Next Article in Journal
Green Growth or Grey Gains: Rethinking Financial Development and Foreign Direct Investment Impacts on Ecological Sustainability in Sub-Saharan Africa
Previous Article in Journal
From Isotopic Evidence to Economic Valuation: A “Water–Carbon–Economy” Nexus Framework for Climate-Resilient Urban Forestry in Southwestern China
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 18
Issue 6
10.3390/su18062781
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Fuzzy Logic-Based Model for Measuring Perception of Urban Spaces During Walking Experience

by
Esra Baran
1,*,
Mehtap Özbayraktar
1 and
Serhat Yılmaz
2
1
Faculty of Architecture and Design, Kocaeli University, Kocaeli 41001, Türkiye
2
Faculty of Engineering, Kocaeli University, Kocaeli 41001, Türkiye
*
Author to whom correspondence should be addressed.
Sustainability 2026, 18(6), 2781; https://doi.org/10.3390/su18062781
Submission received: 11 January 2026 / Revised: 7 February 2026 / Accepted: 25 February 2026 / Published: 12 March 2026
(This article belongs to the Section Sustainable Transportation)
Browse Figures
Versions Notes

Abstract

The perception of urban spaces during walking experiences is influenced by an individual’s social conditions, personal characteristics, and the spatial features of the route being walked. This study proposes a methodological approach that focuses on the individual and spatial factors affecting perception of urban spaces during walking experiences, incorporating subjective data that influence perception. The aim of this research is to measure both qualitative and quantitative data that affect the perception of urban spaces during the walking experience in a comprehensive and systematic manner, using fuzzy logic. Within the scope of the research, a field survey was conducted with 25 university students in Değirmendere Yalı Neighbourhood (Kocaeli) and Bahçelievler Neighbourhood (Yalova), and the data obtained from the survey were analysed using the fuzzy logic method to test the applicability and validity of the method. The results of the analysis performed using the fuzzy logic model showed that individual factors affecting the perception of urban space during the walking experience, the walkability characteristics of the space, and the perceptual characteristics of the space can be comprehensively evaluated using the fuzzy logic method, and that the perception of urban spaces can be measured quantitatively.

1. Introduction

The Perception of urban spaces occurs as a result of the interaction between our body and the environment, and our movements [1]. During the act of walking, individuals discover the sensory, spatial, and social layers of that environment [2,3,4]. While moving in urban space, our mind establishes connections between space and our body and analyses these connections. This dynamic state that urban space acquires with the concept of time during the act of walking enables the individual to discover urban space [3].
The experience of walking and the perception of urban spaces are multidimensional phenomena that encompass spatial, social, and individual parameters [5,6,7]. This complexity creates difficulties in systematically addressing the influence of individual and subjective differences on urban space perception, as well as in jointly evaluating qualitative and quantitative data shaped during walking experiences. In the literature, existing methods for measuring urban space perception during the walking experience involve various limitations, such as detachment from real walking experiences, high costs, the requirement to work with large sample sizes, and difficulties in jointly addressing qualitative and quantitative data. Accordingly, this study seeks to measure urban space perception while walking through fuzzy logic, considering both subjective user-related factors and spatial factors that define the identity of the built environment. The aim is to render qualitative and quantitative data on urban space perception quantitatively measurable within a rule-based fuzzy logic system, thereby contributing to the existing literature. To test the applicability of the proposed method, a field survey was conducted with 25 architecture students in Değirmendere Yalı Neighbourhood (Gölcük, Kocaeli, Turkey) and Bahçelievler Neighbourhood (Centre, Yalova, Turkey), and the survey data were analysed using the fuzzy logic method.
Determining an individual’s perception of the urban environment will contribute to the creation of walkable environments and improve the quality of the built environment. Can individual factors affecting the perception of urban space during the walking experience, the walkability characteristics of the space, and the contribution of the perceptual characteristics of the space to the perception of urban space be measured quantitatively using fuzzy logic? It is anticipated that this article, which seeks to answer this research question, will contribute to the literature in terms of the systematic handling of qualitative data in research conducted at the intersection of environmental psychology and architecture, and will be useful for user-centred urban design approaches.
To determine the factors affecting the perception of urban space during the walking experience, the theoretical framework of the study includes a literature review on urban space perception, the relationship between pedestrian behaviour and urban space perception, the relationship between walkability and urban space perception, the measurement of walkability and urban space perception, and the fuzzy logic method used in the research. In the second part of the article, the methodology of the study and the proposed fuzzy logic model are presented. The third section presents the findings from the field research and the results of the analysis conducted using the fuzzy logic model. The fourth and final section of the study presents the research results.

1.1. Perception of Urban Spaces

Perception is defined as a process that “includes recognizing (being aware of), organizing (gathering and storing), and interpreting (binding to knowledge) objects” in relation to human cognitive processes [8]. Architectural and urban spaces are perceived through senses triggered by stimuli such as light, form, size, sound, texture, colour, smell, and reflection. This information is interpreted through cultural and personal images, forming spatial knowledge by comparison with prior experience [9]. Thus, environmental conditions, social structure, and individual physical and socio-psychological characteristics influence spatial perception [10,11].
Humans are constantly interacting with their physical environment. This interaction occurs through the perception of the surrounding environment by means of the sensory organs. Urban space is collectively experienced by its users through physical and social activities. These experiences transform urban spaces from static entities into dynamic environments that evolve through user experience. Urban spaces are therefore perceived as a result of the experiences users have within them and are mentally constructed by the individual. According to Lang (1987), perception is not a passive but an active process that emerges through a reciprocal interaction between the perceiver and the perceived [12]. In this context, the perception of the body and the image of the world become a continuous experience through movement [1,13]. Consequently, the perception of urban space becomes not merely a visual impression but a dynamic process that is continuously reproduced through the body, movement, and experience. In this regard, walking behaviour serves as an essential means for perceiving urban space.

1.2. The Relationship Between Pedestrian Behaviour and Urban Space Perception

The most fundamental form of behaviour observed in urban space is the act of walking, that is, pedestrian behaviour [14]. Individuals typically walk in urban spaces for three purposes. Walking to get from one place to another is the most basic purpose of walking. Other purposes of walking include walking for sports and exercise, motivated by health considerations, and walking for recreation, i.e., for pleasure and leisure [15,16,17]. In line with these purposes, the act of walking enables the individual to establish cognitive and behavioural relationships with the urban environment.
During walking behaviour, individuals evaluate the urban environment not only through sensory perception but also in relation to individual memory, meaning production, and emotional contexts [18,19]. People establish an emotional relationship with the places they live in, attribute meaning to these places, and form their identities within them [20]. The relationship individuals establish with these places through their experiences is defined as a “sense of place.” The cultural structure, physical structure, social structure, and individual differences in society influence the relationship that individuals establish with a place [21]. In order for an individual to develop a “sense of place” towards the place they live in, they must first explore this place on foot, get to know it, and, through a series of experiences, make sense of it and establish a sense of belonging [19,22]. When evaluated within this theoretical framework, walking behaviour enables individuals to perceive and internalise space.
While moving through urban space, an individual makes behavioural decisions, such as choosing the route to take, orienting towards landmarks, creating shortcuts, and planning movement [23]. The factors influencing an individual’s pedestrian behaviour in urban space can generally be classified under four main categories: the spatial characteristics of the urban environment (perceptual, behavioural, physical features, the city, historical values, etc.), individual characteristics (physiological, demographic, biological), group characteristics (socio-economic, cultural), and regional characteristics (topography, flora-fauna, climate, cultural factors) [24]. The spatial differentiation of urban space, formed through its formal, volumetric, colour, material, and lighting characteristics, affects pedestrian behaviour [25]. Pedestrians navigate in urban space by using information about distance, direction, and geometry. However, an essential factor determining pedestrian behaviour in urban space is the walkability of the pedestrian route. While walking in urban spaces, individuals tend to prefer routes that are both attractive and comfortable for them [26,27,28,29].

1.3. The Relationship Between Walkability and Perception in Urban Spaces

One of the important factors that shape the perception of the urban environment during walking is the context of the pedestrian path, namely, its walkability [30,31]. “Walkability is the extent to which the built environment supports and encourages walking by providing for pedestrian comfort and safety, connecting people with varied destinations within a reasonable amount of time and effort, and offering visual interest in journeys throughout the network” [31]. Researchers have defined walkability through various parameters, including route connections, safety, physical quality, and mixed land use [31]. As asserted by Jin and Kim (2024), the importance of roads that are comfortable, uninterrupted, well-connected, accessible, and safe from traffic and crime for enhancing walkability is paramount [32]. Similarly, in their review of the literature on walkability, Fonseca et al. (2022) identified land use diversity, accessibility, street network connectivity, pedestrian facilities, safety, and street scene design as the salient issues [29].
Walkability is a multidimensional phenomenon encompassing perception alongside the physical and spatial characteristics of place [7]. Investigating the link between street scene quality and walking behaviour, Ewing et al. (2006) proposed a conceptual framework integrating objective and subjective data, identifying perceptual attributes of walkability such as legibility, imaginability, enclosure, human scale, permeability, connectivity, diversity, coherence, and order [33]. Similarly, Fonseca et al. (2022) emphasised six key perceptual qualities influencing walkability at the street scale: aesthetics, human scale, enclosure, complexity, permeability, and imageability [29]. As a result, when studies on walkability in the literature are examined (Supplementary Material SA), it becomes clear that walkability is a multifaceted concept that shapes an individual’s interaction with the urban environment and their walking experience, encompassing its physical, functional, and perceptual dimensions.

