The Threshold Effect in the Street Vitality Formation Mechanism

Abstract

Street vitality has become a crucial metric for smart city management. Classical theories qualitatively explain that street vitality originates from the dynamic interaction between people and spatial carriers, yet the threshold effect within this process has not been addressed, leaving a gap in urban research. This study selects South China, one of China’s most vibrant and globally influential regions, introduces dissipative structure theory based on classical theories, and constructs a threshold effect hypothesis model for the vitality formation mechanism. Through energy efficiency conversion of data and a slope-based method for identifying balanced time periods, the periods of supply–demand balance in energy efficiency were identified, the threshold effect in vitality formation was captured, and critical thresholds were measured. The results indicate the following: (1) the hypothesis model is valid; (2) the threshold effect is inevitable and periodic, primarily occurring on workdays from 12:00 to 13:00 and 18:00 to 19:00, and on rest days from 08:00 to 09:00 and 18:00 to 19:00; and (3) the activation threshold is quantifiable and exhibits volatility, ranging from 0.40 to 1.56, varying specifically by city, season, day type, and street type. This study advances the translation of street vitality research from theory into practice and provides theoretical support and strategic guidance for smart city management globally, particularly in developing countries.

1. Introduction

Scholars worldwide are increasingly focusing on street vitality levels and regard it as a key indicator for assessing the quality of urban public spaces and sustainable development capacity [1,2,3]. This trend indicates that research on street vitality is shifting from theoretical exploration to practical application. However, significant gaps remain in the understanding of the formation mechanisms of street vitality in existing research. Although classical theories qualitatively elucidate that vitality originates from the dynamic interaction between people and spatial carriers (built environment) [4,5,6,7,8], they do not explore what threshold this interaction must reach to generate vitality, which has become a major bottleneck hindering the translation of vitality research from theory to practice.

The innovation and contribution of this study lie in our introduction of dissipative structure theory, using “energy efficiency” as a metaphorical measure of the interaction intensity between people and spatial carriers to reinterpret the formation mechanism of street vitality. Based on this theory, we further constructed a threshold effect hypothesis model for street vitality formation. Through the conversion of valid Point of Interest (POI) data into energy efficiency metrics and a slope-based mathematical identification method, we identified balanced time periods for energy efficiency supply and demand, captured the threshold effect in vitality formation, resolved the issue of measuring the activation threshold, and determined the critical threshold value. This method transcends the intuitive mode of daily observation of urban rhythms, provides a new pathway for measuring street activation thresholds, offers an innovative solution for understanding the formation mechanism and inherent patterns of street vitality, and also supplies a theoretical basis and practical reference for the refined governance of global smart cities [9].

Apart from the introduction, the structure of this paper is as follows: Section 2 is the literature review, systematically reviewing research related to the threshold effect of street vitality. Section 3 constructs the threshold effect hypothesis model and proposes solutions to key problems. Section 4 introduces the study area, data, and research workflow. Section 5 is the results analysis, including the validation of the hypothesis model’s rationality, and the measurement and analysis of thresholds. Section 6 discusses the study’s innovation, contributions, and the accuracy of the results, and puts forward research recommendations, limitations, and prospects. Section 7 is the conclusion.

2. Literature Review

Regarding the threshold effect in the formation mechanism of street vitality, although existing research has not directly addressed this issue, related discussions can be divided into three phases according to the historical development trajectory: the qualitative analysis phase, the transitional phase, and the quantitative empirical phase.

2.1. Qualitative Analysis Phase

In the 20th century, Lewis pointed out that mixed uses could prevent periodic silence on streets by facilitating use by diverse groups at different times, thereby stimulating all-day vitality [10,11]. Jacobs argued that vitality stems from diversity; streets promote interaction among groups through functional mixture and diverse buildings, forming a sustained complex order [4]. Human-scale design further reinforces this. Gehl defined street vitality as the outcome of people-oriented spatial design, formed by promoting interaction between people and the environment and diverse social activities [5]. Whyte believed urban vitality comes from spontaneous social interactions and successful spaces can stimulate gathering and communication, creating a dynamic order [12]. Similarly, Alexander viewed the city as a complex system composed of interrelated subsystems, where the dynamic interaction between spatial structure and social behavior constitutes the source of vitality [7].

Research in this phase laid the foundation for street vitality theory, clearly stating that vitality arises from the dynamic interaction between people and the built environment. Lewis and Jacobs, among others, emphasized the importance of functional mixture and diversity; Gehl and Whyte stressed that people-oriented spatial design promotes social interaction, while Alexander regarded it as a dynamic relationship within a complex system. However, because the research methods relied heavily on subjective observation and lacked quantitative means, it was difficult to accurately measure the dynamic process of human–environment interaction and its threshold effect, which limited further theoretical deepening and application.

2.2. Transitional Phase

Maas argued that vitality stems from the synergy between commercial entertainment mechanisms and human needs [13]. Montgomery emphasized that mixed functions and open spaces stimulate diverse activities, thereby sustaining street vitality [14]. Mehta pointed out that vitality results from the combined effect of elements such as meaningful space, rational utilization, and pleasant form, which must meet users’ functional, social, and leisure needs [15].

During this phase, vitality research represented by Maas, Montgomery, and Mehta, combining qualitative observation, interviews, and quantitative questionnaire analysis, identified key factors of urban vitality and offered practical insights based on Jacobs’ diversity theory, aiding planners in developing intervention strategies that meet diverse human needs. Maas’s synergy model and Montgomery’s emphasis on physical conditions demonstrate the value of explanatory models in revealing the interaction mechanisms among variables. However, these studies mostly focused on measurable variables and were highly context-dependent, limiting the generalizability of their conclusions. More crucially, the combinations of different conditions and their thresholds for vitality formation had not been fully explored.

2.3. Empirical Verification Phase

With the application of multi-source big data such as mobile phone signaling, POIs, street view imagery (SVI), Baidu Heatmap data, and related analytical methods [16,17,18,19,20,21,22,23], the measurement of vitality has shifted from traditional surveys to high-resolution, dynamic analysis, enabling more detailed capture of human activity patterns [23,24,25]. Among these, POI data is widely used to analyze the spatial distribution of urban functions and their impact on vitality [26,27,28,29,30]. For example, Liu et al., combining mobile phone data and POI data, found that vitality exhibits significant spatiotemporal variations and is highly correlated with POI density and entropy [31]. Yue et al. quantified the relationship between mixed land use and vitality, finding that functional richness outperforms Shannon entropy [32]. Meanwhile, the impact of built environment factors on vitality has gradually become a research focus. Existing studies have shown that factors such as population density, land use, and transportation accessibility significantly affect street vitality [19,33,34]. Kang pointed out that different built environment factors affect vitality to varying degrees at different times [24]. Pan et al. found that commercial land use, floor area, and bus stop density can enhance vitality, but mixed land use may have a negative impact [35]. Liu et al. argued that high residential density might suppress vitality, while open spaces and optimized land structure can help enhance it [36]. Rui and Othengrafen further proposed that innovative street design (e.g., smart infrastructure) could improve urban mobility and livability, promoting the formation of sustainable vitality [37]. However, it should be noted that most of these studies are based on static characteristics of the urban environment, while vitality is inherently dynamic. Some scholars have identified a non-linear threshold effect between street vitality and the built environment [21,38,39,40], meaning that certain factors must reach specific critical values to cause significant changes in vitality. However, some studies have begun to question the universal effect of building density and accessibility on vitality [41,42], which may be related to differences in measurement systems or urban development stages. Therefore, comparing the relationship between the built environment and vitality in cities at different development stages will help deepen the understanding of their interaction mechanisms and provide valuable references for the construction and development of emerging cities.

Overall, big data has propelled urban vitality research into a new stage, enabling more refined measurement and factor analysis. However, the current indicator system still relies on data availability and technical pathways, lacking unified theoretical support and a standard framework, making it difficult to verify the completeness and applicability of vitality influencing factors and evaluation metrics. In particular, the effects of building density and accessibility exhibit discrepancies, necessitating multi-scale exploration of the interaction between people and the built environment to establish a universal assessment framework that supports global urban planning.

