly population, determine the screening conditions according to the predicted time period, and obtain the home locations of potential elderly people.

[0042] The fourth step is to predict the elderly population. The number of elderly people in the corresponding area this year is calculated by subtracting the number of elderly people who died from the base number of elderly people in the previous year and adding the net increase in the number of elderly people. (1) The study area is divided into regular grids as spatial analysis units. The elderly people whose residences fall within the grids are summarized to obtain the number of elderly people in the grids, which is used as the base number for elderly prediction. (2) The number of elderly people in each grid corresponding to the target time is predicted according to the elderly population prediction model.

[0043] The fifth step is to acquire data on public service facilities. This includes a collection of various types of public service facilities, or specific types of public service facilities such as tertiary hospitals and parks and green spaces. The relevant data includes the location and service capacity information of the public service facilities. The service capacity reflects the scale of services that the corresponding public facilities can provide to the public. The acquisition method is determined according to actual needs, including direct acquisition in cooperation with relevant competent authorities or self-collection from authoritative public service platforms.

[0044] Step 6: Time cost calculation; obtain the time required for people to travel from their permanent residence to public service facilities, mainly by assessing the path distance and mode of travel between the two; calculate the optimal distance between the two using road network data, and assess the required time information based on the capacity of common transportation tools; or obtain it directly from the route planning interface provided by an authoritative platform.

[0045] Step 7: Accessibility prediction of public service facilities for the elderly; using the Gaussian two-step moving search method, with the predicted number of elderly people as the demand scale, and based on the service scale of public service facilities, calculate the accessibility value of public service facilities for the elderly.

[0046] The specific projection of the elderly population is: the number of elderly people (P) in a local area in the next year. m+1 Equal to the number of elderly people P in year m m Subtract its number of deaths η m In addition to the net increase in the number of people μ m As shown in Formula 1:

[0047] P m+1 =P m -η m +μ m (1)

[0048] Number of deaths η m Based on the base population P of the elderly population in that year m The mortality rate was calculated, and considering that the mortality rate of the elderly was significantly higher than that of the general population, a correction factor α was used to adjust the mortality rate of the general population in that year. m The mortality rate of the elderly is obtained by making corrections; as shown in Formula 2:

[0049] η m =P m ×s m ×α (2)

[0050] Net new number of people μ m The calculation is based on the potential elderly population Δ minus the number of deaths in that year. Considering the difference between the mortality rate of the potential elderly population and the mortality rate of the total population, a correction factor β is used to adjust the mortality rate of the total population s in that year. mAdjustments are made to obtain the potential mortality rate among the elderly; as shown in Formula 3:

[0051] μ m =Δ-Δ×s m ×β (3)

[0052] Annual total population mortality rate s m The mortality rate is estimated based on historical mortality data obtained from the corresponding regional population bulletins or statistical yearbooks; taking a univariate linear regression model as an example, this illustrates how a mortality prediction model is constructed based on historical mortality rates to determine the corresponding annual mortality rate s. m Specifically, as shown in Formula 4:

[0053] s m = a×m+b (4)

[0054] Where a and b are the coefficients to be evaluated, which can be evaluated using the least squares principle based on the mortality rate over the years; while the mortality correction coefficients for the elderly and potential elderly in Formulas 2 and 3 are calculated based on the number of deaths in different age groups and the total number of deaths in the population published by authoritative departments, as shown in Formulas 5 and 6.

[0055]

[0056]

[0057] In the formula, POP0 is the total population, POP1 is the total number of deaths, p0 is the elderly population, p1 is the elderly population (i.e., the number of deaths in the age group of 60 years and older), p'0 is the potential elderly population, which can be determined by subtracting the prediction period from 60 years. For example, if the prediction period is 10 years, it refers to the population in the age group of 50 years and older but less than 60 years old, and p'1 is the number of deaths in the potential elderly population age group.