1.4. Measurement of Walkability and Perception in Urban Spaces

Various techniques are employed to measure users’ perception of the urban environment. Since Lynch (1960), most studies have relied on surveys, with validity enhanced through the use of detailed questionnaires or larger samples [4]. Ramirez et al. (2021) classify perception studies into four groups: qualitative methods, discrete choice models, machine learning, and sequential methodologies [34]. Qualitative approaches based on surveys and interviews limit comparability and generalisation, while discrete choice models segment computer-generated images to assess perception, increasing scalability but also research costs [35,36,37,38]. Machine learning enables the analysis of complex models with many variables using smaller samples [37,39,40,41,42], and sequential methodologies integrate machine learning with econometrics [34]. New technologies, such as wearable physiological sensors (EEG, MRI), eye-tracking, and AI-based applications (VR, AR), are also applied [36,43]. However, these methods primarily address visible elements and remain insufficient in integrating multi-sensory perception or subjective individual and social factors. Comparative information on existing methods for measuring urban space perception is presented in Table 1.
Studies on walkability largely examine walkability and its determinants [31,32,44], with the relationship to perception often discussed under “environmental aesthetics” [5,7,19,33,45,46]. However, existing research remains limited in comprehensively measuring physical parameters, subjective qualities, and socio-individual factors shaping spatial perception during walking. While studies emphasise parameters that enhance walkability, the integrated assessment of factors with different qualities and scales remains challenging [47]. Pedestrian perceptions are generally investigated through qualitative data obtained from surveys, interviews, cognitive mapping, and group studies conducted with users. In contrast, objective elements related to walkability, such as building height, street dimensions, and land use, are measured through GIS or space syntax [29]. Therefore, to systematically evaluate urban space perception during walking, measurement models that incorporate both qualitative and quantitative data are required [48,49]. In addition, perception-related data are often composed of linguistic information. Since linguistic data possess subjective characteristics, problems arise in their comparison and statistical evaluation. At this stage, the fuzzy logic method is utilised to quantify and measure linguistic and subjective data. Based on this premise, the fuzzy logic method is employed in the measurement model proposed in this research due to the advantages it offers, such as its ability to evaluate qualitative and quantitative data together, its flexible structure that allows different parameters to be easily integrated into the system, and its capacity to operate effectively with small data sets.

1.5. Fuzzy Logic Method

Fuzzy logic, introduced by L. A. Zadeh in 1965, differs from classical Aristotelian logic in that it employs linguistic expressions and mimics human decision-making [50]. Unlike binary values of “0/1” or “true/false”, it incorporates intermediate values between [0, 1] [50]. This method enables the representation of human experience and common sense in machine-processable form, facilitating the modelling of perception, inference, and behaviour [51].
The fuzzy logic approach consists of three stages: fuzzification, fuzzy inference, and defuzzification. First, numerical data defined in Aristotelian logic are fuzzified by determining membership functions to form fuzzy sets, which Zadeh (1965) defined as sets with membership degrees in [0, 1] [50]. Second, relationships between inputs and outputs are expressed through rules, constituting the fuzzy decision-making process. Finally, results derived from these rules are converted into a single numerical value via defuzzification [50,52] (as illustrated in Figure 1). Thus, fuzzy logic quantifies verbal and linguistic data beyond Aristotelian logic, enabling inferences comparable to human reasoning.
In architecture, the fuzzy logic method remains relatively underexplored, though it is increasingly applied in studies on walkability and pedestrian behaviour. It is used to quantify perception data from [53,54,55,56,57,58] and to integrate qualitative and quantitative data at multiple scales [59,60,61]. For instance, Soares Müller and Ruiz-Padillo (2025) created a micro-scale accessibility index to evaluate pedestrian infrastructure for people with disabilities, digitising linguistic survey data through fuzzy logic for multi-criteria analysis [58]. Similarly, Nyagah (2015) developed a holistic walkability index incorporating pedestrian safety and accident risk, using fuzzy logic to quantify participants’ linguistic assessments of their walking environment [56]. In summary, studies situated at the intersection of walkability and perception that employ the fuzzy logic method integrate both qualitative and quantitative dimensions within a unified analytical framework. This provides a holistic model for measuring phenomena such as walkability and urban perception.

2. Materials and Methods

In this section of the study, which aims to present a model proposal that measures the perception of urban space formed during the walking experience through the fuzzy logic method, the process of constructing the proposed fuzzy logic model and the methodology of the field study conducted to determine the model’s validity are explained. To identify the factors that affect the perception of urban space during the walking experience, a literature review was conducted within the scope of the research. The factors addressed in the study were then presented as a conceptual framework. In the second stage of the research, the sample was determined; in the third stage, the questionnaire questions were prepared. In the fourth stage, the study areas in which the model would be tested were identified. In the fifth stage, the process of constructing the fuzzy logic model was described. In the final stage of the research, the proposed fuzzy logic model was applied in the selected areas, and the findings obtained were presented (as illustrated in Figure 2).

2.1. Factors Affecting Perception of Urban Spaces During the Walking Experience

As a result of the literature review on perception of urban spaces, pedestrian behaviour in urban spaces, and walkability, factors affecting perception of urban spaces during the walking experience were identified. It was observed that the factors influencing the perception of urban spaces during the walking experience could be grouped into three categories: social factors, individual factors, and spatial factors (as illustrated in Figure 3).
The socio-economic and socio-cultural structure of society that shapes an individual’s perception of space, along with the place of walking within that society’s cultural fabric, is addressed under the heading of social factors [2,3,4,6,10,11,14,21]. Social factors have been excluded from the scope of this research due to the difficulties that would arise from addressing them in different societies and at different socio-economic and cultural levels.
The individual’s socio-psychological structure and physical characteristics, which shape individual differences in perception, are considered individual factors [5,6,11,14,16,18,19,21,22,62]. Individual factors are addressed in the study through participants’ perspectives on walking and their sense of belonging (place sense) towards the place they walk. The perspective on walking aims to determine the individual’s relationship with walking by identifying the purposes and frequency of walking. The individual’s sense of sight and physical characteristics related to walking behaviour were excluded from the scope of the research because they constitute subjects of medical science.
Spatial factors consist of environmental parameters that affect perception of urban spaces, as well as factors that affect the perception of urban space during the walking experience due to walkability. Physical parameters affecting the walkability of urban space include the quality of pedestrian pathway, accessibility, pedestrian density, vehicle density, adaptability to climatic conditions, lighting, noise, slope, etc., which are parameters that make the urban environment suitable for walking [31,32,44,46,47,63]. Within the scope of this research, the focus is on the parameters of the quality of pedestrian pathway, accessibility, pedestrian density, motor vehicle traffic density, and the presence of structural elements such as canopies and coverings. Features affecting the physical comfort of the pedestrian pathway, such as sidewalk width, height, and material, are grouped under a single heading in the study: quality of pedestrian pathway. The accessibility of the pedestrian path for everyone and the absence of obstacles are addressed through the accessibility parameter. The presence of pedestrian density on the walking path and low motor vehicle traffic density is evaluated in the study as spatial parameters affecting the psychological comfort of users. The presence of structural elements such as eaves and coverings that protect individuals from climatic parameters such as rain, wind, and sun during walking behaviour is also considered. Spatial factors affecting walkability, which are subjects of building physics, such as lighting and noise, are excluded from the scope of this study.
Perceptual characteristics influencing the experience of urban space during walking include imageability, legibility, enclosure, human scale, permeability, diversity, coherence, and architectural identity [5,7,19,29,33,45,46,63]. Imageability refers to the memorability and symbolic richness of a space, whereas legibility concerns its clarity and ease of orientation. Enclosure refers to spatial definition through surrounding surfaces, while human scale is assessed by the proportionality of elements to human dimensions. Permeability relates to the degree of connectivity, diversity to visual variety, and coherence to the integration of natural and built features. The presence of a distinctive architectural identity further strengthens memorability and perceptual selectivity.

2.2. Determining the Sample

The sample of the study consists of 25 architecture students aged between 18 and 22 (Age and gender distribution of the participants as shown in Table 2). The reason for selecting architecture faculty students for the study is based on the assumption that these students’ awareness of stimuli in urban spaces can be more easily demonstrated to other participants through expert knowledge. In this way, perception-related variables will be able to be evaluated in a conscious and consistent manner. In addition, the fact that the participants in the sample have similar characteristics in terms of age, educational status, etc., is important for descriptive statistics within the scope of the research. Furthermore, a preliminary questionnaire form was used to ensure that none of the students participating in the study had any vision impairment, such as night blindness or colour blindness, or any physical disability or health problem that would make walking difficult. The main reason for this is to control variables that are not included in the model by keeping the sample homogeneous during the process of testing the fuzzy logic–based methodological approach, which constitutes the primary focus of the study.” Prior to the survey, all participants were asked to complete a written “Participant Information and Consent Form” to ensure voluntary, anonymous participation. Additionally, an ethics committee document was obtained from the relevant institute (Details are presented in the manuscript (with author details)).
To determine the sample size, Neuman’s (2011) approach was adopted, which states that in qualitative research aiming to answer questions such as why, how, and in what way, a fixed sampling rate is unnecessary; instead, a limited number of qualified participants is sufficient [64]. Such studies employ purposive sampling, designed to obtain in-depth information about the subject of research (Maxwell, 1996, cited in [65]. Accordingly, a small sample size allows for a detailed examination [66]. The fuzzy logic method used here, unlike other statistical analyses, enables the assessment of qualitative data with small samples. Based on these principles, the sample group was determined as 25 participants. The aim of the field study was to test the applicability of the model developed to measure urban space perception during the walking experience, rather than to achieve statistical generalisation. Accordingly, the sample size was deliberately kept limited in order to assess the functionality of the model rather than to enable statistical generalisation.