Through a systematic review of existing research, the following conclusions can be drawn: (1) Studies in the qualitative analysis phase laid the classical theoretical foundation for vitality research. However, due to the lack of quantitative methods, the threshold for vitality formation was not explored. (2) In the transitional phase, by combining qualitative observation and quantitative analysis, key physical factors of vitality were identified, promoting the translation of theory into practice, but attention to the threshold effect in vitality formation remained insufficient. (3) In the empirical verification phase, big data and advanced technologies were used to measure vitality and study its relationship with the built environment, revealing the threshold effect between vitality and built environment factors. However, the lack of a unified theoretical framework means the question of what threshold the dynamic interaction between people and spatial carriers must reach to generate vitality remains unresolved.

3. Construction of the Hypothesis Model and Resolution of Key Issues

3.1. Constructing the Hypothesis Model

3.1.1. Theoretical Foundation: Classical Vitality Theory and Dissipative Structure Theory

Jacobs viewed the city as a living organism; Lynch considered the city a support system for human life systems [6], while Alexander regarded the city as a complex system with self-organizing capabilities—an open dissipative structure [7,8,43]. Similarly, many world-renowned design masters hold analogous views. Saarinen believed that cities possess traits convergent with human life forms [44]; Kisho argued that cities exhibit metabolic characteristics unique to living organisms [45]. Therefore, one important consensus in urban cognition is that the city is a human ecosystem with a dissipative structure.

Regarding the mechanism of vitality formation, classical theories unanimously agree that vitality originates from the dynamic interaction process between people and spatial carriers (encompassing natural, economic, social, and cultural dimensions). Jacobs viewed this process as an interaction between residents and the street [4]; Gehl considered it an attraction or repulsion between people and the site [5]; Lynch saw it manifested in the outcomes of human behavior altering the physical characteristics of space [6]; and Alexander regarded it as the interaction between people and physical form through events within space [7]. This process is human-centric, its impetus stemming from the coupling of human needs with natural, economic, social, and cultural driving factors [46,47,48,49]. It exhibits an evolutionary development path, with the ultimate paradigm being the formation of a synergistic and self-regulating dissipative structure among the various factors of people and spatial carriers [50].

Although classical theories qualitatively agree that the vitality formation mechanism aligns with the human ecosystem viewpoint, they cannot quantify this abstract dynamic process, particularly the measurement of “what threshold the interaction must reach to stimulate street vitality.” However, dissipative structure theory provides an approach: Prigogine viewed the city as an ecosystem with a dissipative structure; through continuous exchange of matter and energy with the external environment, when changing conditions reach a certain threshold, self-organization phenomena occur within the system, spontaneously transitioning from disorder to macroscopic order in time, space, and function, forming a new ordered structure [51]. This precisely reveals the essential core of the vitality formation process described by classical theories.

3.1.2. Building a Theoretical Bridge: From Abstract Phenomenon to Measurability

Classical theories point in the direction, and dissipative structure theory reveals the principles, but to apply them in practice, the question of “how to measure” must be solved. In the field of urban studies, Qin et al. constructed a functional diversity measurement model based on dissipative structure and synergy theory, revealing its coupling mechanism with spatial vitality [52]. Qin et al. demonstrated the feasibility of translating this theory into operational and measurable order parameters, addressing the practical limitations of classical theories.

Therefore, using the urban ecosystem as our framework, with humans as the core and the spatial carrier system as external support, we introduce “energy efficiency” as a metaphor (not thermodynamic energy) to measure the intensity of human–space carrier interactions, specifically including the following:

• Human behavioral demand energy efficiency: reflects the intensity of demand from human activities;
• Spatial carrier supply energy efficiency: characterizes the functional response and degree of satisfaction of spatial elements to demand.

3.1.3. Hypothesis Model of Street Space Vitality Formation Mechanism

Using the continuously exchanged energy efficiency between people and spatial carriers as the measurement standard, we construct a hypothesis model of the street space vitality formation mechanism (Figure 1). The model is expressed as follows: people and street spatial carriers form a dynamic interaction structurally, triggering human behavioral demand energy efficiency (Pt) and spatial carrier supply energy efficiency (Ct); when energy efficiency fluctuations reach a certain threshold, the system achieves a dynamic balance between Pt and Ct and the street space becomes activated (Figure 2). Specifically, when the energy efficiency curves are in a parallel interval (ts–te), the system is in supply–demand balance for energy efficiency, and the energy efficiency value of Cts/Pts during this period is the threshold (T) for vitality formation.

Here, it is necessary to explain the key terms introduced:

• Energy efficiency exchange (i.e., dynamic interaction): Refers to the continuous, bidirectional flow and response of energy efficiency between human demand and spatial carrier supply, reflecting the “metabolism” of the street as an ecosystem; vitality stems from synchronous exchange rather than isolated behavior;
• Activation threshold (T): Refers to the critical point at which energy efficiency exchange intensifies to stimulate vitality, marking the system’s transition from a low-vitality state to an orderly active state;
• Balanced time period (ts–te): Refers to the time interval during which supply and demand tend to synchronize, within which the activation threshold can be observed and measured. ts and te mark the start and end moments of this balanced state, respectively.

However, to achieve threshold measurement with this model, three key problems need to be solved: first, how to obtain valid POI data; second, the conversion of phenomenological data to energy efficiency data; and third, how to identify the balanced time period of the two energy efficiency curves.

3.2. Resolution of Key Issues

3.2.1. Acquisition of Valid POI Data

The key to the hypothesis model lies in the identification of supply–demand balance and threshold judgment within energy efficiency fluctuations. However, previous studies mostly relied on static POI data [53,54], overlooking an important fact: different POI elements (e.g., shops) on the same street may be in operational or non-operational states at different times [28]; this dynamic variation directly affects the volatility of Ct. This necessitates capturing the dynamism of POI elements at the data level. However, raw POI data suffers from missing temporal information because some merchants do not provide operating hours. To compensate for this deficiency, we integrated historical business data, industry characteristics, and patterns of human demand to extract standard operating hours for POI subcategories (

Table 1. Standard operating hours for POI subcategories.

POI Subcategories	Operating Hours (Workdays)	Operating Hours (Rest Days)
Bus Stop Related, Express Bus Stop, Regular Bus Stop	05:00–23:00	06:00–23:00
Agricultural and Sideline Products Market, Fruit Market, Aquatic Products and Seafood Market, Vegetable Market	06:00–18:00	06:00–18:00
Fitness Center, Park, Ticket Selling, Bus, Ticket Change, Tourist Special Line Station, Yonghe Soy Milk, Swimming Pool, Bus Station, etc.	06:00–22:00	06:00–22:00
Subway Station	06:00–24:00	06:00–23:00
Railway Station, McDonald’s	06:00–24:00	06:00–24:00
Maxim’s, Fast Food Restaurant, Starbucks Coffee, Maxin, KFC, etc.	07:00–23:00	07:00–23:00
Schools, Primary Schools, Kindergartens	08:00–12:00 14:00–17:00	Close
World Heritage Sites, Botanical Gardens, Scenic Spots, and Historical Sites	08:00–18:00	08:00–19:00
Business Office Buildings, Industrial Buildings, Driving Schools	08:00–20:00	08:00–18:00
Ticket Office, Gynecological Hospital, Newsstand, Laundry, etc.	08:00–20:00	08:00–20:00
Internal Facilities of the School, Higher Education Institutions, etc.	08:00–22:00	09:00–21:00
Convention and Exhibition Centers, Museums, Libraries, Planetariums, Audi Museums, Indoor Booths, Exhibition Halls, Cultural Palaces, etc.	09:00–12:00 15:00–17:00	09:00–18:00
Training Institution	09:00–12:00 15:00–21:00	09:00–18:00
Memorial Hall	09:00–17:00	09:00–16:00
Media Organizations, Publishing Houses, Magazines, Radio Stations, etc.	09:00–17:00	09:00–18:00
Industrial Park, Auto Parts Sales, Saab Repair	09:00–18:00	09:00–17:00
Community Center	09:00–21:00	09:00–22:00
Related to Medicine and Healthcare	09:00–21:00	10:00–20:00
Cinema and Theater-Related, Chess and Card Rooms, Cinemas	09:00–23:00	09:00–23:00
Pedestrian Street, Characteristic Commercial Street	09:00–24:00	09:00–24:00
Chinese Restaurant, Car Club	10:00–22:00	09:00–23:00
Leisure Dining Places, Billiard Halls, Seafood Restaurants	10:00–24:00	10:00–24:00
Nightclub, Disco	20:00–24:00	20:00–24:00
Accommodation Service-Related, Convenience Stores, etc.	00:00–24:00	00:00–24:00
Funeral Facilities, Small Commodity Markets, Moving Companies, etc.	08:00–18:00	08:00–18:00
Bank of China, Traffic Management Institutions, Consumer Associations, Banks, etc.	09:00–12:00 14:00–17:00	Close
China Telecom Business Halls, Life Service Venues, etc.	09:00–18:00	09:00–18:00
Specialty Stores, Bookstores, Shopping-Related Places, etc.	10:00–22:00	10:00–22:00
Etc.	Etc.	Etc.