[0058] The calculation of accessibility to public service facilities specifically includes:

[0059] The first step is to calculate the supply-demand ratio. First, the location of the public service facility is obtained as the supply point j. Then, based on the facility's service standards, the time service threshold d0 of supply point j is determined. Next, all demand points (grid centroids i) within the time threshold d0 of all supply points are searched, and the supply-demand ratio R for each supply point j is calculated. j ,Right now

[0060]

[0061] In the formula, d ij S represents the travel time between demand point i and supply point j; j Let D be the service scale corresponding to point j; i G(d) represents the number of elderly people at demand point i; ij,d0) is a Gaussian equation function, and its calculation formula is:

[0062]

[0063] Step 2: Calculate accessibility; for each demand point i, search for all supply points j within the time threshold d0 of i, and calculate the supply-demand ratio R for each supply point falling within the spatial domain. j Weights are assigned using a Gaussian function, and these weighted supply-demand ratios R are then... j By summing the results, the reachability A of demand point i can be calculated. i The calculation formula is as follows:

[0064]

[0065] In the formula, R j Within the search area centered on demand point i, d ij ≤d0, the supply-demand ratio of public service facility j, A i The larger the value, the better the accessibility.

[0066] This invention uses a city in western my country as an example area, and takes the predicted accessibility of parks and green spaces for the elderly in 2028 (i.e., 10 years later) under a walking pattern as an example to verify the effectiveness of the method for improving the accessibility of public service facilities for the elderly. Mobile phone data was obtained from a mobile operator, consisting of trajectory data from August 1st to August 6th, 2018 (six days). After preprocessing the mobile phone data, the user's residential location was identified. Specifically, this invention considers locations where mobile phone users stayed for more than 3 hours between 0:00 and 6:00 as their residential locations, identifying a total of 255,000 user residential locations. Then, based on age attributes, 7,561 elderly users and 27,287 potential elderly users were selected.

[0067] When calculating the annual mortality rate for the entire population, a model was constructed based on the mortality rates of the past 20 years from the statistical yearbook and population bulletin of a certain city. The corresponding parameters a and b were assessed to be 0.071 and -137.44, respectively. Based on the national mortality population by age and sex from the "China Population and Employment Statistical Yearbook" (2012-2021), α and β for the 10 years were calculated and their average values ​​were taken, which were 5.15 and 0.80, respectively. Based on this, using the number of elderly people in 2018 as the projection base, the number of elderly people in 2028 was projected to be 29,848.

[0068] When acquiring data on public service facilities, the Baidu Maps interface was mainly used to obtain a total of 48 parks in a certain city. The park entrance or centroid was used as the supply point, and the park area was used as its service capacity.

[0069] When calculating time costs, the Baidu Maps API route planning interface is used to calculate the time cost between supply and demand points.

[0070] The results show that by 2028, the accessibility of parks and green spaces for the elderly under a walking pattern will significantly decrease compared to 2018. Under a 30-minute walking condition, considering only areas with available resources, the accessibility of parks and green spaces in 87% of spatial units will decrease by more than 70%. The relative changes are larger in the western and southeastern parts of the study area, followed by the central urban area. These predictions can effectively assist in guiding the planning of relevant public service facilities.

[0071] This invention uses mobile phone location data to identify users' residential locations and predicts the distribution of the elderly population based on the age structure information of mobile phone users and the natural population growth pattern of the region. On this basis, combined with the supply and spatial distribution of public facility resources, the Gaussian two-step move search method is used to predict the accessibility of public service facilities for the elderly in the city.

