Data-driven Optimization Research on Ageing-Responsive Design of Building Space in Urban Elderly Institutions

Kano model is widely used in the field of user needs research, Kano model through the positive and negative set of questions, that is, to study in the “meet the needs” and “do not meet the needs” two kinds of situations, the crowd’s satisfaction level of the project. The Kano model classifies requirements into five categories: essential, desired, charismatic, undifferentiated, and inverse requirements [37]. The Kano model is analyzed as shown in Figure 1.

Figure 1.

Carnot model analysis

M-Must-have needs: needs that the user believes must be met. Satisfying this need does not increase user satisfaction, while not satisfying it will make users feel extremely inconvenient or even unbearable.

O-expected needs: refers to the needs that the user wants to be satisfied. If the need is satisfied, it can make the user feel satisfied. If the need is not satisfied, it will make them dissatisfied, but not to the point of intolerability.

A - Charming needs: refers to the needs that the user does not expect and will be surprised. If the need is met, it will exceed their expectations, and if it is not met, the user will not be dissatisfied.

I-Non-differentiated demand: refers to the demand that users do not pay attention to, the satisfaction of this demand does not affect the user experience.

R-Reverse Requirement: It refers to the requirement that is contrary to the user’s actual intention, and if the requirement is satisfied, the user will be dissatisfied.

Research and analysis of demand is often the first step in good design, whether it is product design, architectural design, intelligent design. Most (80-90%) of the needs analysis studies of architectural space design analyze and summarize the needs based on Maslow’s hierarchy of needs theory, and 30-40% of them combine cluster analysis, survey documents and AHP hierarchical analysis to quantitatively analyze the needs [38]. Analyzing the necessity and importance of needs through Maslow’s hierarchy of needs theory and Kano’s research questionnaire are two important research methods in the field of industrial design and Internet product design.

In this paper, the study of intelligent ageing design in towns and cities requires a scientific research method to accurately determine the practical needs of the user group, and the Kano questionnaire’s positive and negative bidirectional question settings can, to a certain extent, avoid the influence of the degree of perfection of the facilities in the respondent’s living environment on the answer results, thus effectively improving the reliability and validity of the questionnaire. Therefore, this paper chooses the Kano model demand analysis method, through a set of positive and negative questions, to analyze the necessity and importance of each demand. Unlike Maslow’s hierarchy of needs theory, which is a rigid stepwise ascending structure of needs, ERG theory also suggests that people may have more than one type of need at the same time, and that the needs of each level influence each other.ERG theory suggests that if the fulfillment of the needs of the higher level is restricted, then people’s desire for the needs of the lower level will become more intense.

According to the logic of analyzing and comparing indicator acquisition, indicator screening, weight calculation, and indicator classification and processing, this study chooses scientific and reasonable research methods.

The research object of this survey is the demand for ageing-friendly design of architectural space in urban elderly institutions, and its intelligent ageing-friendly refers to the intelligent design aiming at suitable use by the elderly, and its intelligent design measures include intelligent ageing-friendly facilities provided in the scope of public space of urban settlements in the natural physical space, and intelligent ageing-at-home services provided in the virtual information and social space.

The research period is limited to 2021 to 2025. The study focuses on the daily living needs of the self-care and assisted elderly, and the disabled and mentally retarded elderly are not considered as the main subject of this study on intelligent ageing design.

The respondents are people involved in ageing in place activities. This study starts from the perspective of aging in place, and the main respondents of this demand survey are 45-59 years old prospective elderly group, 60-75 years old “young elderly” group, community operators and medical caregivers.

The research on virtual information spaces is limited to the application layer services of the smart settlement management platform. The measurement of demand research is limited to the intelligent facilities for the elderly provided in the natural physical space within the public space of urban settlements, and the intelligent home care services provided in the virtual information and social space.

Based on the environmental monitoring and control system to complete the design of green environment at home, the green environment is in line with the green standards of the indoor environmental state, based on the green environmental standards designed environmental monitoring and control system, environmental monitoring and control system shown in Figure 3. Combined with the figure it can be seen that the system includes sensor module, data processing module, environmental regulation module and other key parts: (1) the system uses multiple sensors to collect real-time indoor temperature and humidity, fine particulate matter, formaldehyde, VOCs information to determine the current indoor environmental level, the original environmental data collected is sent to the environmental monitoring center, when various types of monitoring indicators exceed the normal standard thresholds to the user sends exceeding the standard alarm information for the user to improve the indoor environment to provide a basis for the user. Users should improve the indoor environment to provide a baseline, while recording data for later environmental analysis backup. (2) Intelligent control module: on the one hand, this module of the cell phone APP set up a user center, personal preferences, environmental data mining and other functions, through the calculation of human comfort index to form an automated control of environmental regulation equipment. On the other hand, the data processing module data as the basis for artificial regulation and control, intelligent regulation and control module including new wind system, intelligent air conditioning, air purifier, intelligent curtains and other intelligent equipment, through the cell phone APP interface, according to the ventilation, light, air pollutants monitoring results of the flexible use of intelligent regulation and control equipment will be adjusted to the reasonable value of the indicators, to create a green environment at home. (3) The sensor module transmits the information to the data processing module for noise reduction, supplementation, deletion, format conversion, classification and other processing, in order to make more efficient and accurate environmental regulation judgment.