2.3. Survey Design

Questions were prepared to determine how walkable participants found the area, the participants’ perception of the area, their perspective on walking, and the current perceptual and walkability characteristics of the places they walked in. Referring to similar studies in the field [32,67,68], the survey form asked participants to rate the questions on a scale ranging from 0 to 100. This approach allows participants to express their opinions in a more precise and continuous manner rather than limited categories. Another reason for preferring this scale is that it provides a structure suitable for defining the membership functions used in the fuzzy logic model. While this scale offers an intuitive evaluation range for participants, it is also compatible with the mathematical structure of the model. A value of 0 on the scale represents ‘disagree/low’, while a value of 100 represents ‘agree/high’. This facilitates the separation of the obtained data into membership functions in fuzzy logic evaluation.
In preparing the survey questions, attention was generally paid to identifying elements that affect the perception of urban spaces during the walking experience outlined in Figure 3. Question 1 of the survey form sought to determine how often and for what purpose (transportation, sports and exercise, pleasure, and leisure) participants walked, as well as their perspectives on walking. The answers to this question were prepared specifically for this question, as linguistic expressions such as “rarely,” “moderately,” and “very often.” The second question asked participants how walkable they found the route they walked. In the third question, an attempt was made to determine how participants perceived the route using semantic differential scales, which involved evaluating the route through the given adjective pairs (ordinary–unique, uninteresting–interesting, irregular–regular, undefined boundaries–defined boundaries, cramped–spacious, incompatible–compatible). The semantic differential scale used in this question is one of the methods employed in human–environment research to evaluate users’ responses to the environment [69]. By employing adjective pairs with opposite meanings at both ends, participants were asked to indicate a value within the “0–100” range, where the negatively connoted adjective corresponded to “0” and the positively connoted adjective to “100”. The adjective pairs preferred here were determined in relation to the concepts specified as the perceptual characteristics of space in the conceptual framework (as illustrated in Figure 3), making use of the adjective pairs employed in the studies of [70,71,72,73,74,75,76] (Supplementary Material SB). The fourth question in the questionnaire asked participants to rate the route they walked on a scale of “0–100” based on the specified parameters. In this question, participants evaluated the route they walked based on the perceptual and walkability characteristics of the space specified in the conceptual framework, and an attempt was made to determine the participants’ level of belonging to this area. To briefly summarize the survey questions, Questions 1 and 4 in the questionnaire were asked to generate the input values for the fuzzy logic method used in the research. Questions 2 and 3, on the other hand, were asked to determine how the routes walked were perceived by the participants and to assess their walkability.
The relationship between the questionnaire items and the factor groups was addressed through the conceptual framework of the research (Figure 3). The questions included under Question 4 of the questionnaire, which assess the physical quality of the route, accessibility, pedestrian density, motor vehicle traffic intensity, and the level of elements such as canopies that may provide protection against seasonal conditions, are related to the walkability factors group (Questions 4.1, 4.2, 4.3, 4.4, and 4.5). Perceptibility factors were addressed in the questionnaire design through the levels of imageability, legibility, enclosure, human scale, permeability, diversity, coherence, and architectural identity included under Question 4 (Questions 4.6, 4.7, 4.8, 4.9, 4.10, 4.11, 4.12, and 4.13). Questions 1 and 4.14 in the questionnaire are related to the individual factors group defined in the conceptual framework and were designed to measure participants’ perspectives on walking and their sense of place attachment. A schematic representation illustrating the relationship between the questionnaire items and the factor groups, as well as which parameter is measured by each question in the questionnaire design process, is presented in Figure 4.

2.4. The Case Study Areas

Within the scope of the research, field studies were conducted in two different settlements in order to test the applicability and validity of the proposed fuzzy logic model. As the aim of the field study was to examine whether the proposed fuzzy logic model produces consistent and valid results, two settlements with similar urban identity characteristics were selected. Accordingly, Değirmendere Yalı Neighbourhood in the Gölcük district of Kocaeli Province and Bahçelievler Neighbourhood in the Central district of Yalova Province were determined as the study areas. Both settlements are located along the coast of the Marmara Sea and are characterised as former coastal towns that were significantly affected by the earthquake in 1999, resulting in the loss of their coastal town identity. In both settlements, the existing building stock predominantly consists of residential and mixed residential–commercial structures (as shown in Table 3).
Değirmendere Yalı Neighbourhood (Kocaeli) is a settlement with visual diversity, featuring old wooden houses, wooden sculptures, and fruit trees, and has an area of 0.31 km2. Değirmendere Yalı Neighbourhood has a population of 4985, and its housing stock consists of 1–2 or 3–4-story reinforced concrete buildings [77]. The layout of the streets and avenues in the neighbourhood has an organic structure due to the sloping hills located in the south of the neighbourhood (as illustrated in Figure 5).
Bahçelievler Neighbourhood (Yalova), on the other hand, is a place with visual diversity due to its landscape elements and sculptures located on the shore, and it has a grid-like settlement plan. Bahçelievler Neighbourhood (Yalova) has an area of 1.31 km2, and the section in the western part of the neighbourhood, which includes greenhouse areas and agricultural lands, was excluded from the scope of the study, focusing instead on the location where the residential fabric of the neighbourhood is located. Thus, two study areas were created that are similar in terms of both population density and urban identity characteristics. In this neighbourhood, which is one of the city’s old settlements, the buildings are generally 4-, 5-, and 6-story reinforced concrete structures [78] (as illustrated in Figure 6).
Within the scope of the field study, all participants walked the same predetermined route. This approach was adopted based on the anticipation that allowing participants to experience different routes could cause difficulties in the measurability of the model and in the generalisation of the research findings. In both neighbourhoods, the selected routes are located along the coastal zone and are frequently used by local residents. Both routes are closed to vehicular traffic and consist of linear paths of approximately 1 km in length. The gradient of both routes ranges between 1% and 1.5. During the walking experience along both routes, landscape elements, sea views, sculptures, and urban furniture are present. These elements contribute to differences in visual diversity between the routes (as illustrated in Figure 5, Figure 6 and Figure 7). Detailed photographs and schematic sections of the walked routes are presented in Supplementary Material SD.

2.5. Development of the Fuzzy Logic Model

This section presents the fuzzy logic model, including its inputs, outputs, rules, and membership functions. MATLAB R2024 b Fuzzy Logic Tool and Simulink toolbox were used to implement the model. These programs were chosen because the MATLAB R2024 b Fuzzy Logic Toolbox allows for the definition of membership functions, the creation of a rule base, and the precise execution of defuzzification processes. MATLAB R2024 b Simulink was chosen to create a model that obtains a single output for each urban perception value modelled under different groups (individual factors, walkability factors, and perceptual factors). For the creation of the model, inputs, outputs, and membership functions were first defined, followed by fuzzification, rule base creation, fuzzy inference, and defuzzification. The model was established using the Mamdani-type fuzzy logic method. The Operational system of the fuzzy logic model proposed in the study is illustrated in Figure 8.
The parameters influencing urban space perception during walking, identified through a literature review and conceptual framework, form the basis of the study’s inputs. Since modelling all 17 inputs individually created excessive rules, they were grouped into three categories: individual factors, walkability factors, and perceptibility factors. Five walkability factors were considered: pedestrian pathway quality, accessibility, human density, vehicle traffic density, and structural elements such as eaves and coverings. Perceptibility factors included eight parameters: imageability, legibility, enclosure, human scale, permeability, diversity, coherence, and architectural identity. Individual factors comprised four inputs: sense of belonging, walking for transportation, walking for sports and exercise, and walking for leisure (as illustrated in Figure 9).
The input and output variables presented in Table 4 and Table 5 were obtained by following the fundamental stages of the fuzzy logic method. First, the data collected through a questionnaire as part of the field study constituted the input values. Participants rated the parameters included in the questionnaire for the route they walked on a scale where ‘0’ represented the lowest and ‘100’ the highest level of evaluation. For the inputs expressing the individual’s perspective on walking (‘walking for transportation’, ‘walking for sport and exercise’, and ‘walking for pleasure and leisure’), the linguistic data expressed in the questionnaire as ‘rarely’, ‘moderately frequently’, and ‘very frequently’ to describe weekly walking frequency were converted into numerical values (The linguistically expressed data were converted into quantitative inputs as follows: ‘rarely’ = 33.3, ‘moderately frequently’ = 66.6, and ‘very frequently’ = 99.9). In the second stage of the study, the input values were subjected to the fuzzification process. In this process, three linguistic membership sets—‘low’, ‘medium’, and ‘high’—were defined for each input variable, and the membership functions were constructed using an equally spaced approach covering the ranges of 0–33.3, 33.4–66.6, and 66.7–100. The reason for dividing the inputs into three ranges was to avoid increasing the number of rules in the fuzzy logic model, thereby significantly reducing both the structural complexity and the computational load of the model [79] (as shown in Table 4 and Figure 10)
In this study, as the modelling of subjective and perception-based variables, such as urban space perception and the walking experience, was aimed at, Gaussian-type membership functions were preferred for both input and output variables. Gaussian membership functions allow the transitions between linguistic variables to be represented in a non-abrupt and continuous manner rather than through sharp boundaries. In addition, the balanced and symmetrical structure offered by Gaussian functions, as well as their intuitive interpretability, constitute an advantage for the study [80,81] (as illustrated in Figure 10 and Figure 11).
During the fuzzy inference stage, the rule bases corresponding to three input groups—walkability factors, perceptual factors, and individual factors—were processed separately, and a single output value (score) was obtained for each group. At this stage, a total of 6885 rules were generated using the MATLAB R2024 b Fuzzy Logic Toolbox, comprising 243 rules for the walkability factors of the space, 6561 rules for the perceptual factors of the space, and 81 rules for the individual factors. No weighting was applied in the formulation of the rules, and all inputs were assumed to have equal importance (the first ten rules related to walkability factors, perceptual factors, and individual factors are presented in Supplementary Material SC). In accordance with the nature of the fuzzy logic method, the rules were formulated by considering all possible combinations of the linguistic membership sets of each parameter obtained from the literature review and defined within the conceptual framework of the study (as illustrated in Figure 3). One example rule for each group is presented below;
  • Individual Factors Rule 1: If the individual’s sense of belonging is low, and they rarely walk for transportation, sport and exercise, as well as leisure and strolling purposes, then the perception of the urban space formed during the walking experience is negative.
  • Perceptual Factors Rule 1: If the pedestrian pathway’s level of imageability is low, the level of legibility is low, the level of enclosure is low, the level of human scale is low, the level of permeability is low, the level of diversity is low, the level of coherence is low, and the level of unique architectural identity is low, then the perception of the urban space formed during the walking experience is negative.
  • Walkability Factors Rule 1: If the pedestrian pathway’s level of physical quality is low, the level of accessibility is low, the level of human density on the pathway is low, the density of motor vehicle traffic is low, and the number of structural elements such as eaves and coverings is low, then the perception of the urban space formed during the walking experience is negative.
The output value obtained during the fuzzy inference stage was converted into singular quantitative values within the range of 0–100 through the defuzzification process. In the final stage of the model, using the MATLAB R2024 b Simulink Toolbox, the walkability score, perceptibility score, and individual factors score were multiplied by a weighting coefficient of 0.333 in order to assume an equal effect of each input group on the final output, and were then summed to complete the system modelling and to calculate the final output value representing urban space perception during the walking experience. Through this weighting approach, equal contribution from the three input groups was intended, and the final output value was maintained within the range of 0–100. The final output value was linguistically associated, in accordance with the membership sets defined in Table 5, with perception levels of ‘negative’ within the range of 0–33.3, ‘neither negative nor positive’ within the range of 33.4–66.6, and ‘positive’ within the range of 66.7–100 (as illustrated in Figure 12).