Claims

1. A method for predicting the accessibility of public service facilities for the elderly, characterized in that, Includes the following steps: The first step is to clean the mobile phone location data; This includes the following aspects: (1) removing duplicate values; (2) deleting null values; (3) deleting typical error values, which refers to records with abnormal latitude and longitude, anonymous user ID, and time. The second step is to identify the user's home location. This is done by identifying the areas where the user spends the most time at night. Specifically, the location where the user stays for the longest time is identified based on the length of time the user spends in each place. If the location identified by the user exceeds a certain duration threshold or is the location where the user spends the most time, then that location is identified as the user's home location. Otherwise, the location closest to the centroid of these locations is selected as the user's home location. The third step is to identify the elderly and potential elderly population; (1) Identify the existing elderly population, that is, among the users whose home locations are identified, users aged 60 or above are identified as elderly and their home locations are obtained; (2) Identify the potential elderly population, determine the screening conditions according to the predicted time period, and obtain the home locations of potential elderly people. The fourth step is to predict the elderly population. The number of elderly people in the corresponding area this year is calculated by subtracting the number of elderly people who died from the base number of elderly people in the previous year and adding the net increase in the elderly population. (1) The study area is divided into regular grids as spatial analysis units. The elderly people whose residences fall within the grids are summarized to obtain the number of elderly people in the grids, which is used as the base number for elderly prediction. (2) The number of elderly people in each grid corresponding to the target time is predicted according to the elderly population prediction model. The fifth step is to acquire data on public service facilities. This includes a collection of various types of public service facilities, or specific types of public service facilities such as tertiary hospitals and parks and green spaces. The relevant data includes the location and service capacity information of the public service facilities. The service capacity reflects the scale of services that the corresponding public facilities can provide to the public. The acquisition method is determined according to actual needs, including direct acquisition in cooperation with relevant competent authorities or self-collection from authoritative public service platforms. Step 6: Time cost calculation; obtain the time required for people to travel from their permanent residence to public service facilities, mainly by assessing the path distance and mode of travel between the two; calculate the optimal distance between the two using road network data, and assess the required time information based on the capacity of common transportation tools; or obtain it directly from the route planning interface provided by an authoritative platform. Step 7: Predicting the accessibility of public service facilities for the elderly; Using the Gaussian two-step move search method, with the predicted number of elderly people as the demand scale, the accessibility value of public service facilities for the elderly is calculated based on the service scale of public service facilities. The elderly population forecast specifically refers to the number of elderly people in a local area in the next year. equal to the elderly population Subtract its death toll In addition to the net increase in the number of people As shown in Formula 1: （1） Death toll Based on the base number of elderly people in that year The mortality rate was calculated based on the elderly mortality rate, and a correction factor was used because the mortality rate of the elderly was significantly higher than that of the general population. The mortality rate of the entire population in that year The mortality rate of the elderly is obtained by making corrections; as shown in Formula 2: （2） Net new number of people Based on the potential elderly population of that year The calculation is performed by subtracting the number of deaths, taking into account the difference between the potential mortality rate of the elderly population and the total population mortality rate; therefore, a correction factor is used. The mortality rate of the entire population in that year Adjustments are made to obtain the potential mortality rate among the elderly; Specifically, as shown in Formula 3: （3） Annual total population mortality rate The mortality rate was estimated based on historical mortality rates obtained from the corresponding regional population bulletins or statistical yearbooks. Taking a univariate linear regression model as an example, this explains how to construct a mortality prediction model based on historical mortality rates to determine the mortality rate for a given year. ; Specifically, as shown in Formula 4: （4） in and The coefficient to be evaluated can be assessed using the least squares principle based on the mortality rate over the years; while the mortality correction coefficients for the elderly and potential elderly in Formulas 2 and 3 are calculated based on the number of deaths in different age groups and the total number of deaths in the population published by authoritative departments, as shown in Formulas 5 and 6. （5） （6） In the formula, Total population Total number of deaths For the elderly population, This refers to deaths among the elderly, specifically those aged 60 and above. The term "potential elderly population" is specifically determined by subtracting the projection period from 60 years old. For example, if the projection period is 10 years, it refers to the population aged 50 to 60. This refers to the number of deaths in the potential elderly population age group.

2. The method for predicting the accessibility of public service facilities for the elderly according to claim 1, characterized in that, The calculation of accessibility to public service facilities specifically includes: The first step is to calculate the supply-demand ratio; this begins by identifying the locations of public service facilities as the supply points. Then, based on the facility service standards, the supply points are determined. Time service threshold And search for time thresholds for all supply points. The demand points within the range are the centroids of the grid. Calculate each supply point supply and demand ratio ,Right now （7） In the formula, For demand points and supply points Travel time between them; for The service scale corresponding to the point; For demand points The number of elderly people; The function is a Gaussian equation function, and its calculation formula is: （8） Step 2: Calculate reachability; for each demand point Search all in Time threshold Supply points within the scope For the supply-demand ratio of each supply point within the spatial domain Weights are assigned using a Gaussian function, and these weighted supply-demand ratios are then... The demand points are calculated by summing the results. Accessibility The calculation formula is as follows: （9） In the formula, Based on demand points Within the central search area, Public service facilities The supply-demand ratio The larger the value, the better the accessibility.

Images