) and used Python 3.9 and machine learning packages to map workday and rest day time fields to corresponding POI categories, ultimately obtaining valid POI data for each moment.

3.2.2. Conversion of Phenomenological Data to Energy Efficiency Data

(1)

Necessity of Conversion

According to the hypothesis model, a city, as an ecosystem with a dissipative structure, derives its vitality from the dynamic interaction between the human life system and the spatial carrier system. This process pertains to the metabolism and exchange of energy efficiency, essentially representing the citizens’ comprehensive energy efficiency demand for multiple elements of the urban built environment [55]. A single phenomenological element in the built environment (e.g., a restaurant, cafe) cannot reflect the metabolic and exchange relationships between systems and thus cannot be directly used to measure inter-system interaction. Using “energy efficiency” as a metaphorical measure instead of element data shifts the basis of calculation from phenomenological elements to energy efficiency data, thereby revealing the essence of system interaction. Therefore, measurement results based on energy efficiency no longer reflect the manifestation of vitality but rather the degree to which the spatial carrier system meets the energy efficiency demands of the human life system. The measurement of vitality has changed from measuring the phenomenon of vitality to measuring the degree of matter and energy exchange between systems (Figure 1).

(2)

Measurement Based on Human Needs: An Overview of Maslow’s Hierarchy Theory

In the hypothesis model, people and spatial carriers are always two interacting basic systems; if energy efficiency, rather than elements, must be used as the means to measure inter-system interaction, then measurement can only be based on human needs. Maslow’s hierarchy of needs theory provides the theoretical basis for this.

Maslow’s hierarchy of needs is a psychological theory that explains human motivation through a five-level pyramid, ranging from basic survival to personal fulfillment [56]. The specific levels, from lowest to highest, are shown in Figure 3.

• Bottom-level needs: Include physiological needs (individual survival, bodily existence) and safety needs (physical safety, health security), which are the primary conditions for human survival.
• Middle-level needs: Cover belongingness and love (emotional connection, social interaction) and esteem needs (value affirmation, achievement recognition), primarily related to an individual’s emotional health.
• High-level needs: Namely self-actualization needs, involve the realization of individual potential, the manifestation of creativity, and higher-level personal growth.

(3)

Economic Reinterpretation and Mapping to POI

Maslow described human needs as a hierarchical relationship from low to high, making human needs very abstract [56] and unable to be directly linked to carrier element data, making the degree of interaction between people and spatial carriers difficult to measure. Shi et al., from an economic perspective, pointed out that Maslow’s hierarchical needs are actually a manifestation of consumption upgrading [57]. Using the extension principle of Taylor and Houthakker [58], the hierarchy of needs can be mapped to actual consumption expenditure categories, which are then linked to POI elements that support such consumption, thereby achieving a connection from abstract needs to spatial carriers. To unify dimensions, human needs are summarized into three basic energy efficiency types:

• Physical energy: Corresponds to physiological and safety needs. Mapped POI examples: restaurants, vegetable markets, bus stops, hospitals, etc.
• Emotional energy: Corresponds to esteem, belongingness, and love needs. Mapped POI examples: cinemas, casual dining venues, fitness centers, community centers, etc.
• Intellectual energy: Corresponds to self-actualization needs. Mapped POI examples: museums, libraries, training institutions, convention and exhibition centers, etc.

As shown in Figure 4, connecting lines lead from abstract human needs to objective expenditure categories, simulating that the satisfaction of a particular human need is supplied by an associated combination of multiple phenomenological elements. The measurement results no longer reflect the performance of vitality elements at the phenomenological level, but rather the diversity and mixture of the interaction between people and spatial carriers at an essential level, which reflects the innovation in vitality measurement. To simplify statistics, valid POIs are categorized into energy efficiency types based on the dominant need standard, forming the spatial carrier supply energy efficiency.

(4)

Conceptual Explanation

It must be specifically pointed out that “energy efficiency” in the hypothesis model is not a thermodynamic concept; it is a metaphor set to express the degree of interaction between people and spatial carriers, representing the general term for the functions or services that spatial carrier elements can provide to meet human needs. The study selected POI data as the primary street phenomenon data rather than others, mainly because other data (e.g., SVI data) have relatively small fluctuation amplitudes at different time points, which are almost negligible compared to the dynamic changes in POI. However, this does not deny the other values of SVI data; it is just that in the vitality formation mechanism, due to its weak volatility, SVI data has a very limited impact on the measurement of the activation threshold.

3.2.3. Identification of Balanced Time Periods for Dual Energy Efficiency Curves

The hypothesis model emphasizes that vitality originates from the dynamic interaction between people and spatial carriers. When the changes in human behavioral demand energy efficiency (Pt) and spatial carrier supply energy efficiency (Ct) reach a certain threshold, the system undergoes fluctuations, achieving a transition from disorder to order, i.e., the street space is activated. To verify and apply this hypothesis, it is necessary to precisely identify the activation threshold at which Pt and Ct reach it and the corresponding time period from continuous time-series data.

For this purpose, we propose a mathematical model for balanced time period identification. Based on changes in the time-series slope, a dynamic analysis of the similar trends and balanced state of the two energy efficiencies is conducted, extracting phases and key nodes where the system approaches equilibrium from the data, providing a mathematical basis for identifying activation points. The specific principles of the mathematical modeling are detailed in

Table 2. Mathematical modeling principles for balanced time period identification.

Procedure		Description	Formula	Calculation Method
1	Slope Calculation	Compute slopes between each pair of adjacent time points in time-series Pt and Ct	$S_{P_t}={\displaystyle \frac{P_{t+1}-P_t}{∆t}}$ $S_{C_t}={\displaystyle \frac{C_{t+1}-C_t}{∆t}}$	t = 5, 6, …, 23, where ∆t represents the temporal increment, typically set as one time point.
2	Equilibrium Detection	$\mathrm{Define} \mathrm{parallelism} \mathrm{conditions} \mathrm{for} \mathrm{two} \mathrm{slopes} S_{P_t}$ $\mathrm{and} S_{C_t}$	$\mathrm{Condition} 1:S_{P_t}·S_{C_t}>0$ $\mathrm{Condition} 2:{|S}_{P_t}-S_{C_t}|<ϵ,$ $ϵ=0.05$	$\mathsf{ϵ} = 0.05 \mathrm{was} \mathrm{determined} \mathrm{as} \mathrm{the} \mathrm{optimal} \mathrm{threshold} \mathrm{through} \mathrm{cross}-\mathrm{validation}. \mathrm{First}, \mathrm{establish} \mathrm{a} \mathrm{candidate} \mathrm{threshold} \mathrm{range}, \mathrm{e}.\mathrm{g}., \mathsf{ϵ}\in${0.01, 0.02, …, 0.1}. The threshold ϵ is adjusted to accommodate slope variations across different data scenarios. Time points t satisfying both conditions are considered candidate points for balanced segments.
3	Continuous Balanced Segment Merging	For consecutive time points {t1, t2, …, tk} meeting the conditions, merge them into a single period {ts, te}, where ts = t1 and te = tk+1.
4	Threshold Calculation	Compute the ratio during balanced time period ts	$T={\displaystyle \frac{C_{t_{\mathrm{s}}}}{P_{t_{\mathrm{s}}}}}$ Calculation proceeds when denominator ≠ 0	Denominator represents human behavioral demand energy efficiency (Pt), numerator denotes spatial carrier supply energy efficiency (Ct). Only roads exhibiting vitality (i.e., with human activities) qualify as streets; otherwise, they remain merely roads, with zero vitality rendering the denominator mathematically invalid.

.

4. Study Area, Data, and Research Workflow

4.1. Study Area

This research uses South China as its study area. This region is among China’s most dynamic areas of global impact and serves as a crucial engine for the nation’s economic development [59,60]. Cities within the area, including Guangzhou, Shenzhen, Dongguan, Foshan, Nanning, Haikou, and Sanya (Figure 5), cover various tiers of urban development.