3. Results

3.1. Results from Survey Data

Based on the data obtained from field studies, a comparative analysis was performed for the spatial parameters of Değirmendere Yalı Neighbourhood (Kocaeli) and Bahçelievler Neighbourhood (Yalova), comparing the results for the same participants who walked in both settlements. Accordingly, participants found the Bahçelievler Neighbourhood (Yalova) to be more walkable than the Değirmendere Yalı Neighbourhood (Kocaeli) (as shown in Table 6). When evaluating the walkability parameters of both settlements, the physical quality level of the pedestrian pathway and the intensity of motor vehicle traffic were found to be “high” for Bahçelievler Neighbourhood (Yalova) and “average” for Değirmendere Yalı Neighbourhood (Kocaeli), based on the average values given by the participants. The accessibility level of the pedestrian path, pedestrian density, and the number of structural elements such as eaves and coverings that protect individuals from climatic conditions were rated as “average” for both settlements (as shown in Table 7).
When examining both settlements in terms of the perceptibility characteristics of the space, the level of imageability, level of enclosure, level of diversity, and level of coherence of Bahçelievler Neighbourhood (Yalova) were assessed as high, while the level of imageability of Değirmendere Yalı Neighbourhood (Kocaeli) was assessed as average. The legibility level, human scale level, and permeability level of both settlements were found to be high, while the original architectural identity level was found to be average. The level of belonging felt by participants during their walking activity in these settlements was higher in the Bahçelievler Neighbourhood (Yalova) than in the Değirmendere Yalı Neighbourhood (Kocaeli) (as shown in Table 7).
When the perception of urban space in Bahçelievler Neighbourhood (Yalova) and Değirmendere Yalı Neighbourhood (Kocaeli) is comparatively evaluated through adjective pairs, participants described Bahçelievler Neighbourhood as unique, regular, and compatible. In contrast, they evaluated Değirmendere Yalı Neighbourhood as neither ordinary nor unique, neither irregular nor regular, and neither incompatible nor compatible. Both settlements were described as neither uninteresting nor interesting, defined in terms of boundaries, and spacious. When the two settlements are compared based on the mean values of perception, Bahçelievler Neighbourhood has an average of 68.43, while Değirmendere Yalı Neighbourhood has an average of 59.07. It is observed that the perception of urban space is higher in the Bahçelievler Neighbourhood (as shown in Table 8).
When the mean values of the survey data are examined, it is observed that the Bahçelievler Neighbourhood was found by university students to be both more walkable and to have a higher urban space perception score. In addition, the participants’ sense of belonging to the place they walked was also higher in the Bahçelievler Neighbourhood. In this context, it is seen that there is a positive relationship among the parameters of walkability, perceptibility, and sense of belonging.

3.2. Results from the Evaluation Using the Fuzzy Logic Model

In the fieldwork conducted in Değirmendere Yalı Neighbourhood (Kocaeli) and Bahçelievler Neighbourhood (Yalova), the answers given by each participant to the survey for the route they experienced on foot constitute the input values of the model (Supplementary Material SE Tables S1 and S2). Each value in the range of “0–100” given by the participants in the survey was transferred to the system model designed in the MATLAB R2024 b Simulink program. Thus, for each participant’s walked route, the walkability factor value, the perceptibility factor value, and the individual factors score were calculated quantitatively, and the total perception of urban spaces value was determined. According to the model designed using fuzzy logic, the obtained perception of urban spaces value was linguistically expressed as negative if it was in the range of “0–33.3”, neither negative nor positive if it was in the range of “33.4–66.6”, and positive if it was in the range of “66.7–100”.
According to the fuzzy logic evaluation, in Değirmendere Yalı (Kocaeli), 3 participants rated urban space perception as “positive,” while 22 rated it “neither negative nor positive,” with an average score of 56.042 (The detailed participant-based numerical outputs obtained from the fuzzy logic model are presented in the Supplementary Material SE Table S3). Bahçelievler (Yalova) showed similar results: 3 participants rated it “positive” and 22 “neither negative nor positive,” with an average of 60.21 (Supplementary Material SE Table S4). Thus, both settlements are linguistically perceived as “neither positive nor negative” (as shown in Table 9). This result is an expected outcome within the methodological framework of the study. It demonstrates that the model applied in two different study areas with similar urban identity characteristics is able to evaluate perception-based data in a consistent manner.
When the results of the analysis conducted using the fuzzy logic method are evaluated, parallels can be observed with the survey findings. The Bahçelievler neighbourhood (Yalova) exhibits higher walkability parameter values compared to the Değirmendere Yalı neighbourhood (Kocaeli), and the walkability parameter score calculated using the fuzzy logic method is likewise higher. Similarly, both the urban space perception value obtained from the survey results and the urban space perception value calculated through fuzzy logic are higher in the Bahçelievler neighbourhood (Yalova). The survey findings also revealed that participants experienced a stronger sense of belonging in the Bahçelievler neighbourhood, which was reflected in the fuzzy logic results. The value of Individual Factors is higher in the Bahçelievler neighbourhood (Yalova) (as illustrated in Figure 13).
In addition, a correlation analysis was conducted in order to examine the relationship between the urban space perception results measured using the fuzzy logic method and those obtained through the conventional questionnaire method. As the skewness and kurtosis values of the data did not fall outside the range of −2 to +2, and the p-values obtained from the Kolmogorov–Smirnov and Shapiro–Wilk normality tests were not smaller than 0.05, the assumption of normality was satisfied; accordingly, the Pearson correlation analysis method was employed. According to the results, a statistically significant, moderate positive correlation was found in Bahçelievler Neighbourhood (Yalova) (r = 0.418, p = 0.038). In Değirmendere Yalı Neighbourhood (Kocaeli), a statistically significant, strong positive correlation was identified (r = 0.720, p < 0.001) (as shown in Table 10).

4. Discussion and Conclusions

This study aims to present an original model that employs the fuzzy logic method to measure urban space perception within the context of the walking experience. Existing methods used to evaluate urban space perception during the walking experience—such as questionnaire-based qualitative studies, discrete choice models, machine learning/deep learning approaches, and sensor-based methods—generally require studies to be conducted with large sample sizes in order to generalise the results, thereby creating high cost-related constraints. Moreover, these studies, which are typically carried out within virtual or simulated environments, tend to overlook data related to real walking experiences associated with senses other than vision. The difficulties encountered in evaluating subjective and qualitative data using these methods constitute the main reason for the preference for the fuzzy logic method employed in this research. While the applicability of the proposed model with small samples and its reliance on walking experiences conducted in real environments represent significant advantages, the need to address the model within context-specific conditions constitutes an important limitation (as shown in Table 11).
In the existing literature, studies that utilise fuzzy logic at the intersection of walkability and perception predominantly focus on walkability criteria and pedestrian networks [53,54,55], perceptions of traffic safety [56,57], accessibility [58], and the combined evaluation of qualitative and quantitative data at various scales [59,60,61]. However, studies that address the relationship between perception and walkability holistically, including individual factors, are limited. The original model proposed in this study introduces a methodology that integrates individual and spatial factors using fuzzy logic.
Within the scope of the research, the study addressed the question of whether individual factors influencing urban space perception during the walking experience, the walkability characteristics of space, and perceptual features can be quantitatively assessed using fuzzy logic. Field data collected from the Bahçelievler neighbourhood (Yalova) and the Değirmendere Yalı neighbourhood (Kocaeli), selected as the study areas, were analysed using the proposed fuzzy logic model. The findings provide supportive evidence for the study’s hypothesis, suggesting that urban space perception can be quantitatively examined through the application of the fuzzy logic method. The analysis revealed variations in perception levels among participants, and the fuzzy logic approach enabled a more detailed examination of individual differences than conventional mean-based survey analyses. In addition, the perceptual values of urban space derived from the survey data were found to be largely consistent with those obtained through the fuzzy logic analysis. The results further indicate that the Bahçelievler neighbourhood exhibited higher spatial parameter values and was perceived more positively. Overall, the findings suggest that increases in the values of spatial parameters influencing urban space perception during the walking experience are associated with more positive perceptions of urban space.
The model proposed in this study enables the evaluation of subjective data and provides a user-centred spatial assessment. Therefore, the proposed model can be employed to compare different pedestrian routes, monitor perceptual changes over time, or evaluate the impact of design interventions. Moreover, the suggested fuzzy logic model can also be utilised in walkability indices or urban environmental perception indices that measure pedestrians’ perceptions of spatial quality levels.