Figure 3.

Environmental monitoring and control system

In order to monitor accidental falls occurring in the home activities of the elderly in real time, a wearable fall detection device is designed, and the architecture of the human wearable health monitoring system is shown in Fig. 4. The system mainly contains acceleration sensors and inclination sensors to collect the behavioral information of the monitored user, use the fall detection algorithm to judge the fall behavior, and send the fall alarm information to the guardian when the fall detection criteria are met [39].The GPS module collects the user’s current location information and sends it together with the alarm information to the client, which is convenient for the caregiver to take timely rescue measures.

Figure 4.

Human wearable health monitoring system architecture

The fall detection algorithm operates based on three key variables: regional acceleration intensity, linear acceleration signal intensity, and human body inclination angle, and the three variables are defined to be represented by μ, λ, and γ, and the idea of fall detection is as follows: use λ behavioral feature quantities for the initial screening of the fall behavior, and then use μ and γ behavioral feature quantities for the precise detection.

Step1: Define the suspected fall threshold w₁, precise fall detection threshold w₂, and tilt angle threshold w₃.

Step2: Use wearable fall detection equipment to dynamically collect human behavior data, if λ the characteristic quantity is greater than the threshold will be judged as a possible fall behavior, into the next step of the operation, and vice versa to give up the operation of the data. Among them, the linear acceleration signal strength calculation method is shown in Equation (1): (1) λ=x(i)2+y(i)2+z(i)2

In the formula, the acceleration values in the X, Y, and Z directions collected by the i st acceleration sensor are expressed as x(i), y(i), and z(i), respectively.

Step3: Judge the characteristic quantity μ and the precise fall detection threshold w₂, μ is calculated as shown in Eq. (2): (2) μ=1W∑i=1W−1(x(i)+y(i)+z(i))

Where the total amount of data collected is described by W. When the amount of features μ is larger continue to execute the next step, and vice versa go back to Step1 until the conditions are met and then go to the next step.

Step4: Compare the threshold w₃ with the body inclination γ and calculate the expression as follows: (3) λ=arcsin=y(i)x(i)2+y(i)2+z(i)2

If the angle of inclination of the human body in the direction of the trunk and the center of gravity is below 60°, the human body is defined as “upright”. If the angle of inclination is close to 90°, the body is considered to be in a “horizontal” state, which is also a sign of a fall. Therefore, when the angle of inclination of the human body is 60° or more, it is considered that the human body is characterized by a fall. When the angle of inclination of the human body γ is larger, continue to the next step, and vice versa, go back to Step1.

Step5: Get the behavioral feature quantity that meets the fall threshold condition, and at this moment, start the warning mechanism to send the fall alert to the client.

After the fall detection device starts working, the system is initialized and the LED module displays the current time and the state of the old man. The threshold is set to 0.9g, if the collected a is less than 0.9g, then the old man is judged to be in a weightlessness-like process. Next determine whether a collision occurs (a>2g) If it is determined that no collision occurs, then send a warning III (the elderly may be at risk of falling). If the data return to normal within 10s after the collision, it is judged that the old man gets up on his own or is rescued, then it sends warning II. If there is almost no fluctuation of a value and obvious change of θ value in 10s, it can be recognized that the old man has fallen and cannot get up, then send warning I. The above three levels of different degrees of warning are sent to the guardian’s cell phone through the positioning module and communication module to locate and alarm information, warning will trigger the voice broadcast module alarm, to ensure that the elderly get timely assistance. Within 15s of the voice alarm, if the old man retains consciousness or is rescued, he can manually contact the alarm.

The experimental results are shown in Table 7. The experiment proves that this system has a high success rate of 100% for walking, sitting down, standing up, going downstairs, class weightlessness process, and standing up after falling detection.The number of false alarms for elevator movement is lower, and the success rate is 95%.The test results show that the system can distinguish between elderly people in daily activities or in fall status, with feasibility and practicality.