Considering the cities’ location, tier, and the latest socioeconomic indicators from the first-year post-pandemic recovery (2023) (

Table 3. Urban location, tier classification, and socioeconomic indicators.

Location	City	City Tier/Functional Profile	GDP (CNY Billion)	Year-End Permanent Resident Population (Million People)
Guangdong–Hong Kong–Macao Greater Bay Area	Guangzhou	First-Tier City/Commercial Hub	3035.57	18.83
	Shenzhen	First-Tier City/Tech Innovation Hub	3460.64	17.79
	Dongguan	New First-Tier City/Manufacturing Base	1143.81	10.49
	Foshan	Second-Tier City/Key Manufacturing and Tech Base	1327.61	9.62
Guangxi Regional Hub	Nanning	Second-Tier City/Political-Economic-Cultural Center	546.91	8.94
Hainan Free Trade Port	Haikou	Third-Tier City/Major Port	235.84	3.00
	Sanya	Third-Tier City/International Tourism City	97.13	1.11

City tiers were determined with reference to the 2023 China City Commercial Charm Rankings, functional profiles were based on urban functions, industrial status, and economic and land use structure, while GDP and population data were obtained from the 2023 Statistical Yearbook.

), seven cities in this region were selected as samples. This offers strong referential value for cities in developing countries and facilitates the extension of research findings to other developing nations or economies with similar development levels worldwide, promoting global urban research and management.

4.2. Phenomenological Data

The primary datasets employed in this study include vector road network data, Baidu Heatmap values, and POI data (

Table 4. Summary of phenomenological data.

Phenomenological Data Type	Data Content	Interpretation and Explanation	Data Source
Baidu Heatmap Data	Heatmap data at 19 time points (05:00–23:00 UTC+8) for sampling days across four quarters	Records of human activities (work, commuting, leisure, social interactions) in spatiotemporal contexts driven by socioeconomic and environmental factors.	Baidu Maps https://huiyan.baidu.com/ (accessed on 10 April 2023–16 January 2024)
Validated POI Data	Operational POIs (e.g., restaurants, offices, malls) across 19 daily time points (05:00–23:00)	Physical/functional entities hosting human activities and influencing behavioral-energy patterns.	Amap Open Platform https://mobile.amap.com/ (accessed on 1 June 2022–31 December 2023)

).

4.2.1. Vector Road Network Data

Using Baidu Map imagery, road networks were digitized along centerlines via ArcGIS Pro 3.0.2, then segmented according to geographic position, road morphology, intersection configuration, and functional characteristics of street frontages, ultimately constructing an urban vector network comprising main roads, secondary roads, and branch roads.

4.2.2. Baidu Heatmap Data

Heatmap data for the study area was acquired based on Baidu’s spatiotemporal big data service platform “Baidu Huiyan”. The specific time range for data collection spans spring, summer, and autumn of 2023, as well as winter of 2024, with one consecutive week per quarter (including five workdays and two rest days) selected for analysis. The collected data records the frequency of Baidu SDK calls on a 200 m resolution grid, covering 19 time points per day (05:00–23:00 UTC+8).

4.2.3. POI Data

In recent years, POIs (Points of Interest) have become a crucial data source for acquiring spatial characteristic information, covering various functional locations including commercial establishments and public service facilities, and are annotated with latitude–longitude coordinates [61]. This study obtained Amap POI data through official paid channels, which include not only location data but also attributes like category, name, and business hours, providing critical support for the research.

4.3. Research Workflow

To clearly demonstrate the research framework, we developed a workflow diagram (Figure 6) comprising six key steps:

Step 1: Address the issues of acquiring valid POI data and converting phenomenological data into energy efficiency data (corresponding to Section 3.2.1 and Section 3.2.2). Using Python 3.9 and machine learning packages, map the workday and rest day operating hour fields (Table 1) to corresponding POI categories, determine their operational status at 19 time points (05:00–23:00), and thereby extract valid POI data (Table 4). On this basis, convert the valid POI data into spatial carrier supply energy efficiency data (Figure 4 and

Table 5. Summary of energy efficiency data.

Energy Efficiency Data Type	Description	Data
Human Behavioral Demand Energy Efficiency	Concentration of population behavioral activities	Heatmap values at 19 timepoints (05:00–23:00) for workdays/rest days across four seasons (spring, summer, autumn, winter)
Spatial Carrier Supply Energy Efficiency	Characterizes the functional response and degree of satisfaction of spatial elements to demand	Operational POIs with energy efficiency at 19 timepoints (05:00–23:00) for workdays/rest days

). Simultaneously, set vitality impact indicators based on classical vitality theory indices:

• Diversity (H): Namely the Shannon Diversity Index, used to measure the diversity of different categories (e.g., POI types) within an area.
• Intensity (D): The concentration density of POI elements within a certain length range.
• Evenness (E): Calculated based on the Shannon diversity index, used to measure the uniformity of the distribution of different POI categories.
• Concentration of Population Behavioral Activity (P): The spatial distribution density of population activities.

Step 2: Calculate the H, D, and E indices, then employ the entropy-weighted TOPSIS method for comprehensive quantitative analysis to obtain the spatial carrier supply energy efficiency (Ct) for the 19 time points of the sample streets. This method combines the advantages of the entropy weight method and the TOPSIS method, which can enhance the accuracy and rationality of measurement [62], and has been widely applied in studies on urban vitality and livability assessment [63,64,65,66]. Simultaneously, using the Kernel Density analysis tool in ArcGIS Pro 3.0.2, extract the concentration of population behavioral activity at 19 time points on sample days (one week per quarter). However, given that anomalies may exist at specific time points, to enhance the stability and representativeness of the results, we calculated the average concentration of population behavioral activity at each time point across sample days as the relative concentration P for each street time point, using P to represent the human behavioral demand energy efficiency (Pt); the specific calculation formula is detailed in

Table 6. Vitality impact indicators and calculation formulas.

Energy Efficiency	Index	Computational Formula	Description
Human Behavioral Demand Energy Efficiency	Relative Population Activity Concentration (P)	$P_t={\displaystyle \frac{\sum _{i=1}^d{x}_{t,i}}{d}}$	$\mathrm{Average} \mathrm{heatmap} \mathrm{values} \mathrm{across} 19 \mathrm{timepoints} \mathrm{for} \mathrm{workdays}/\mathrm{rest} \mathrm{days} \mathrm{over} \mathrm{four} \mathrm{quarters}. Pt \mathrm{denotes} \mathrm{mean} \mathrm{heatmap} \mathrm{value} \mathrm{at} \mathrm{time} \mathrm{point} t, \mathrm{where} x_{t,i}$ represents value on day i at time t, with d being total days (5 workdays, 2 rest days).
Spatial Carrier Supply Energy Efficiency	Diversity (H)	$H=-{\sum }_{i=1}^n{p}_i·\ln \left(p_i\right)$	$p_i$ $\mathrm{is} \mathrm{proportion} \mathrm{of} \mathrm{category} i (\mathrm{POI} \mathrm{count}/\mathrm{total} \mathrm{POIs}), n \mathrm{is} \mathrm{category} \mathrm{count} (n = 8), \ln (p_i$) is the natural logarithm of the proportion of category i.
	Density (D)	$D={\displaystyle \frac{N}{L}}$	$\mathrm{Using} \mathrm{linear} \mathrm{density} \mathrm{for} \mathrm{intensity}: N={\sum }_{i=1}^8{n}_i$ $(\mathrm{total} \mathrm{valid} \mathrm{POIs} \mathrm{across} 8 \mathrm{categories}), D={\displaystyle \frac{N}{L}}$ (D: intensity, L: street length in meters).
	Evenness (E)	$E={\displaystyle \frac{H}{H_{\mathrm{m}\mathrm{a}\mathrm{x}}}}$	$H_{\mathrm{m}\mathrm{a}\mathrm{x}}=\ln \left(n\right)$$(n = 8 \mathrm{active} \mathrm{categories}), \mathrm{E}: \mathrm{evenness}, \mathrm{H}: \mathrm{diversity}, {\mathrm{H}}_{\mathrm{m}\mathrm{a}\mathrm{x}}$: maximum diversity potential.
	Entropy-weighted TOPSIS Model	$C_t={\displaystyle \frac{\sqrt{{\sum }_{j=1}^n{W}_j{(Z_{ij}-Z_j^{-})}^2}}{\sqrt{{\sum }_{j=1}^n{W}_j{(Z_{ij}-Z_j^{+})}^2}+\sqrt{{\sum }_{j=1}^n{W}_j{(Z_{ij}-Z_j^{-})}^2}}}$	$Z_{ij}$$: \mathrm{Standardized} \mathrm{value} \mathrm{eliminating} \mathrm{dimensional} \mathrm{units}; W_j$$\mathrm{entropy} \mathrm{weight} \mathrm{for} j\text{-}\mathrm{th} \mathrm{indicator}; Z_j^{-}$$: \mathrm{Negative} \mathrm{ideal} \mathrm{solution} \left(\mathrm{minimum}\right); Z_j^{+}$: Positive ideal solution (maximum).