Limitations and Future Research

As a result of the study, the proposed fuzzy logic-based model provides a methodological framework for the systematic and quantitative evaluation of perception-related data obtained during the walking experience. This model can be applied in studies conducted in similar urban contexts. However, by incorporating additional spatial parameters that fall outside the methodological framework proposed in this study and that influence the perception of urban spaces during the walking experience (such as topographic data, slope, noise, safety, and lighting) as input variables, the proposed model can also be applied in different urban contexts. The flexible structure of the model developed using the fuzzy logic method allows the inclusion of different parameters as inputs in subsequent studies.
Within the scope of this research, social factors, although recognised as important variables that directly influence urban space perception, were excluded from the study. As the reliable assessment of social parameters requires larger and more heterogeneous samples representing different social groups, the analysis of these factors should be addressed within the framework of a separate and comprehensive research design. In particular, in order to examine socio-psychological parameters affecting individual perception more comprehensively in future studies, collaboration with various disciplines such as sociology and psychology is essential.
In this study, the homogeneity of the sample, consisting of university students, limits the evaluation of perceptual differences related to demographic variables such as age, gender, occupation, and household income, as well as socio-economic variables. In future studies to be conducted with larger and more heterogeneous samples encompassing different age groups and varying socio-economic levels, the effects of age, gender, occupation, physical disability, and other demographic parameters on urban space perception can be examined in detail through the fuzzy logic model. The representativeness of this experimental study employing the fuzzy logic method will be enhanced in subsequent research by expanding the sample size and testing the model in contexts with different urban identities. All these aspects constitute the main limitations that should be taken into consideration when evaluating the findings obtained from the research. In future studies, these issues may be addressed in order to develop a more comprehensive approach to the relationship between perception and walkability.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/su18062781/s1. SA: Walkability Parameters, SB: Semantic Differential Scales, SC: Rules Section, SD: Information on Research Areas, SE: Input and Output Values Evaluated Using the Fuzzy Logic Method, Figure S1: Schematic section of the route walked by participants in Değirmendere Yalı Neighbourhood (Kocaeli), Figure S2: Schematic section of the route walked by participants in Bahçelievler Neighbourhood (Yalova), Figure S3. (a): Views of the sea encountered along the streets in Değirmendere Yalı Neighbourhood (Kocaeli), Figure S3 (b): Views of the sea encountered along the streets in Bahçelievler Neighbourhood (Yalova), Figure S4. (a): Some sculptures located in Değirmendere Yalı Neighbourhood (Kocaeli), Figure S4. (b): Some sculptures located in Bahçelievler Neighbourhood (Yalova), Table S1: Input Values Obtained in Değirmendere Yalı Neighbourhood (Kocaeli), Table S2: Input Values Obtained in Bahçelievler Mahallesi Neighbourhood (Yalova), Table S3. Perceived value of perception of urban spaces in Değirmendere Yalı Neighbourhood according to fuzzy logic evaluation, Table S4. Perception of urban spaces value according to fuzzy logic assessment in Bahçelievler Neighbourhood.

Author Contributions

Conceptualization, E.B. and M.Ö.; methodology, E.B. and S.Y.; software, E.B. and S.Y.; validation, E.B., M.Ö. and S.Y.; formal analysis, E.B.; investigation, E.B.; resources, E.B. and M.Ö.; data curation, E.B.; writing—original draft preparation, E.B.; writing—review and editing, M.Ö. and S.Y.; visualization, E.B.; supervision, M.Ö. and S.Y.; project administration, E.B.; funding acquisition, E.B. All authors have read and agreed to the published version of the manuscript.

Funding

This study was conducted as part of a PhD thesis and was funded by the Scientific and Technological Research Council of Türkiye (TÜBİTAK), 1002-A Programme, Grant No. 223K272.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Kocaeli University Institute of Science (No.E-20189260-100-688666 Date: 26 November 2024).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors upon request.

Acknowledgments

The authors are grateful to all those who contributed to the development of the research through their insights, feedback, and academic support, as well as to the university students who kindly participated in the survey.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
TÜİKTurkish Statistical Institute
TÜBİTAKThe Scientific and Technological Research Council of Türkiye