. Finally, form two curves, Pt-Tt and Ct-Tt, to analyze the dynamic evolution of the supply–demand relationship.

Step 3: At the macro level, using all streets as samples, conduct correlation analysis between Pt and Ct with SPSS software (Version 27). If significant positive correlation exists, this proves energy efficiency supply–demand balance between humans and spatial carriers, empirically validating the hypothesis model at the macro level.

Step 4: At the micro level, first sort streets in descending order based on their average daily heatmap value $\overline{P}$, obtaining the street sequence X₁, X₂, …, X_n. Then, for each street X_i, analyze the correlation (e.g., Pearson coefficient) between Pt and Ct using SPSS. This computes correlation coefficient r_i and its significance p-value for each street. If the coefficients and significance satisfy r₁ > r₂ > … > r_n and p₁ < p₂ < … < p_n, with r₁ > 0 and p₁ < 0.01, then energy efficiency supply–demand between humans and spatial carriers achieves asymptotic balance, empirically validating the hypothesis model at the micro level.

Step 5: Under the premise that the rationality of the hypothesis model has been empirically validated at both macro and micro levels, test for the existence of the threshold effect based on the mathematical model for identifying balanced time periods proposed in Section 3.1.3 (Table 2). First, perform Min-Max normalization on the energy efficiency data to eliminate dimensional effects, then proceed with calculations according to the steps outlined in Table 2.

Step 6: For the values in the approaching segment of the threshold data series, use the sliding window method and the range discrimination criterion to calculate the representative threshold. The specific steps are as follows:

(1)

Sliding Window Traversal

With window length set to 10, slide over the original sequence with step size 1, extracting each 10-length subsequence $T_i=\left\{T_i,T_{i+1},\cdots ,T_{i+10-1}\right\}$.

(2)

Range Discrimination

For each subsequence $T_i$, calculate the range $R_i$:

$$\begin{array}{c}{R}_i=max\left(T_i\right)-min\left(T_i\right)\end{array}$$

(1)

where $max\left(T_i\right)$ = maximum value in the window; $min\left(T_i\right)$ = minimum value in the window.

If $R_i<\delta$, δ = 1, the subsequence is identified as an approaching segment.

(3)

Threshold Determination

For all sliding window positions meeting the condition, mark their corresponding sequence segments as approaching phases and calculate the median threshold value of the approaching phases.

5. Results Analysis

5.1. Empirical Validation of the Street Vitality Formation Hypothesis Model at the Macro Level

Through Step 3, the correlation between Pt and Ct was analyzed at the macro level. As shown in

Table 7. Results of macro-level correlation analysis.

		Guangzhou	Shenzhen	Dongguan	Foshan	Nanning	Haikou	Sanya
Workdays	r (p)	0.593 ** (0.000)	0.538 ** (0.000)	0.432 ** (0.000)	0.419 ** (0.000)	0.566 ** (0.000)	0.598 ** (0.000)	0.640 ** (0.000)
Rest days	r (p)	0.588 ** (0.000)	0.567 ** (0.000)	0.491 ** (0.000)	0.486 ** (0.000)	0.567 ** (0.000)	0.585 ** (0.000)	0.631 ** (0.000)

**. Correlation is significant at the 0.01 level (2-tailed).

, all seven cities exhibited significant positive correlations (p < 0.01). On workdays, r-values ranged from 0.419 to 0.640, while on rest days they ranged from 0.486 to 0.631, indicating strong correlations. This suggests the energy efficiency exchange between humans and spatial carriers approaches supply–demand balance during vitality formation. The street vitality formation hypothesis model has been empirically validated at the macro level.

5.2. Empirical Validation of the Street Vitality Formation Hypothesis Model at Micro Level

Through Step 4, the correlation between Pt and Ct was analyzed at the micro level. As shown in Figure 7, with the increase in the daily average thermal value $\overline{\mathrm{P}}$ of streets, the r-value increases significantly while the p-value decreases markedly and eventually approaches 0.00, indicating that streets with higher vitality exhibit increasingly balanced energy efficiency exchange between people and spatial carriers. Moreover, in high-vitality streets, most cities exhibit r-values exceeding 0.611 (workdays) and 0.708 (rest days), with all p-values below 0.01. Guangzhou’s r-values approach 0.788 (workdays) and 0.855 (rest days), demonstrating a robust positive correlation (Figure 7 and

Table 8. Statistics of asymptotic values from micro-level correlation analysis.

		Guangzhou	Shenzhen	Dongguan	Foshan	Nanning	Haikou	Sanya
Workdays	r (p)	0.788 (0.001)	0.677 (0.005)	0.676 (0.004)	0.611 (0.005)	0.682 (0.004)	0.634 (0.006)	0.669 (0.004)
Rest days	r (p)	0.855 (0.000)	0.787 (0.001)	0.728 (0.003)	0.724 (0.002)	0.767 (0.003)	0.734 (0.002)	0.708 (0.001)

). This indicates that energy efficiency exchange between people and spatial carriers achieves supply–demand balance in high-vitality streets. The hypothesis model of street vitality formation is empirically validated at the micro level.

5.3. The Threshold Effect in Street Vitality Formation Mechanism Inevitably Exists and Is Measurable

5.3.1. Validation of the Inevitable Existence of the Threshold Effect in Street Vitality Formation Along the Time Dimension

During Step 5, when identifying the balanced time periods, the times of street activation thresholds were marked and statistically analyzed. As shown in Figure 8, the horizontal axis represents different time periods of street activation, arranged in descending order of their occurrence frequency; the primary vertical axis indicates the number of occurrences for each period, and the secondary vertical axis shows the percentage of occurrence for each period. The results show significant differences in the activation frequency of streets across different time periods, with the activation phenomenon being more concentrated and prominent in some periods.

Further analysis of the time periods with the highest street space activation frequency, as shown in

Table 9. The two most frequent activation time periods in street spaces.

Season	Day Type	Frequency Ranking	Guangzhou	Shenzhen	Dongguan	Foshan	Nanning	Haikou	Sanya
Winter	Workdays	First/ Second	12:00–13:00 18:00–19:00	12:00–13:00 18:00–19:00	18:00–19:00 12:00–13:00	18:00–19:00 12:00–13:00	12:00–13:00 18:00–19:00	12:00–13:00 07:00–08:00	12:00–13:00 17:00–18:00
	Rest days	First/ Second	18:00–19:00 08:00–09:00	18:00–19:00 08:00–09:00	18:00–19:00 21:00–22:00	18:00–19:00 21:00–22:00	18:00–19:00 21:00–22:00	08:00–09:00 18:00–19:00	21:00–22:00 18:00–19:00
Spring	Workdays	First/ Second	12:00–13:00 18:00–19:00	12:00–13:00 18:00–19:00	18:00–19:00 12:00–13:00	18:00–19:00 12:00–13:00	12:00–13:00 18:00–19:00	07:00–08:00 12:00–13:00	12:00–13:00 18:00–19:00
	Rest days	First/ Second	18:00–19:00 08:00–09:00	18:00–19:00 08:00–09:00	18:00–19:00 08:00–09:00	18:00–19:00 08:00–09:00	18:00–19:00 07:00–08:00	18:00–19:00 08:00–09:00	18:00–19:00 08:00–09:00
Summer	Workdays	First/ Second	12:00–13:00 18:00–19:00	12:00–13:00 18:00–19:00	18:00–19:00 12:00–13:00	18:00–19:00 12:00–13:00	12:00–13:00 18:00–19:00	12:00–13:00 07:00–08:00	12:00–13:00 18:00–19:00
	Rest days	First/ Second	18:00–19:00 08:00–09:00	18:00–19:00 07:00–08:00	18:00–19:00 07:00–08:00	18:00–19:00 08:00–09:00	18:00–19:00 07:00–08:00	18:00–19:00 08:00–09:00	18:00–19:00 08:00–09:00
Autumn	Workdays	First/ Second	12:00–13:00 18:00–19:00	12:00–13:00 13:00–14:00	18:00–19:00 12:00–13:00	18:00–19:00 12:00–13:00	12:00–13:00 18:00–19:00	12:00–13:00 07:00–08:00	12:00–13:00 18:00–19:00
	Rest days	First/ Second	18:00–19:00 08:00–09:00	18:00–19:00 08:00–09:00	18:00–19:00 08:00–09:00	18:00–19:00 08:00–09:00	18:00–19:00 21:00–22:00	18:00–19:00 08:00–09:00	18:00–19:00 21:00–22:00

, reveals that the times with the highest activation frequency across cities throughout the year are concentrated at 12:00–13:00 and 18:00–19:00, showing high consistency especially on spring workdays. Although the activation times in summer and autumn show slight differences, they still follow a similar trend, reflecting the high stability of street vitality formation under the city’s daily operational rhythm. This finding resonates with Lefebvre’s rhythm analysis, which posits that urban life is constituted by the interaction of rhythmic cycles and multiple temporalities [67]. This also indicates that street activation is not occurring by chance, but rather a common self-organization phenomenon of the system after reaching the energy efficiency threshold within specific time periods.

On rest days, although crowd activity patterns change and the distribution of vitality periods becomes more diversified, frequently appearing in non-working hours such as 08:00–09:00 and 21:00–22:00, 18:00–19:00 remains a high-frequency period for vitality formation in all cities, reflecting the universal pattern of crowd activity during this time. Furthermore, some cities frequently show activation in the early morning like Haikou at 07:00–09:00 and Sanya at 08:00–09:00, reflecting the influence of seasonal climate and tourist city rhythms on the timing of energy efficiency fluctuations. This indicates that although the specific timing of street activation adjusts slightly due to city attributes, climate, and social rhythms, the underlying mechanism of reaching an energy efficiency threshold triggering self-organization still holds.

5.3.2. Measurability of Thresholds in Street Vitality Formation Mechanism

Through Step 5, the activation threshold (T) for the sample streets was obtained. It should be noted that for sample streets with weak supply–demand equilibrium (e.g., streets in a disordered state throughout the day), no distinct balanced time period emerged; thus, threshold points could not be identified. However, in streets with strong supply–demand equilibrium, there are usually multiple consecutive or proximate balanced time periods corresponding to multiple threshold points. To ensure analytical consistency, we took the arithmetic mean of these proximate thresholds and recorded it as the representative threshold ($\overline{T}$) for that street.

As shown in Figure 9, with the increase in the daily average heat value ($\overline{P}$), the activation threshold demonstrates a relatively consistent trend across different cities, seasons, day types, and street types. This indicates that regardless of changes in spatial attributes, the street system commonly exhibits enhanced vitality after reaching a certain energy efficiency level, reflecting a certain self-organization regularity. It is noteworthy that the threshold trend in winter significantly differs from that in spring, summer, and autumn. Affected by the dual impact of low temperatures and the Spring Festival travel migration, the overall activity level of urban streets decreases in winter. Even streets that usually exhibit high vitality see their energy efficiency supply–demand equilibrium weakened in winter, resulting in the activation threshold showing higher instability.

The above results verify that the activation threshold proposed in our hypothesis model not only indeed exists but is also quantifiable and verifiable. Simultaneously, it shows that the formation of the threshold is significantly influenced by the external environment, such as seasonal climate fluctuations and holiday population movements; these factors alter the energy efficiency exchange of the street system, thereby affecting the stability of the threshold.

5.4. Measurement Results of Threshold Values

Through Step 5, we empirically demonstrate that the threshold for vitality formation on individual streets exists and is measurable. Building on this, Step 6 measures the threshold across four key dimensions: city tier (first-tier/new first-tier/second-tier/third-tier), seasonal variation, day type (workdays/rest days), and street type (main road/secondary road/branch road). The results indicate the following:

(1)

The threshold for street spatial vitality formation is not a single fixed value but exhibits a distinct fluctuation range of 0.40–1.56 (Figure 10);

(2)

The threshold for street spatial vitality formation exhibits volatility across city tiers, seasons, day types, and street types (Table 10). Using standard deviation (SD) to represent volatility, the specific analysis is as follows:

Figure 9. Approaching thresholds visualization for city street samples.

Figure 9. Approaching thresholds visualization for city street samples.

For first-tier cities, main road threshold fluctuation ranges between 0.07 and 0.08. Workday main roads show a mean of 0.54 (SD = 0.07), while rest days show a mean of 0.55 (SD = 0.08), indicating high stability. Concurrently, secondary roads exhibit relatively steady threshold fluctuations, with rest day SD = 0.06 and workday SD = 0.07. These values are lower than branch roads and approximate main road levels, signifying comparable stability. In contrast, branch road fluctuation is significantly higher than main/secondary roads, with an SD of 0.14, reflecting greater susceptibility to disturbance in marginal spaces. Overall, thresholds in first-tier cities are concentrated with low fluctuation, signifying system stability.

Figure 10. Activation thresholds by city tier, season, day type, and street type.

Figure 10. Activation thresholds by city tier, season, day type, and street type.

In new first-tier cities, main road threshold fluctuation ranges between 0.03 and 0.10. Rest day main roads have a relatively high mean of 0.74 (SD = 0.10), whereas workday main roads show a mean of 0.54 (SD = 0.03), significantly lower than rest days, revealing clear temporal heterogeneity. Secondary roads, however, exhibit greater stability, workday SD = 0.02 and rest day SD = 0.03, with fluctuation significantly lower than main and branch roads, demonstrating good regulatory capacity. Branch road thresholds show increased fluctuation on workdays (SD = 0.09) but become more concentrated on rest days (SD = 0.04). This indicates that while new first-tier cities are slightly less stable than first-tier cities, they still exhibit strong overall self-organizing capacity, with secondary roads showing particularly good stability.

In second-tier cities, main road threshold fluctuation widens to 0.14–0.19. Rest days show a mean of 0.71 (SD = 0.19); workdays show a mean of 0.75 (SD = 0.14). Secondary roads show rest day SD = 0.11 and workday SD = 0.08, indicating intermediate, slightly elevated fluctuation, yet still distinct from main and branch roads. Branch road thresholds exhibit smaller fluctuation, with rest day SD = 0.05 and workday SD = 0.06. This suggests declining stability in second-tier city street systems, particularly heightened sensitivity on main roads, likely due to their high traffic volume and greater susceptibility to incidents and peak travel periods, whereas branch roads serve localized areas with less fluctuation.

Third-tier cities show the highest main road threshold fluctuation: rest day SD = 0.38 and workday SD = 0.37, with maximum values reaching 1.56 (rest days) and 1.29 (workdays), significantly exceeding all road types in other cities, indicating weak regulatory capacity under extreme conditions and substantial uncertainty. Although secondary roads show lower fluctuation than main roads (workday SD = 0.08, rest day SD = 0.05), they remain somewhat unstable compared to the same street type in higher-tier cities. Branch roads also exhibit substantial fluctuation: rest day SD = 0.15 (max = 0.95); on workdays, fluctuation peaks with SD = 0.21, significantly higher than first-tier city branch roads (0.14) and second-tier city branch roads (0.06). This indicates that streets in third-tier cities are significantly affected by holidays, seasons, etc., possess limited regulatory capacity, and exhibit considerable uncertainty.

Table 10. Descriptive statistics of street activation thresholds by city tier.

	Minimum	Maximum	Mean		SD
	Statistics	Statistics	Statistics	Standard Error	Statistics
First-Tier City main road (workdays)	0.47	0.66	0.54	0.03	0.07
First-Tier City secondary road (workdays)	0.42	0.63	0.52	0.02	0.07
First-Tier City branch road (workdays)	0.42	0.76	0.57	0.05	0.14
First-Tier City main road (rest days)	0.46	0.70	0.55	0.03	0.08
First-Tier City secondary road (rest days)	0.40	0.58	0.49	0.02	0.06
First-Tier City branch road (rest days)	0.45	0.82	0.63	0.05	0.14
New First-Tier City main road (workdays)	0.50	0.58	0.54	0.02	0.03
New First-Tier City secondary road (workdays)	0.58	0.63	0.61	0.01	0.02
New First-Tier City branch road (workdays)	0.61	0.81	0.75	0.05	0.09
New First-Tier City main road (rest days)	0.67	0.88	0.74	0.05	0.10
New First-Tier City secondary road (rest days)	0.57	0.63	0.59	0.01	0.03
New First-Tier City branch road (rest days)	0.65	0.73	0.71	0.02	0.04
Second-Tier City main road (workdays)	0.60	0.98	0.75	0.05	0.14
Second-Tier City secondary road (workdays)	0.44	0.64	0.55	0.03	0.08
Second-Tier City branch road (workdays)	0.61	0.79	0.72	0.02	0.06
Second-Tier City main road (rest days)	0.53	1.09	0.71	0.07	0.19
Second-Tier City secondary road (rest days)	0.60	0.85	0.72	0.04	0.11
Second-Tier City branch road (rest days)	0.79	0.94	0.86	0.02	0.05
Third-Tier City main road (workdays)	0.42	1.29	0.80	0.13	0.37
Third-Tier City secondary road (workdays)	0.57	0.80	0.67	0.03	0.08
Third-Tier City branch road (workdays)	0.44	0.92	0.70	0.07	0.21
Third-Tier City main road (rest days)	0.51	1.56	0.87	0.13	0.38
Third-Tier City secondary road (rest days)	0.57	0.73	0.65	0.02	0.05
Third-Tier City branch road (rest days)	0.52	0.95	0.72	0.05	0.15

Table 10. Descriptive statistics of street activation thresholds by city tier.

	Minimum	Maximum	Mean		SD
	Statistics	Statistics	Statistics	Standard Error	Statistics
First-Tier City main road (workdays)	0.47	0.66	0.54	0.03	0.07
First-Tier City secondary road (workdays)	0.42	0.63	0.52	0.02	0.07
First-Tier City branch road (workdays)	0.42	0.76	0.57	0.05	0.14
First-Tier City main road (rest days)	0.46	0.70	0.55	0.03	0.08
First-Tier City secondary road (rest days)	0.40	0.58	0.49	0.02	0.06
First-Tier City branch road (rest days)	0.45	0.82	0.63	0.05	0.14
New First-Tier City main road (workdays)	0.50	0.58	0.54	0.02	0.03
New First-Tier City secondary road (workdays)	0.58	0.63	0.61	0.01	0.02
New First-Tier City branch road (workdays)	0.61	0.81	0.75	0.05	0.09
New First-Tier City main road (rest days)	0.67	0.88	0.74	0.05	0.10
New First-Tier City secondary road (rest days)	0.57	0.63	0.59	0.01	0.03
New First-Tier City branch road (rest days)	0.65	0.73	0.71	0.02	0.04
Second-Tier City main road (workdays)	0.60	0.98	0.75	0.05	0.14
Second-Tier City secondary road (workdays)	0.44	0.64	0.55	0.03	0.08
Second-Tier City branch road (workdays)	0.61	0.79	0.72	0.02	0.06
Second-Tier City main road (rest days)	0.53	1.09	0.71	0.07	0.19
Second-Tier City secondary road (rest days)	0.60	0.85	0.72	0.04	0.11
Second-Tier City branch road (rest days)	0.79	0.94	0.86	0.02	0.05
Third-Tier City main road (workdays)	0.42	1.29	0.80	0.13	0.37
Third-Tier City secondary road (workdays)	0.57	0.80	0.67	0.03	0.08
Third-Tier City branch road (workdays)	0.44	0.92	0.70	0.07	0.21
Third-Tier City main road (rest days)	0.51	1.56	0.87	0.13	0.38
Third-Tier City secondary road (rest days)	0.57	0.73	0.65	0.02	0.05
Third-Tier City branch road (rest days)	0.52	0.95	0.72	0.05	0.15

6. Discussion

6.1. Innovation and Distinctiveness of the Research

6.1.1. Innovatively Introducing Dissipative Structure Theory, Providing a New Perspective for Research

We introduce dissipative structure theory to construct a hypothesis model for vitality formation mechanisms, providing a new perspective for understanding the dynamic interactions between humans and spatial carriers in street spaces. The core contribution of this hypothesis model lies in transforming abstract and immeasurable dynamic interactions into quantifiable energy efficiency, overcoming the limitations of classical theories and innovatively resolving the measurement challenge of abstract, dynamic interactions between humans and spatial carriers. Through correlation analysis of human behavioral demand energy efficiency (Pt) and spatial carrier supply energy efficiency (Ct) in Section 5.1 and Section 5.2, the rationality of this hypothesis model is empirically validated. This elevates research on street vitality formation mechanisms from phenomenological observation to theoretical exploration of intrinsic principles. Therefore, based on this hypothesis model and using energy efficiency exchange as the quantitative benchmark, we empirically demonstrate the threshold effect in street vitality formation mechanisms, directly responding to the theoretical core of vitality formation.

6.1.2. Using a Single Street as the Analysis Unit and Basing Analysis on Valid POI Data Are Distinctive Features of This Study

Previous vitality studies mostly focused on the city or regional level [68,69], often neglecting the micro-level dynamics and humanized interactions at the street level. This study uses a single street as the analysis unit to capture the threshold of supply–demand balance between human behavior and spatial carrier elements at a micro scale, addressing the gap in existing research. Meanwhile, through comprehensive analysis of historical business data, industry characteristics, and human demand patterns, standardized processing of operating hours for street POI data was performed, obtaining valid POI data which accurately reflects the temporal volatility of spatial carrier supply.

6.1.3. The Slope-Based Mathematical Model for Identifying Balanced Time Periods May Be Another Innovation of This Study

One of the key issues of this study was how to identify the balanced time periods of energy efficiency supply and demand during the dynamic interaction between people and spatial carriers. Previous studies largely relied on methods such as machine learning (e.g., GBDT, XGBoost) [21,70] or econometric models (e.g., GWR, SDM) [17,71], which could not solve this key problem. In contrast, this study innovatively proposed a slope-based mathematical model for identifying balanced time periods, which accurately identified the balanced time periods of energy efficiency supply and demand, solved the effective capture of the threshold effect, revealed the key points of vitality formation, and provided a new method for understanding the complex dynamics of human–street space interaction; this may be another innovation of this study.

6.2. On Accuracy, Precision, and Variability of Threshold Measurements

The accuracy of activation threshold measurement needs to be discussed in the following three aspects:

Firstly, the volatility of street element data is inevitable, while the volatility of POI data is relative. Besides POI data, other data such as street view images (SVI) also exhibit some volatility; however, compared to the volatility of POI data, the volatility of street SVI data and other data is not significant and is almost negligible. This study’s use of valid POI data as the primary spatial carrier data does not mean that SVI data or other street element data do not affect the accuracy of the measurement results, rather methods to identify their volatility have not yet been found. This limitation may have some impact on precision, but it does not affect the accuracy of the research results.

Secondly, the fundamental data for threshold measurement all come from within the street scope and do not involve relevant data outside the street boundaries. However, this does not mean that factors outside the street scope have no impact on the threshold effect of street vitality formation. In recent years, the proximate spatial effects [72] and spatial spillover effects [73] of streets have been proven to exist. Therefore, the threshold effect discussed in this study is premised on internal factors dominating street vitality formation; if external factors become the dominant factors in street vitality formation, the accuracy of the threshold measurement results in this paper is uncertain.

Finally, the threshold measurement results are influenced by sample differences and urban renewal [74] and cannot serve as a universal benchmark. On one hand, the operating hours of POIs are a standardized template integrating historical records, industry characteristics, and human demand patterns, not real-time data for specific cities. While this ensures cross-scale feasibility, it may underestimate regional differences. On the other hand, the current thresholds are derived from street data of seven sample cities; if more sample cities are included, the precision of the measurement results is expected to improve further. It must also be emphasized that the data for this study were collected from South China. If future studies collect data in other regions with significantly different living demand habits (e.g., developing countries in Southeast Asia, the Middle East, etc.), the activation thresholds measured may change considerably. Nevertheless, these potential differences do not affect the research value of this paper, as the threshold effect in the vitality formation mechanism has been empirically demonstrated.

6.3. Recommendations and Implications Based on the Results

6.3.1. Urban Governance Should Pay More Attention to Medium- and Low-Vitality Areas

According to the analysis results in Section 5.1, Pt and Ct show a significant positive correlation in all sample cities, indicating a tendency toward supply–demand balance between them. And this analysis result was obtained based on including a large number of medium- and low-vitality sample streets. If the analysis focused solely on high-vitality streets, the correlation between Pt and Ct would be more significant, and there would be a stronger synergistic relationship between them. This shows that in street spaces, the dynamic interaction between people and spatial carriers moving towards synergy is an inevitable trend, and in this process, medium- and low-vitality urban areas may be more prone to generating a threshold effect. Therefore, urban governance should pay more attention to medium- and low-vitality areas, enhancing their responsiveness to human behavioral demands by optimizing the energy efficiency supply in these spaces.

6.3.2. Urban Management Needs to Focus on Specific Time Periods

The analysis results in Section 5.3.1 indicate that street vitality exhibits a distinct temporal rhythm, and the threshold effect is more likely to occur during specific time periods (e.g., 12:00–13:00 and 18:00–19:00 on workdays, and 08:00–09:00 and 21:00–22:00 on rest days), which is consistent with the urban life rhythms proposed by Smith, Hetherington [75], and Lefebvre [67]. During these critical periods, it is easier to achieve a supply–demand balance between Pt and Ct, thereby effectively enhancing urban vitality. Therefore, urban management should focus on the configuration of spatial elements and dynamic regulation during these specific time periods. For example, smart street infrastructure supported by the Internet of Things (IoT), such as citywide Wi-Fi, can be responsively regulated according to different circadian rhythms, combining efficiency improvements with enhanced livability, thereby promoting sustainable urban transformation [37].

6.3.3. Urban Development Level Influences System Resilience and Threshold Stability

The analysis results in Section 5.4 reveal that the threshold for street vitality formation is not fixed but fluctuates between 0.40 and 1.56, with the amplitude of fluctuation significantly influenced by city tier, season, day type, and street type (Figure 10). The thresholds for main and secondary roads in first-tier cities exhibit higher stability and concentration, indicating strong system regulation capability. Although the overall stability of new first-tier cities is slightly inferior to that of first-tier cities, their secondary roads show good regulatory capacity and self-organization level, demonstrating a resilient foundation in their spatial systems. The threshold volatility in second-tier cities increases further, showing higher sensitivity on main roads especially, indicating a beginning decline in their system stability. Third-tier cities, however, exhibit the most significant threshold fluctuations and uncertainty, with notably insufficient regulatory capacity, especially on high-grade streets. This indicates that cities at different development levels exhibit distinct differences in developmental resilience, and various countries and regions should formulate strategies tailored to local conditions: First-tier cities can leverage their strong system stability to promote structural spatial redevelopment. New first-tier cities can capitalize on the regulatory elasticity of their secondary roads to facilitate flexible governance. Second-tier cities need to enhance traffic management and risk control on their main roads. Third-tier cities, however, need to focus on improving regulatory mechanisms to cope with uncertainty, such as strengthening emergency response capabilities for seasonal changes or sudden traffic flows.

6.4. On the Contributions of the Research

6.4.1. The Research Findings Possess Strong Generalizability

The study area selected for this paper includes both world-class economic development engine regions—such as the Guangdong–Hong Kong–Macao Greater Bay Area—and emerging free trade ports with great development potential that are rapidly rising—like the Hainan Free Trade Port. Therefore, its research conclusions and recommendations can be generalized to other regions and cities worldwide with similar development levels. Although South China overall has high vitality, the hypothesis model based on dissipative structure theory is also applicable to low-vitality areas, albeit the thresholds in these regions might be higher; this provides a foundation for future cross-regional comparative studies.

6.4.2. Our Work May Stimulate Global Theoretical Discussions on Urban Vitality

Although the existing literature generally agrees that “vitality originates from the interaction between people and spatial carriers,” there is still a lack of consensus or even attention on the key question of “what threshold the interaction must reach to stimulate vitality.” Building upon classical vitality theories, this study reconstructed a hypothesis model via dissipative structure theory and, using valid POI data and a mathematical model for identifying balanced time periods, achieved the measurement of the threshold effect in the vitality formation mechanism; no similar research has been found in the past literature. Therefore, this study is expected to inspire theoretical discussion and empirical follow-up on the urban vitality formation mechanism globally.

6.5. Limitations

6.5.1. Non-Linear Relationships Not Considered

Our work is premised on Pt and Ct exhibiting a linear influence, and the threshold effect was verified based on this. However, existing studies have indicated that the effect of Pt on Ct also involves non-linear effects [39,76,77]. Therefore, the conclusions of this paper hold only when Pt and Ct have a linear relationship; whether they apply to non-linear scenarios requires further verification.

6.5.2. Data Granularity Needs Improvement

Although our economic mapping, based on Taylor and Houthakker’s extension principle and Maslow’s hierarchy of needs, effectively links abstract human needs to consumption expenditure categories and POI elements, it may oversimplify subtle socio-cultural dynamics and potentially underestimate unrecorded elements, such as real-time interactions or the informal economy. Furthermore, reliance on digital resources like Baidu Heatmap and Amap POI lacks sufficient temporal and qualitative resolution to fully capture these complexities, especially under unconventional conditions such as unexpected events. Future research should integrate qualitative methods and diverse data streams (e.g., Dianping, Meituan) to enrich this framework, ensuring that metaphors enhance rather than obscure the true complexity of urban vitality, remaining in line with sustainable urban theory.

6.5.3. Potential Disagreements on the Accuracy of Energy Efficiency Categorization

In the process of converting phenomenological element data into energy efficiency data, due to the abstract nature of “self-actualization” in Maslow’s theory, categorizing expenditures other than physical and emotional energy as “intellectual energy” might make this category too broad, affecting classification accuracy and consequently introducing potential bias into threshold measurement.

6.5.4. Other Research Limitations

On one hand, challenges in converting accessibility metrics into energy efficiency measurements preclude the incorporation of accessibility indicators in this study. On the other hand, the mathematical model for balanced time period identification developed in this research, is predicated on linear interactions between two variables, offering interpretability and operational utility, yet relative to more advanced non-linear modeling approaches, exhibits discernible limitations and shortcomings.

6.6. Future Outlook

Future research will advance in four aspects: First, incorporating more granular street types (e.g., commercial streets and historical streets) to explore how cultural and economic contexts moderate vitality thresholds, enhancing applicability across different urban forms. Second, expanding data sources by combining field surveys and real-time trajectories to build a dynamic monitoring system, compensating for the omission of informal activities and unexpected events, and enhancing model interpretability. Third, optimizing energy efficiency data processing to refine the classification logic of the “intellectual” dimension. Fourth, quantifying the relationship between accessibility and energy efficiency, exploring potential influencing factors, and introducing non-linear machine learning models to improve the ability to identify complex vitality evolution patterns.

7. Conclusions

Based on classical vitality theories and introducing dissipative structure theory, this study constructed a threshold effect hypothesis model for the street vitality formation mechanism. By using valid POI data to characterize the functional state of spatial carriers and performing energy efficiency conversion and employing a slope-based mathematical model to identify balanced time periods for energy efficiency supply and demand, the abstract dynamic interaction process between people and spatial carriers was quantitatively measured, thereby empirically demonstrating the threshold effect in the vitality formation mechanism. The research results indicate that street vitality relies on the dynamic balance of energy efficiency exchange between people and spatial carriers. In high-vitality streets, the correlation coefficients between Pt and Ct were r > 0.611 on workdays and r > 0.708 on rest days, both significant (p < 0.01), validating the rationality of the hypothesis model. The study successfully identified the street activation threshold (T) and the corresponding time periods; high-frequency periods were concentrated at 12:00–13:00 and 18:00–19:00 on workdays, and 08:00–09:00 and 18:00–19:00 on rest days. The threshold is not fixed but fluctuates under the influence of factors such as city tier, season, day type, and street type, with the threshold variation range based between 0.40 and 1.56. In summary, through theoretical innovation and empirical analysis, this study revealed the threshold effect on the street vitality formation mechanism, providing a new perspective and method for urban vitality research and also offering important theoretical and practical references for global urban spatial planning and sustainable development.