References

  1. Pallasmaa, J. Tenin Gözleri: Mimarlık ve Duyular; İnka Yayınları: İzmir, Turkey, 2019. [Google Scholar]
  2. Gehl, J. İnsan İçin Kentler; Koç Üniversitesi Yayınları: İstanbul, Turkey, 2020. [Google Scholar]
  3. Pinder, D. Arts of Urban Exploration. Cult. Geogr. 2005, 12, 383–411. [Google Scholar] [CrossRef]
  4. Lynch, K. Kentin İmgesi; Başaran, İ., Ed.; İş Bankası Yayınları: İstanbul, Turkey, 1960. [Google Scholar]
  5. Erturan Topgül, G.A. Kentlerde Yürünebilirliğin Çok Boyutlu Yaklaşım ve Hareketli Yöntemlerle Analizi. Ph.D. Thesis, Mimar Sinan Fine Arts University, İstanbul, Turkey, 2021. [Google Scholar]
  6. Jaśkiewicz, M.; Besta, T. Is Easy Access Related to Better Life? Walkability and Overlapping of Personal and Communal Identity as Predictors of Quality of Life. Appl. Res. Qual. Life 2014, 9, 505–516. [Google Scholar] [CrossRef]
  7. Wang, H.; Yang, Y. Neighbourhood Walkability: A Review and Bibliometric Analysis. Cities 2019, 93, 43–61. [Google Scholar] [CrossRef]
  8. Wang, L.; Ma, L.; Yang, J.; Wu, J. Human Somatosensory Processing and Artificial Somatosensation. Cyborg Bionic Syst. 2021, 2021, 9843259. [Google Scholar] [CrossRef]
  9. Oliva, A.; Mack, M.L.; Shrestha, M.; Peeper, A. Identifying the Perceptual Dimensions of Visual Complexity of Scenes. In Proceedings of the 26th Annual Meeting of the Cognitive Science Society, Chicago, IL, USA, 4–7 August 2004. [Google Scholar]
  10. Mostafavi, A. Architecture, Biometrics, and Virtual Environments Triangulation: A Research Review. Archit. Sci. Rev. 2022, 65, 504–521. [Google Scholar] [CrossRef]
  11. Rapoport, A. Human Aspects of Urban Form; Pergamon Press: Oxford, UK, 1977. [Google Scholar]
  12. Lang, J.T. Creating Architectural Theory: The Role of the Behavioral Sciences in Environmental Design; Van Nostrand Reinhold: New York, NY, USA, 1987. [Google Scholar]
  13. Ocakçı, M. Kent İmgesi (İmajı). In Kentsel Planlama Ansiklopedi Sözlük; Ersoy, M., Ed.; Ninova Yayıncılık: İstanbul, Turkey, 2016. [Google Scholar]
  14. Rapoport, A. History and Precedent in Environmental Design; Springer: Boston, MA, USA, 1990. [Google Scholar] [CrossRef]
  15. Forsyth, A.; Hearst, M.; Oakes, J.M.; Schmitz, K.H. Design and Destinations: Factors Influencing Walking and Total Physical Activity. Urban Stud. 2008, 45, 1973–1996. [Google Scholar] [CrossRef]
  16. Saelens, B.E.; Handy, S.L. Built Environment Correlates of Walking: A Review. Med. Sci. Sports Exerc. 2008, 40, S550. [Google Scholar] [CrossRef] [PubMed]
  17. Sugiyama, T.; Neuhaus, M.; Cole, R.; Giles-Corti, B.; Owen, N. Destination and Route Attributes Associated with Adults’ Walking: A Review. Med. Sci. Sports Exerc. 2012, 44, 1275–1286. [Google Scholar] [CrossRef] [PubMed]
  18. Farkic, J.; Peric, D.; Lesjak, M.; Petelin, M. Urban Walking: Perspectives of Locals and Tourists. Geogr. Pannonica 2015, 19, 212–222. [Google Scholar] [CrossRef]
  19. Mehta, V. Walkable Streets: Pedestrian Behavior, Perceptions and Attitudes. J. Urban. Int. Res. Placemaking Urban Sustain. 2008, 1, 217–245. [Google Scholar] [CrossRef]
  20. Göregenli, M. Çevre Psikolojisi: İnsan Mekân İlişkileri; İstanbul Bilgi Üniversitesi Yayınları: İstanbul, Turkey, 2010. [Google Scholar]
  21. Deutsch, K.; Goulias, K. Investigating the Impact of Sense of Place on Travel Behavior Using an Intercept Survey Methodology; University of California Transportation Center: Berkeley, CA, USA, 2009. [Google Scholar]
  22. Farr, D. Sustainable Urbanism: Urban Design with Nature; John Wiley & Sons: Hoboken, NJ, USA, 2011. [Google Scholar]
  23. Montello, D.R. Navigation. In The Cambridge Handbook of Visuospatial Thinking; Cambridge University Press: Cambridge, UK, 2005; pp. 257–294. [Google Scholar] [CrossRef]
  24. Halu, Z.Y. Transactional Approach for Walkable Urban Spaces: Hierarchy of Walking Needs. J. Environ. Prot. Ecol. 2019, 20, 302–312. [Google Scholar]
  25. Arthur, P.; Passini, R. Wayfinding: People, Signs, And Architecture; Focus Strategic Communications: Brantford, ON, Canada, 2002. [Google Scholar]
  26. Guo, Z.; Loo, B.P.Y. Pedestrian Environment and Route Choice: Evidence from New York City and Hong Kong. J. Transp. Geogr. 2013, 28, 124–136. [Google Scholar] [CrossRef]
  27. Wozniak, M.; Filomena, G.; Wronkowski, A. What’s Your Type? A Taxonomy of Pedestrian Route Choice Behaviour in Cities. Transp. Res. Part F Traffic Psychol. Behav. 2025, 109, 1257–1274. [Google Scholar] [CrossRef]
  28. Siewwuttanagul, S.; Hayashida, Y.; Inohae, T. Identifying Pedestrian Movement Behaviour Using Object Detection Methods and Land-Use Agglomeration Analysis. ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci. 2018, 4, 123–128. [Google Scholar] [CrossRef]
  29. Fonseca, F.; Ribeiro, P.J.G.; Conticelli, E.; Jabbari, M.; Papageorgiou, G.; Tondelli, S.; Ramos, R.A.R. Built Environment Attributes and Their Influence on Walkability. Int. J. Sustain. Transp. 2022, 16, 660–679. [Google Scholar] [CrossRef]
  30. Gehl, J. Life Between Buildings: Using Public Place; Island Press: Washington, DC, USA, 2011; Volume 297. [Google Scholar]
  31. Southworth, M. Designing the Walkable City. J. Urban Plan. Dev. 2005, 131, 246–257. [Google Scholar] [CrossRef]
  32. Jin, S.; Kim, E.J. Correlation of the Walk Score and Environmental Perceptions with Perceived Neighborhood Walkability: The Quantile Regression Model Approach. Sustainability 2024, 16, 7074. [Google Scholar] [CrossRef]
  33. Ewing, R.; Handy, S.; Brownson, R.C.; Clemente, O.; Winston, E. Identifying and Measuring Urban Design Qualities Related to Walkability. J. Phys. Act. Health 2006, 3, S223–S240. [Google Scholar] [CrossRef]
  34. Ramírez, T.; Hurtubia, R.; Lobel, H.; Rossetti, T. Measuring Heterogeneous Perception of Urban Space with Massive Data and Machine Learning: An Application to Safety. Landsc. Urban Plan. 2021, 208, 104002. [Google Scholar] [CrossRef]
  35. Chen, N.; Wang, L.; Xu, T.; Wang, M. Perception of Urban Street Visual Color Environment Based on the CEP-KASS Framework. Landsc. Urban Plan. 2025, 259, 105359. [Google Scholar] [CrossRef]
  36. Huang, C.; Wei, F.; Han, Q.; Xu, J.; Qiu, S.; Ban, X.; Huang, Y.; Huang, T. Visual and Emotional Interaction between People and Post-Industrial Riverscape Based on the Significance of “Original—New Placement”. Ecol. Indic. 2024, 163, 112135. [Google Scholar] [CrossRef]
  37. Shi, J.; Yan, Y.; Li, M.; Zhou, L. Measuring the Convergence and Divergence in Urban Street Perception among Residents and Tourists through Deep Learning: A Case Study of Macau. Land 2024, 13, 345. [Google Scholar] [CrossRef]
  38. Sun, D.; Zhou, F.; Lin, J.; Yang, Q.; Lyu, M. A Study on Landscape Feature and Emotional Perception Evaluation of Waterfront Greenway. Environ. Res. Commun. 2024, 6, 095023. [Google Scholar] [CrossRef]
  39. Freitas, F.; Berreth, T.; Chen, Y.-C.; Jhala, A. Characterizing the Perception of Urban Spaces from Visual Analytics of Street-Level Imagery. AI Soc. 2023, 38, 1361–1371. [Google Scholar] [CrossRef]
  40. Gönül, A.; Durak, S. On Measuring the Change in Historical City Centres: An Attempt at Comparing Human Perception and Deep Learning through Visual Quality of Street Space. Multimed. Tools Appl. 2025, 84, 39283–39305. [Google Scholar] [CrossRef]
  41. Ogawa, Y.; Oki, T.; Zhao, C.; Sekimoto, Y.; Shimizu, C. Evaluating the Subjective Perceptions of Streetscapes Using Street-View Images. Landsc. Urban Plan. 2024, 247, 105073. [Google Scholar] [CrossRef]
  42. Qin, Y.; Wu, X.; Yu, T.; Jiang, S. Enhancing Student-Centered Walking Environments on University Campuses through Street View Imagery and Machine Learning. PLoS ONE 2025, 20, e0321028. [Google Scholar] [CrossRef]
  43. Luo, S.; Shi, J.; Lu, T.; Furuya, K. Sit down and Rest: Use of Virtual Reality to Evaluate Preferences and Mental Restoration in Urban Park Pavilions. Landsc. Urban Plan. 2022, 220, 104336. [Google Scholar] [CrossRef]
  44. Lee, S.; Lee, C.; Nam, J.W.; Abbey-Lambertz, M.; Mendoza, J.A. School Walkability Index: Application of Environmental Audit Tool and GIS. J. Transp. Health 2020, 18, 100880. [Google Scholar] [CrossRef] [PubMed]
  45. Li, Y.; Yabuki, N.; Fukuda, T. Measuring Visual Walkability Perception Using Panoramic Street View Images, Virtual Reality, and Deep Learning. Sustain. Cities Soc. 2022, 86, 104140. [Google Scholar] [CrossRef]
  46. Park, S.; Deakin, E.; Lee, J.S. Perception-Based Walkability Index to Test Impact of Microlevel Walkability on Sustainable Mode Choice Decisions. Transp. Res. Rec. 2014, 2464, 126–134. [Google Scholar] [CrossRef]
  47. Lo, R.H. Walkability: What Is It? J. Urban. Int. Res. Placemaking Urban Sustain. 2009, 2, 145–166. [Google Scholar] [CrossRef]
  48. Sallis, J.F.; Cervero, R.B.; Ascher, W.; Henderson, K.A.; Kraft, M.K.; Kerr, J. An Ecological Approach to Creating Active Living Communities. Annu. Rev. Public Health 2006, 27, 297–322. [Google Scholar] [CrossRef]
  49. Yıldırım, Ö.C.; Çelik, E. Understanding Pedestrian Behavior and Spatial Relations: A Pedestrianized Area in Besiktas, Istanbul. Front. Archit. Res. 2023, 12, 67–84. [Google Scholar] [CrossRef]
  50. Zadeh, L.A. Fuzzy Sets. Inf. Control 1965, 8, 338–353. [Google Scholar] [CrossRef]
  51. Tanaka, K.; Sugeno, M. Stability Analysis and Design of Fuzzy Control Systems. Fuzzy Sets Syst. 1992, 45, 135–156. [Google Scholar] [CrossRef]
  52. Mert, Z.G.; Yilmaz, S. Fuzzy Modeling Approach Based on Property Location Quality for Grading Neighborhood Level of Family Housing Units. Expert Syst. Appl. 2009, 36, 3603–3613. [Google Scholar] [CrossRef]
  53. AlShareef, F.; Aljoufie, M. Identification of the Proper Criteria Set for Neighborhood Walkability Using the Fuzzy Analytic Hierarchy Process Model: A Case Study in Jeddah, Saudi Arabia. Sustainability 2020, 12, 9286. [Google Scholar] [CrossRef]
  54. Doğan, U. Examining Urban Design Characteristics of City Centers Using Walkability Criteria: Case of Turkey. J. Urban Plan. Dev. 2021, 147, 04021003. [Google Scholar] [CrossRef]
  55. Fonseca, F.; Ribeiro, P.; Jabbari, M.; Petrova, E.; Papageorgiou, G.; Conticelli, E.; Tondelli, S.; Ramos, R. Smart Pedestrian Network: An Integrated Conceptual Model for Improving Walkability. In Proceedings of the International Conference on Society with Future: Smart and Liveable Cities, Braga, Portugal, 4–6 December 2020; pp. 125–142. [Google Scholar] [CrossRef]
  56. Nyagah, P. A Multi Procedural Approach to Evaluating Walkability and Pedestrian Safety. Ph.D. Thesis, University of Nevada, Las Vegas, NV, USA, 2015. [Google Scholar]
  57. Oestreich, L.; Torres, T.B.; Ruiz-Padillo, A. Fuzzy Analysis of Students’ Perception of Traffic Safety in School Environments: The Case of a Small Brazilian City. Int. J. Inj. Contr. Saf. Promot. 2021, 28, 255–265. [Google Scholar] [CrossRef] [PubMed]
  58. Soares Müller, A.P.; Ruiz-Padillo, A. Fuzzy Micro-Scale Accessibility Indexes from the Perception of Pedestrians with Disabilities: Case Study of a Medium-Sized Latin American City. Case Stud. Transp. Policy 2025, 21, 101534. [Google Scholar] [CrossRef]
  59. Jabbari, M.; Fonseca, F.; Ramos, R. Combining Multi-Criteria and Space Syntax Analysis to Assess a Pedestrian Network: The Case of Oporto. J. Urban Des. 2018, 23, 23–41. [Google Scholar] [CrossRef]
  60. Jabbari, M.; da Fonseca, F.P.; Ramos, R.A.R. Assessing the Pedestrian Network Conditions in Two Cities: The Cases of Qazvin and Porto. In Urban Heritage Along the Silk Roads; Springer: Berlin/Heidelberg, Germany, 2020; pp. 229–245. [Google Scholar] [CrossRef]
  61. Yıldız, B.; Cağdaş, G. Agent-Based Modeling for User Movements Using Fuzzy. JCoDe J. Comput. Des. 2021, 2, 243–264. [Google Scholar]
  62. Lee, E.; Dean, J. Perceptions of Walkability and Determinants of Walking Behaviour among Urban Seniors in Toronto, Canada. J. Transp. Health 2018, 9, 309–320. [Google Scholar] [CrossRef]
  63. Blečić, I.; Congiu, T.; Fancello, G.; Trunfio, G.A. Planning and Design Support Tools for Walkability: A Guide for Urban Analysts. Sustainability 2020, 12, 4405. [Google Scholar] [CrossRef]
  64. Neuman, W.L. Social Research Methods: Qualitative and Quantitative Approaches; Pearson Education Limited: Essex, UK, 2011. [Google Scholar]
  65. Özdemir, M. Nitel Veri Analizi: Sosyal Bilimlerde Yöntembilim Sorunsalı Üzerine Bir Çalışma. Eskişehir Osman. Üniversitesi Sos. Bilim. Derg. 2010, 11, 323–343. [Google Scholar]
  66. Baltaci, A. Nitel Araştırma Süreci: Nitel Bir Araştırma Nasıl Yapılır? Ahi Evran Üniversitesi Sos. Bilim. Enstitüsü Derg. 2019, 5, 368–388. [Google Scholar] [CrossRef]
  67. Özyılmaz Küçükyağcı, P. Kent Meydanlarının Mekân Tasarımı Niteliklerinin Bulanık Mantık Ile Değerlendirilmesi. Ph.D. Thesis, Istanbul Technical University, İstanbul, Turkey, 2020. [Google Scholar]
  68. Wang, R.; Ren, S.; Zhang, J.; Yao, Y.; Wang, Y.; Guan, Q. A Comparison of Two Deep-Learning-Based Urban Perception Models: Which One Is Better? Comput. Urban Sci. 2021, 1, 3. [Google Scholar] [CrossRef]
  69. Polatoğlu, Ç. Mimarlıkta Görsel Etki Değerlendirme Yöntem ve Teknikleri; Yıldız Teknik Üniversitesi Basım Yayın Merkezi: İstanbul, Turkey, 2012. [Google Scholar]
  70. Küller, R. (Ed.) Beyond Semantic Measurement. In Architectural Psychology, Proceedings of Lund Conference; Studentlitteratur AB: Lund, Sweden, 1973. [Google Scholar]
  71. Yildiz Kuyrukçu, E.; Özdemir Erdoğan, T. Perceptional Differences in Architectural Facade Perception Due to Architectural Education. Int. Ref. J. Des. Archit. 2021, 23, 114–145. [Google Scholar] [CrossRef]
  72. Aytuğ, A. Mimaride Doku Kullanımının Psikolojik Etkileri Üzerine Bir Araştırma. Stud. Psychol. 1987, 17, 37–46. [Google Scholar]
  73. Ertürk, S. Mekan Bileşenlerinin Tasarımında Doku Boyutu; KTÜ İnşaat ve Mimarlık Fakültesi Yayınları: Trabzon, Turkey, 1979. [Google Scholar]
  74. Hershberger, R.G.; Cass, R.C. Predicting User Responses to Buildings. In Man-Environment Interactions: Evaluations and Applications; Cambridge University Press: Cambridge, UK, 1974; pp. 117–143. [Google Scholar]
  75. Temel, S.C.; Kuru, R.; Seçal, S. Tarihi Çevrede Eski Yeni Birlikteliğinin Öznel İzlenimler Üzerinden Değerlendirilmesi: Divanyolu Caddesi (Beyazıt-Sirkeci Aksı) Örneği. Sanat Tasarım Derg. 2021, 27, 379–405. [Google Scholar] [CrossRef]
  76. Yıldırım, K.; Arslan, H.D. Perceptual Evaluation of Traditional Turkish House Façade. ICONARP Int. J. Archit. Plan. 2021, 9, 742–768. [Google Scholar] [CrossRef]
  77. TÜİK. Available online: https://biruni.tuik.gov.tr/medas/?kn=95&locale=tr (accessed on 15 September 2025).
  78. Yalova Municipality. Yalova İli, Merkez İlçesi, Bahçelievler Mahallesi 1/1000 Ölçekli Uygulama İmar Planı Revizyonu Plan Açıklama Raporu; Yalova Municipality: Yalova, Türkiye, 2022.
  79. Özyılmaz Küçükyağcı, P.; Ocakçı, M. A Mamdani-Based Fuzzy Logic Model for Evaluating the Design Quality of Urban Squares. ICONARP Int. J. Archit. Plan. 2025, 13, 286–312. [Google Scholar] [CrossRef]
  80. Shamoi, P.; Toganas, N.; Muratbekova, M.; Kadyrgali, E.; Yerkin, A.; Igali, A.; Ziyada, M.; Adilova, A.; Karatayev, A.; Torekhan, Y. COLIBRI Fuzzy Model: Color Linguistic-Based Representation and Interpretation. IEEE Access 2025, 13, 205932–205956. [Google Scholar] [CrossRef]
  81. Akdemir, H.G. On Generalized Picture Fuzzy Numbers with Gaussian Membership Functions. J. Syst. Sci. Syst. Eng. 2025. [Google Scholar] [CrossRef]
Sustainability 18 02781 g001
Figure 1. Fuzzy logic working system [52].
Figure 1. Fuzzy logic working system [52].
Sustainability 18 02781 g001
Sustainability 18 02781 g002
Figure 2. Schematic representation of the methodological stages of the study.
Figure 2. Schematic representation of the methodological stages of the study.
Sustainability 18 02781 g002
Sustainability 18 02781 g003
Figure 3. The conceptual framework of the study.
Figure 3. The conceptual framework of the study.
Sustainability 18 02781 g003
Sustainability 18 02781 g004
Figure 4. Schematic representation of the design process of the questionnaire items.
Figure 4. Schematic representation of the design process of the questionnaire items.
Sustainability 18 02781 g004
Sustainability 18 02781 g005
Figure 5. Route walked in Değirmendere Yalı Neighbourhood (Gölcük/Kocaeli).
Figure 5. Route walked in Değirmendere Yalı Neighbourhood (Gölcük/Kocaeli).
Sustainability 18 02781 g005
Sustainability 18 02781 g006
Figure 6. Route walked in Bahçelievler Neighbourhood (Centre/Yalova).
Figure 6. Route walked in Bahçelievler Neighbourhood (Centre/Yalova).
Sustainability 18 02781 g006
Sustainability 18 02781 g007
Figure 7. (a) The route walked by participants in Değirmendere Yalı Neighbourhood (Kocaeli); (b) The route walked by participants in Bahçelievler Neighbourhood (Yalova).
Figure 7. (a) The route walked by participants in Değirmendere Yalı Neighbourhood (Kocaeli); (b) The route walked by participants in Bahçelievler Neighbourhood (Yalova).
Sustainability 18 02781 g007
Sustainability 18 02781 g008
Figure 8. The Operational system of the fuzzy logic model proposed in the study.
Figure 8. The Operational system of the fuzzy logic model proposed in the study.
Sustainability 18 02781 g008
Sustainability 18 02781 g009
Figure 9. (a) Individual Factors inputs in MATLAB Fuzzy Logic Toolbox; (b) Walkability Factors inputs in MATLAB Fuzzy Logic Toolbox; (c) Perceptibility Factors inputs in MATLAB Fuzzy Logic Toolbox.
Figure 9. (a) Individual Factors inputs in MATLAB Fuzzy Logic Toolbox; (b) Walkability Factors inputs in MATLAB Fuzzy Logic Toolbox; (c) Perceptibility Factors inputs in MATLAB Fuzzy Logic Toolbox.
Sustainability 18 02781 g009
Sustainability 18 02781 g010
Figure 10. (a) The membership functions of fuzzy sets for inputs in MATLAB and the appearance of low value ranges; (b) medium value ranges; (c) high value ranges.
Figure 10. (a) The membership functions of fuzzy sets for inputs in MATLAB and the appearance of low value ranges; (b) medium value ranges; (c) high value ranges.
Sustainability 18 02781 g010
Sustainability 18 02781 g011
Figure 11. (a) The membership functions of fuzzy sets for outputs in MATLAB, and the appearance of low value ranges; (b) medium value ranges; (c) high value ranges.
Figure 11. (a) The membership functions of fuzzy sets for outputs in MATLAB, and the appearance of low value ranges; (b) medium value ranges; (c) high value ranges.
Sustainability 18 02781 g011
Sustainability 18 02781 g012
Figure 12. System model established using the MATLAB R2024 b Simulink program.
Figure 12. System model established using the MATLAB R2024 b Simulink program.
Sustainability 18 02781 g012
Sustainability 18 02781 g013
Figure 13. Fuzzy Logic Results in Değirmendere and Bahçelievler Neighbourhood.
Figure 13. Fuzzy Logic Results in Değirmendere and Bahçelievler Neighbourhood.
Sustainability 18 02781 g013
Table 1. Comparative table of existing methods for measuring urban space perception.
Table 1. Comparative table of existing methods for measuring urban space perception.
MethodsSurvey-Based Qualitative StudiesDiscrete Choice ModelsMachine Learning/Deep Learning ApproachesSensor-Based Methods (VR, EEG, Eye Tracking, etc.)
Type of DataSurvey/Interview Data
Qualitative Data
Quantitative Data		Image-Based Data
Image Segmentation		Image-Based Data
Physiological Data
Quantitative Data		Physiological Data
Quantitative Data
StrengthsAllows the direct expression of individual perceptionsEnables the analysis of the effects of visual variables on perceptionIt can model complex relationships with more variables.It can directly measure perceptual responses.
LimitationsConducting studies with large sample sizes is required in order to generalise the results.
The quantification of subjective/linguistic data related to perception is difficult.		A large amount of data needs to be handled.
High Implementation Cost.
Detachment from the real walking experience.
Addresses the non-visual and qualitative (subjective) components of perception in a limited manner.		Addresses the non-visual and qualitative (subjective) components of perception in a limited manner.High Implementation Cost,
Limited sample size and representativeness issues,
Typically based on a virtual environment,
Detachment from the real walking experience.
Table 2. Age and gender distribution of the participants.
Table 2. Age and gender distribution of the participants.
Gender/Age1819202122
Female12532
Male15531
Table 3. Information on the study areas.
Table 3. Information on the study areas.
LocationPopulationArea Size (km2)Population Density (Persons/km2)Location IDSettlement PlanFunctional Feature
Değirmendere Yalı Neighbourhood (Gölcük, Kocaeli)49850.3116,081-Coastal
-Earthquake		-Organic Settlement Plan-Residential + Commercial
Bahçelievler Neighbourhood (Center, Yalova)12,5961.31
(0.80) a		9615
(15,745) b		-Coastal
-Earthquake		-Grid Settlement Plan-Residential + Commercial
a It is the area size measured using residential area boundaries as a reference. b Population density measured by reference to residential area boundaries.
Table 4. Inputs, membership values, and linguistic equivalents.
Table 4. Inputs, membership values, and linguistic equivalents.
InputsMembership Functions
0–33.333.4–66.666.7–100
Physical Quality Level of the Pedestrian PathLowAverageHigh
Accessibility Level of the Pedestrian PathLowAverageHigh
Human DensityLowAverageHigh
Motor Vehicle Traffic DensityLowAverageHigh
Number of Structural Elements such as Eaves, Roofing, etc.LowAverageHigh
Level of ImageabilityLowAverageHigh
Legibility LevelLowAverageHigh
Enclosure LevelLowAverageHigh
Human Scale LevelLowAverageHigh
Permeability LevelLowAverageHigh
Diversity LevelLowAverageHigh
Coherence LevelLowAverageHigh
Architectural Identity LevelLowAverageHigh
Sense of Belonging LevelLowAverageHigh
Walking for Transportation PurposesRarelyModeratelyVery frequently
Walking for Sports and Exercise PurposesRarelyModeratelyVery often
Walking for Pleasure and LeisureRarelyModeratelyVery often
Table 5. Outputs, membership values, and linguistic equivalents.
Table 5. Outputs, membership values, and linguistic equivalents.
OutputsMembership Functions
0–33.333.4–66.666.7–100
Perception of Urban Spaces During Walking ExperienceNegativeNeither Negative
Nor Positive		Positive
Table 6. Average walkability values for Değirmendere Yalı Neighbourhood (Kocaeli) and Bahçelievler Neighbourhood (Yalova).
Table 6. Average walkability values for Değirmendere Yalı Neighbourhood (Kocaeli) and Bahçelievler Neighbourhood (Yalova).
LocationBahçelievler Neighbourhood (Yalova)Değirmendere Yalı Neighbourhood (Kocaeli)
Walkability Score82.8078
Table 7. Bahçelievler Neighbourhood (Yalova) and Değirmendere Yalı Neighbourhood (Kocaeli) average values given to parameters (0–33.3: Low, 33.4–66.6: Average, 66.7–100: High).
Table 7. Bahçelievler Neighbourhood (Yalova) and Değirmendere Yalı Neighbourhood (Kocaeli) average values given to parameters (0–33.3: Low, 33.4–66.6: Average, 66.7–100: High).
FactorsParametersBahçelievler Neighbourhood (Yalova)Value RangeDeğirmendere Yalı
Neighbourhood (Kocaeli)		Value Range
Walkability FactorsPhysical Quality Level of the Pedestrian Path69.6High58.40Average
Accessibility Level of the Pedestrian Path55.16Average50.80Average
Population Density55.44Average55.60Average
Motor Vehicle Traffic Density67.00High63.20Average
Number of Structural Elements such as Eaves, Roofing, etc.47.00Average39.20Average
Perceptual FactorsLevel of Imageability71.40High49.60Average
Legibility Level83.80High80.40High
Enclosure Level69.00High63.40Average
Human Scale Level75.20High71.20High
Permeability Level74.60High73.20High
Diversity Level71.00High50.80Average
Coherence Level72.20High58.28Average
Architectural Identity Level57.00Average49.00Average
Individual F.Sense of Belonging Level66.6High52.80Average
Table 8. Comparison of perception values based on adjective pairs in Bahçelievler Neighbourhood (Yalova) and Değirmendere Yalı Neighbourhood (Kocaeli) (0–33.3: Low, 33.4–66.6: Average, 66.7–100: High).
Table 8. Comparison of perception values based on adjective pairs in Bahçelievler Neighbourhood (Yalova) and Değirmendere Yalı Neighbourhood (Kocaeli) (0–33.3: Low, 33.4–66.6: Average, 66.7–100: High).
Semantic Differential ScalesBahçelievler Neighbourhood (Yalova)Value RangeDeğirmendere Yalı Neighbourhood (Kocaeli)Value Range
ordinary–striking66.00High50.40Average
uninteresting–interesting52.40Average41.60Average
irregular–regular67.60High64.80Average
undefined boundaries–defined boundaries72.80High67.60High
cramped–spacious78.00High68.80High
incompatible–compatible73.80High61.20Average
Overall Average68.43High59.07Average
Table 9. Perception of urban spaces results measured using fuzzy logic in Değirmendere Yalı Neighbourhood (Kocaeli) and Bahçelievler Neighbourhood (Yalova) (0–33.3: Negative, 33.4–66.6: Neither Negative Nor Positive, 66.7–100: Positive).
Table 9. Perception of urban spaces results measured using fuzzy logic in Değirmendere Yalı Neighbourhood (Kocaeli) and Bahçelievler Neighbourhood (Yalova) (0–33.3: Negative, 33.4–66.6: Neither Negative Nor Positive, 66.7–100: Positive).
LocationWalkability Factor
Value		Perceptibility Factor ValueIndividual Factors ValueFuzzy Logic ResultLinguistic
Equivalent
Bahçelievler
Neighbourhood (Yalova)		18.39821.37820.14660.21Neither Negative Nor Positive
Değirmendere Yalı Neighbourhood
(Kocaeli)		17.3719.3219.3656.042Neither Negative Nor Positive
Table 10. Correlation table of perception results measured using the fuzzy logic method and the questionnaire method.
Table 10. Correlation table of perception results measured using the fuzzy logic method and the questionnaire method.
LocationPearson Correlation Coefficientp
Value
Değirmendere Yalı Neighbourhood (Kocaeli)0.720<0.001
Bahçelievler Neighbourhood (Yalova)0.4180.038
Table 11. Comparative information on the proposed fuzzy logic–based model and other methods.
Table 11. Comparative information on the proposed fuzzy logic–based model and other methods.
MethodsOther Methods.
(Survey, Discrete Choice Models, Machine Learning, Sensor-Based Methods)		Fuzzy Logic-Based Models
Type of DataSurvey/Interview Data
Qualitative Data
Quantitative Data
Image-Based Data
Image Segmentation
Physiological Data		Survey/Interview Data
Quantitative Data
Qualitative Data
StrengthsAllows the direct expression of individual perceptions
Enables the analysis of the effects of visual variables on perception
It can model complex relationships with more variables.
It can directly measure perceptual responses.		It can incorporate uncertainty and subjective evaluations into the model’s structure.
Can be applied with small samples.
Is based on real walking experiences.
It can model perception in a multidimensional and holistic way.
It provides a systematic approach to measuring perception.
Due to the flexible structure of the model, different parameters can be integrated into the system.
LimitationsConducting studies with large sample sizes is required in order to generalise the results.
The quantification of subjective/linguistic data related to perception is difficult.
High Implementation Cost.
Detachment from the real walking experience.		The sensitivity of the proposed model to context-specific characteristics.
An increase in the number of rules generated and the computational load as the number of inputs integrated into the model increases.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Baran, E.; Özbayraktar, M.; Yılmaz, S. A Fuzzy Logic-Based Model for Measuring Perception of Urban Spaces During Walking Experience. Sustainability 2026, 18, 2781. https://doi.org/10.3390/su18062781

AMA Style

Baran E, Özbayraktar M, Yılmaz S. A Fuzzy Logic-Based Model for Measuring Perception of Urban Spaces During Walking Experience. Sustainability. 2026; 18(6):2781. https://doi.org/10.3390/su18062781

Chicago/Turabian Style

Baran, Esra, Mehtap Özbayraktar, and Serhat Yılmaz. 2026. "A Fuzzy Logic-Based Model for Measuring Perception of Urban Spaces During Walking Experience" Sustainability 18, no. 6: 2781. https://doi.org/10.3390/su18062781

APA Style

Baran, E., Özbayraktar, M., & Yılmaz, S. (2026). A Fuzzy Logic-Based Model for Measuring Perception of Urban Spaces During Walking Experience. Sustainability, 18(6), 2781. https://doi.org/10.3390/su18062781

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop