Spatial Vitality Assessment of Urban Post-Industrial Landscapes Using Multi-Source Data: A Case Study of Beijing Shougang Park

Abstract

Transforming the global legacy of abandoned industrial landscapes into vibrant, sustainable urban assets presents a critical yet complex opportunity, requiring solutions that simultaneously honor heritage and meet evolving urban demands. As multifunctional public spaces, their vitality significantly affects spatial quality and user engagement. We investigate the spatial vitality of post-industrial landscapes through a multi-source data framework, using Beijing’s Shougang Park as a case study. Integrating spatial syntax, point-of-interest (POI) analysis, and Baidu Heat Map data, the research constructs a comprehensive evaluation model encompassing spatial accessibility, functional diversity, heritage openness, and crowd dynamics. The findings reveal a marked spatial imbalance in accessibility, with global integration values ranging from 0.09 to 0.29 and a low intelligibility coefficient of 0.09, underscoring a mismatch between spatial structures and modern functional demands. The study identifies dynamic openness of heritage spaces and integrated community functions as key drivers for revitalization. Optimization simulations demonstrate that restructuring road networks significantly enhances spatial integration, increasing the global integration range to 0.10–0.87. This research contributes a replicable, data-driven framework for assessing and guiding the renewal of legacy industrial sites, offering valuable insights for post-industrial urban regeneration and heritage-based development.

1. Introduction

‘Post-industrial landscape’ refers to abandoned industrial sites transformed through multidisciplinary planning, design, and ecology into multifunctional spaces for environmental, cultural, and economic renewal [1,2,3]. Within sustainable urban regeneration processes, the redevelopment of vacated abandoned industrial sites facilitates the expansion of urban green public spaces [4]. As a distinctive category within these spaces, post-industrial landscapes enhance historical and cultural value through preserving and repurposing industrial remnants, thereby continuing site heritage while elevating esthetic significance and fostering local identity [5]. Consequently, cities worldwide increasingly preserve existing industrial sites, transforming them into post-industrial attractions characterized by cultural and creative branding [6,7,8].

The large-scale suburbanization of cities, urban economic recession, rising unemployment among industrial workers, permanent closures of heavy industrial plants, and the broader ecological movement have collectively driven the rapid development of post-industrial landscape parks [9]. Concurrently, topics such as sustainable urban development, climate change, and ecological restoration have garnered sustained attention in landscape architecture. As both historical objects [10] and central components of urban regeneration planning [11], post-industrial landscapes have become widely studied subjects, with global scholarly publications in this field accelerating significantly. Over the years, existing research in landscape architecture has frequently focused on its development trends and design paradigms. For instance, Li et al. [12] performed a quantitative global systematic review of post-industrial landscape regeneration to visualize and analyze international cooperation, research trends, and sustainability challenges across multiple dimensions. Zheng et al. [13] explained that urban post-industrial landscapes have unrealized ecological potential. Fu et al. [14] analyze the regional design strategy of post-industrial transformation in the Ruhr area of Germany to provide new ideas and reference for the current brownfield regeneration in China. Dogan et al. [15] explored strategies for restoring urban greening in post-industrial landscapes through nature-based solutions (NBS). Starczewski et al. [16] analyzed green spatial transformations across 32 major Polish cities, including 12 post-industrial cities, addressing a literature gap in green space research related to urban resilience.

With the increasing maturity of design frameworks and technological advances, Design strategies for post-industrial landscapes have been more widely applied in practice [17,18]. Design objectives have consequently shifted from single industrial waste land reuse to the integration of environmental, social, and economic multi-dimensional goals, as evidenced by Loures, Luís’ studies on post-industrial landscapes as drivers for urban redevelopment and public participation [11,19]. Post-industrial landscapes are no longer merely instruments for landscape transformation under industrial heritage preservation; rather, they are increasingly studied as complex, dynamic, open, multidimensional, and independent urban systems [20]. Concurrently, data surveys and fieldwork on implemented projects reveal that some completed sites fail to achieve planned outcomes, exhibit low user satisfaction with park usability, and demonstrate insufficient spatial vitality. Jiang et al. [21] identify that the Cotton 3 Creative Block’s external spaces suffer from limited vegetation diversity, monotonous greening, and scarce heritage features, weakening its attractiveness. Consequently, revitalizing diverse post-industrial landscape spaces and effectively transforming them into areas that foster urban vitality has become a central research focus. We address this need by providing clearer assessment and guidance for enhancing spatial vitality through a micro-level community perspective, specifically targeting connections between post-industrial landscapes, community spaces, and user populations.

Spatial quality serves as a key indicator of both urban development and residents’ well-being [22]. Numerous scholars have studied methods to measure and evaluate public space vitality. For instance, research has explored the relationship between street audiovisual environments and cycling behavior [23], examined how greening characteristics, spatial quality, and children’s subjective perceptions in urban park play spaces collectively affect physical activity levels [24], aiming to enhance spatial quality and revitalize these areas. Furthermore, urban green spaces (UGS) are indispensable components of cities, crucial to sustainable urban development [25], and significantly benefit residents’ physical and psychological well-being [26]. Post-industrial landscapes, as a distinct form of urban green space, promote citizens’ outdoor activities and social interactions; the vitality within their service areas reflects their influence and the quality of renewed urban spaces.

Previous research on spatial vitality evaluation predominantly focuses on social attributes of space, physical spatial forms, and field observation of crowd activity characteristics [27,28]. However, these evaluation systems exhibit subjectivity, while reliance on field methods to capture behavioral patterns inherently constrains the spatial and temporal scope of studies [29].

Concurrently, the development of urban society and information technology has generated massive multi-source urban data, enabling more systematic and scientific urban spatial research. Methods for data analysis and integration—such as spatial syntax and geographic information systems (GIS)—provide essential theoretical and technological support for urban spatial analysis [30]. Consequently, many scholars have shifted from traditional non-data-driven research approaches (e.g., field surveys, questionnaires, subjective evaluations) to big-data-driven thinking, integrating emerging fields like urban computing, artificial intelligence, and big data analytics with urban research [31]. For instance, Li et al. [32] employed GIS to map tourism spaces in Gulangyu’s historical sites and examine relationships between street network accessibility and urban structure. Similarly, Delclòs-Alió analyzed Barcelona’s urban vitality using Jane Jacobs’s theory, GIS data, and smartphone tracking experiments [33].

Spatial syntax theory, grounded in graph theory and urban morphology, examines the relationship between human activities and spatial configurations by segmenting space into lines or points and analyzing their topological and geometric relationships [34]. This theory has advanced significantly in recent years, with its application growing as a method for quantifying space and enabling more scientific investigations of spatial relationships. For instance, spatial syntax has been used to study the interaction between people and community spaces, the relationship between human emotional well-being and blue-green spaces, the impact of roads on urban structural connectivity, tourist path selection tendencies, urban spatial vitality, and the spatial relationships between culture, settlements, and social spatial structures [35,36,37,38,39,40].

Although spatial vitality assessments using multi-source data have evolved into a systematic framework within urban studies [29,31,41], existing approaches remain limited when applied to post-industrial landscapes. Conventional methods frequently overlook the fundamental tension between the preserved heritage elements in these landscapes and contemporary vitality demands. Building upon foundational spatial vitality analysis, we integrate heritage elements into its framework. This provides a targeted diagnostic approach for identifying vitality constraints during the spatial renewal of post-industrial landscapes, ultimately guiding interventions for their preservation and revitalization.

Based on the research background presented above, we address a key question aimed at advancing the systematic development of post-industrial landscapes in future urban construction: How can the spatial vitality of post-industrial landscapes be quantitatively assessed to analyze the connection between their spatial and social attributes and enhance their spatial quality?

Based on the research questions raised, using Beijing Shougang Park North as a case study, this research proposes a quantitative analysis and evaluation framework for the spatial vitality of post-industrial landscapes based on multi-source data. Through a multidimensional exploration of its spatial characteristics, we aim to uncover the economic patterns and social logic underlying different spatial forms.

2. Materials and Methods

2.1. Study Site and Service Areas

In accordance with the Beijing City Master Plan (2016–2035) and the Beijing Xishan Yongding River Cultural Belt Protection and Development Plan (2018–2035), industrial heritage preservation will be prioritized. This includes promoting the adaptive reuse of sites such as Shougang and the Beijing West Coal Mine, establishing a model for industrial heritage revitalization in western Beijing. As a nationally recognized example of successful adaptive reuse, Beijing Shougang Park’s redevelopment experience holds significant implications for industrial heritage research and practice in China [42]. Consequently, we focus on Shougang Park’s northern sector (total area: 291 ha, South District not yet completed and open to the public), located in Beijing’s Shijingshan District (Figure 1). Methodologically, we applied the distance attenuation principle using a standard pedestrian walking speed of 80 m/min. This established that facilities beyond 2400 m (i.e., a 30 min walk) exert negligible influence on the origin point [43]. Thus, the study area was delineated as follows:

• Using the park’s main entrance/exit (East Gate) as the origin;
• Defining primary boundaries via 15 and 30 min walking isochrones;
• Excluding areas with access barriers (e.g., construction zones, traffic restrictions).

This delineation represents Shougang Park’s effective service area based on pedestrian isochrone accessibility. The study area and its corresponding served residential population are detailed in

Table 1. Scope of the study and its attributes under different isochronous circles.

Definition	Area	Population
15 min	1.78 square kilometers	23,683 1
30 min	9.03 square kilometers	134,826

¹ Data from Open Router Services.

.

2.2. Data Resource

2.2.1. Urban Street Data

Urban street data were collected from OSM (open street map) [44], and combined with Baidu map to obtain the satellite map of Shijingshan District in Beijing and the current situation of the street for cross-calibration, the geometric length of the road to the drawing axis was calculated and subsequently imported into DepthmapX 0.8.0 software for analysis. DepthmapX is an open-source and multi-platform spatial analysis software for spatial networks of different scales. The software was originally developed by Alasdair Turner from the Space Syntax group as Depthmap, now open-source and available as DepthmapX.

2.2.2. POI Data

POI is also known as point of interest data. POI data is point data of real geographic entities, containing spatial and attribute information, characterized by high precision, comprehensive coverage, fast updating, and a large amount of data, etc., and it is widely used in urban research [45]. The POI point data of Beijing was obtained through the API interface of Gaode map, downloaded, pre-processed, and imported into ArcGIS10.7 software for visualization and representation.

2.2.3. Baidu Heat Map Data

Baidu Heat Map is a new big data visualization product launched by Baidu in 2011, which is based on the geographic location data of cell phone users on the LBS platform, and based on the location information carried by smartphone users when accessing Baidu products (e.g., search, maps, weather, and music, etc.), it is used to calculate the density of the crowd in each area of the clustered area and the flow of the crowd by clustering according to the location The calculation results reflect the spatial differences in the flow of people with different colors and brightness [46].

2.3. Research Methods

2.3.1. Spatial Syntax

At the urban design level, spatial syntax analysis can assess the feasibility of planning proposals and support refinements to urban plans, ensuring they align with developmental needs and objectives. We investigate spatial vitality under varying conditions through spatial accessibility analysis, structured into the following three aspects.

• Topological Accessibility—Global Integration

Topological accessibility is evaluated using the integration measure derived from the Axial Model. Integration quantifies how centrally or peripherally the space is located relative to all other spaces within the system, reflecting its potential to attract movement flows. This measure comprises both global and local integration. Local integration utilizes a defined radius parameter (R), corresponding to the spatial scale of relevant movement flows. Global integration is calculated using the following formula [47]:

$$Intergration={\displaystyle \frac{2(MD-1)}{(n-2)}}$$

(1)

In the above equation, n is the total number of axes or nodes, and MD is the mean depth value.

• Geographic accessibility—standardized angle of choice

The standardized calculation method of the spatial syntax Segment Model, representing the latest advancement in syntactic theory, enables comparative studies of data across different scales and spatial types. This method shifts model data analysis from value-based to rate-based comparison. Furthermore, it replaces the topological distance used in traditional syntactic models with angular and metric distances for scale measurement, offering a better alignment with actual spatial scales [48]. The primary syntactic parameter utilized in the present research is Normalized Angular Choice (NACH), calculated as follows:

$$Nach=\mathit{log}\left[{\displaystyle \frac{\mathit{value}\left(“\mathit{T}1024 \mathit{Choice\; Rn\; metric}”\right)+1}{\mathit{value}\left(“\mathit{T}1024 \mathit{Total\; Depth\; Rn\; metric}”\right)+3}}\right]$$

(2)

The Rn metric in the equation refers to the scale distance incorporated in the computation, while T1024 indicates a 1024-step topological radius analysis. This substantial radius value is typically employed for system-level global analysis to capture the overall structural characteristics of the entire spatial network.

• Perceived Accessibility—Intelligibility

Perceived accessibility is measured by the degree of intelligibility in the axial model, which analyzes how well residents perceive the urban road network within a certain range of space. Mathematically, synergy is understood as the correlation between local and global variables, reflecting how an individual understands the global system when experiencing a local space [34]. Synergy is positively correlated with the ability of residents to perceive the entire urban space through the local urban space; typically, synergy is calculated as the correlation R between global integration and local integration, expressed by the following formula [49]:

$$Synergy={\displaystyle \frac{{\left[\sum \left({Int}_r-\overline{{Int}_r}\right)\left({Int}_n-\overline{{Int}_n}\right)\right]}^2}{\sum {\left({Int}_r-\overline{{Int}_r}\right)}^2\sum {\left({Int}_n-\overline{{Int}_n}\right)}^2}}$$

(3)

In the formula, ${Int}_r$ and ${Int}_n$ are denoted for local integration and global integration.

2.3.2. Spatial Kernel Density Analysis

Kernel Density Estimation (KDE) in GIS quantifies feature density within radial neighborhoods to analyze spatial integration, synergy, and dispersion patterns in POI distributions. This method visually represents spatial accessibility while characterizing POI clustering through density surfaces [50]. Centered on specified point locations, KDE calculates feature density within defined bandwidths to objectively indicate aggregation or dispersion characteristics. Higher density values correspond to denser POI distributions, while lower values indicate greater dispersion [45].

2.3.3. Spatial Facility Mix Index

The Spatial Facility Mix Index (SFMI) is a measure of the distribution and diversity of different types of facilities within a region. It reflects the diversity and mix of facilities in the region, and can also be used to assess whether the region has more comprehensive service functions. The higher the entropy value, the more evenly distributed the types of facilities in the region and the higher the degree of mix [50].

$$H=-\sum _{i=1}^N\left(p_i\times \mathit{ln}\left(p_i\right)\right)$$

(4)

In the above formula, H is the facility mixing degree (entropy value), N is the total number of facility types, and $P_i$ is the proportion of the $i^{th}$ facility type relative to the region’s total facilities.

2.4. Research Framework

The current work employs Beijing Shougang Park as a case study to analyze and evaluate the spatial vitality of urban post-industrial landscapes using multi-source data. The framework comprises four components (Figure 2). Part I defines the study scope and selects the research area based on the research questions. Part II details data acquisition: retrieving road network data from OpenStreetMap, generating axial maps using CAD2018 software for spatial syntax analysis in DepthmapX, collecting POI data via the Gaode API, gathering crowd density data from Baidu Heatmap, and performing kernel density analysis in ArcGIS. Part III entails data coupling analysis to explore intrinsic relationships within spatial vitality by integrating these multi-source datasets. Part IV involves evaluation and discussion: importing all data into ArcGIS for visualization and quantitative analysis, assessing Shougang Park’s spatial vitality characteristics against established spatial metrics, and analyzing future development pathways for post-industrial urban sites.

In urban regeneration and post-industrial landscape regeneration, public space vitality serves as both a core indicator of spatial quality and a critical metric for social benefits and sustainable operation. We propose a three-dimensional measurement framework encompassing spatial accessibility, functional serviceability, and the relationship between heritage landscapes and population density (

Table 2. Evaluation dimensions of space vitality.

Dimension	Indicator	Technical Means
Spatial accessibility	Topological accessibility, Geographic accessibility, Perceived accessibility	Spatial syntax
Functional serviceability	Facility mixing degree, Industry distribution	Kernel density analysis
Heritage Landscape and Crowd Density	Degree of openness 1, conservation utilization patterns, Spatial and spatio-temporal distribution of people	Field research and Baidu Heat Map

¹ Degree of openness refers to the degree to which industrial heritage sites are open to the public, including whether admission is free, whether there are time or functional restrictions, etc.

). Within this framework, spatial accessibility serves as the foundational dimension, directly influencing individuals’ willingness and ability to reach specific spaces. Functional diversity constitutes the core attraction, determining the types of activities that occur once people arrive. Unique heritage elements, along with the ways they are utilized and made accessible, may significantly shape the spatial morphology, dynamic distribution of crowds, and the occurrence of activities within these landscapes. These three dimensions and their corresponding indicators are distinct yet interrelated. They establish a comprehensive framework for assessing the spatial vitality of post-industrial landscapes.

3. Results

3.1. Spatial Accessibility Within the Service Area of Shougang Park

3.1.1. Topological Accessibility—Global Integration Degree

Aligned with space syntax theory, we employ axial modeling to quantify Global Integration as a topological accessibility metric. This index characterizes a space’s potential to attract through-movement via standardized spatial depth calculations. Topological analysis proves more suitable than physical distance measurements for examining connectivity in community-scale networks. Depth constitutes the fundamental computational output in spatial syntax parsing (Figure 3), while integration—derived from depth normalization—measures arrival-destination potential. Higher integration values correspond to greater system-wide accessibility [51].

The spatial depth within the service area shows a trend of decreasing from the surrounding main roads to the center, and the higher the global integration degree, the lower the spatial depth. Shougang Park has a higher overall integration degree in the entire service area, and the potential of space to attract traffic varies, showing spatial differentiation characteristics (Figure 3), with the values of global integration degree distributed from 0.09 to 0.29, with an average value of 0.20. In addition to urban arterial roads, the highest values within the study area are located in Shougang Park’s Repair Shop Street (a street within the park), as well as in the areas connected with urban arterial roads, such as Gucheng West Road and Beixin’an Middle Road (Community Road 1 and Community Road 2), in addition to better accessibility to the roads around Beijing No. 11 School, another core area within the study area. The lowest values are located in the Shijingshan Park area as well as the Xin’an Urban Green Center Park area (Urban Parks 1 and 2). The results of global integration indicate that Repair Shop Street is an important street for contacting Shougang Park inside and outside the city, except for the urban arterial Beixin’an Road. Due to the opening of the rail transit M11 Winter Olympics Spur and other transportation routes in 2022, the new Shougang Station strengthens the transportation contact between Shougang Park and the surrounding communities and promotes the smooth movement of people and goods, and its connection with the surrounding transportation is gradually strengthened with the increasing accessibility of the residents to Shougang Park.

3.1.2. Geographic Accessibility—Standardized Choice

Standardized choice measures the accessibility of a space when used for transportation and is more effective for analyzing the relationship between urban street network morphology and land use, as it better aligns with actual travel patterns. Measuring angular depth at different radii and for different travel modes can evaluate space usage for various modes. Higher choice indicates a space has greater potential to attract through traffic [43,52].

Under vehicular traffic (r = 3000), the streets with the highest choice are the urban arterials around Shougang Park (Hushi Road, Beixin’an Road, and Shijingshan Road) and Qunminghu Street within the park, followed by Beixin’an Middle Road and Gucheng West Road in the Gucheng Neighborhood (Figure 4). Conversely, under pedestrian traffic (r = 300), the streets with the highest choice are Hushi Road, Beixin’an Road, and Shijingshan Road in Shougang Park. Besides overlapping with some high-choice vehicular streets, the street exhibiting the highest local choice for pedestrians is South Ergaolu Road, followed by South Wuyi Theater Road, and parts of the community roads articulated by Beixin’an Road.

According to spatial syntax theory, street choice is closely related to its topological connectivity within the global or local network. Shougang Park’s arterial streets with high choice under vehicular traffic (r = 3000) exhibit a high degree of integration, demonstrated by their multi-level connections to the urban backbone network. These streets become the optimal topological paths for traversing vehicular traffic by integrating the large-radius network to form trunk routes within the area. Streets with high choice under pedestrian traffic (r = 300) (e.g., South Ergaolu Road) rely more on localized visual accessibility. Short-radius choice analysis shows these streets are directly connected to neighborhood entrances and public facilities via a dense network of side streets, creating a visually permeable system that meets pedestrians’ need for intuitive wayfinding.

3.1.3. Perceived Accessibility—Intelligibility

Intelligibility describes the degree to which individuals perceive the global spatial configuration based on local spatial experience. This relationship is quantified by the linear regression equation y = ax + b (where a is the coefficient and b is a constant), visualized in a scatter plot with local connectivity on the x-axis and global integration on the y-axis. An R² > 0.5 indicates a well-organized, easily navigable space with relatively high Intelligibility. Conversely, R² < 0.5 signifies a weakly comprehensible spatial system characterized by poor connectivity between individual spaces and their surroundings, hindering overall spatial understanding [53].

Figure 5 shows the low linear regression fit (Intelligibility R² = 0.09), indicating very weak spatial legibility within Shougang Park’s service area. This makes it difficult for people to infer the global spatial organization from local experience, increasing disorientation risk. Prevalent street closures within Shougang Park have resulted in an irregular overall street distribution, preventing local spaces from integrating effectively into the broader spatial system. Consequently, streets in the study area exhibit poor wayfinding potential and low recognizability, impeding neighborhood integration and development.

3.2. Functional Serviceability Within the Service Area of Shougang Park

3.2.1. Mix of Facilities Within the Service Area

Point-of-interest (POI) data for Beijing’s Shijingshan District in 2021 were obtained by accessing the Gaode Map Open Platform API. The original Gaode POI classification system is complex, comprising 23 primary, 267 secondary, and 869 tertiary categories. Adopting Professor Long Ying’s POI classification framework and accounting for Beijing’s urban development characteristics [54], we reclassified the captured data into three major categories: commercial shopping, commuting office, and service facilities (

Table 3. Classified statistics of POI data within the service area of Shougang Park.

Primary Category	Sub-Category	Specific Types	Count	Sub-Total
Commercial Facilities	Catering Services	Chinese restaurants, International restaurants, Fast-food restaurants, Cafes, Tea houses, Juice bars, Bakeries, Dessert shops, etc.	392	950
	Shopping Services	Shopping malls, Convenience stores, Electronics markets, Supermarkets, Flower and bird markets, etc.	532
	Accommodation Services	Hotels, Tourist lodges	26
Commuting and Office	Business and Residential	Industrial parks, Office buildings, Residential compounds	155	451
	Corporate Enterprises	Companies, Factories	296
Service Facilities	Sports and Recreation	Sports venues, Entertainment venues, Leisure facilities, Theaters, etc.	91	324
	Daily Life Services	Shared facilities, Travel agencies, Public toilets, Newsstands, etc.	216
	Scenic Spots	Tourist attractions, Parks, Plazas	17
Grand Total				1725

).

The functional business types within the service area of Shougang Park are diverse, covering a wide range of fields such as business offices, science and technology industry, sports industry, culture and tourism, and commercial support. The mutual integration and synergistic development of these sectors not only enhances the comprehensive competitiveness of the region but also makes Shougang Park gradually develop into an exemplary case of urban rejuvenation. Table 2 summarizes the number of different POI types within the service area of Shougang Park. The findings suggest that the highest number of POIs is for shopping services, with a total of 532 POIs; followed by catering services and living services, with 392 and 296 POIs, respectively, and the lowest number of POIs is for accommodation services, with only 26 POIs. This shows that commercial shopping has become the main daily activity of the residents in the region.

In the mixing degree calculation of the business within the service scope of Shougang Park, the entropy value is 1.74, indicating that the functional mixing degree of facilities in the region is at a medium level, and there exists a certain degree of diversity in the distribution of facility types, but there are also some facility types occupying a large proportion of the total number of facilities, and certain functional types, such as food and beverage services (0.337) and service facilities (0.227), occupying a relatively large proportion of the total number of facilities. The functions of the facilities within the service area of Shougang Park are relatively more concentrated, and there is a certain trend of homogenization, especially in terms of fewer facilities for recreation, entertainment, and offices.

3.2.2. Distribution of Businesses in the Service Area

Spatial kernel density analysis reveals distinct spatial distribution patterns among different commercial formats within the study area, with their spatial characteristics significantly correlated with regional functional transformation, transportation networks, and policy orientation. Analysis of Points of Interest (POIs) within the service area shows that commercial formats exhibit an overall disparate distribution pattern (Figure 6). Spatially, commercial service points extend linearly along major streets. Peak density zones demonstrate significant spatial coupling with subway stations and arterial road networks, underscoring the fundamental role of transportation accessibility in driving commercial location choices. Concurrently, the CHANG’AN MILLS, developed within Shougang Park through industrial heritage renovation, has evolved into a regional commercial cluster center. This highlights the value-added commercial effects generated by the functional repurposing of industrial relics within the context of urban regeneration.

Commuter-oriented office formats, in contrast, display a distribution pattern characterized by peripheral concentration and internal sparsity (Figure 7). High-density office clusters are primarily located within mature business districts surrounding Shougang Park. Their spatial selection reflects a tripartite driver logic: (1) accessibility benefits derived from proximity to arterial roads and public transit nodes; (2) policy-driven development of office building clusters; and (3) market demand-driven functional synergy, creating complementary relationships with commercial facilities. Within Shougang Park, however, office development remains relatively weak, constrained by the spatial characteristics of the industrial heritage and their alignment with current functional positioning. Existing planned office zones also face challenges, such as high conversion costs and fragmented transportation networks, leading to their current vacant status. The current Regulatory Plan for Shougang Park’s North Zone allocates 62% of land to commercial/cultural tourism uses, while office land comprises only 18%, with Floor Area Ratios (FAR) capped between 1.5 and 2.0. This creates a policy gap compared to the surrounding business districts (FAR 4.0+). Consequently, the park’s current development focus leans more heavily towards fostering commercial and cultural–tourism functions derived from the adaptive reuse of its industrial heritage.

Service facilities exhibit a multi-nuclei distribution pattern (Figure 8). Community-level facilities are distributed based on the community life circle principle, with high coverage of essential daily service facilities such as convenience stores, shared equipment, and public toilets. Core high-density areas within the park concentrate around industrial heritage revitalization nodes—such as blast furnaces and silos—demonstrating the integration of heritage preservation and public services. Tourism service facilities diverge from traditional single-core clustering, forming distinctive service clusters in the northeastern area (blast furnace complex) and southwestern area (ski jump platform-silo complex). Through spatially integrated units linking attractions and services, these nodes achieve value co-creation between service facilities and heritage landscapes.

3.3. The Coupling of Crowd Activity and Heritage Space in Shougang Park

As the most identifiable spatial element in the transformation of post-industrial landscapes, industrial heritage not only carries the historical memory of the site’s industrial civilization but also constitutes the spatial substrate of contemporary urban regeneration. It can shape differentiated attractiveness depending on varying spatial reconfigurations and degrees of openness of the industrial heritage. Based on this, the scope of this phase of the study is narrowed down to examine the relationship between heritage space and crowd distribution characteristics within Shougang Park.

3.3.1. Temporal and Spatial Characteristics of Crowd Distribution in Shougang

In urban life, the most significant difference in population activities is primarily reflected in regular activities (commuting) and free activities (leisure time). Therefore, this paper analyzes eight time slots from 8:00 to 20:00 on 6 April 2025 (a rest day) and 9 April 2025 (a weekday) in the northern area of Shougang Park, Shijingshan District, Beijing, with updates every 4 h. Using Baidu Heatmap data (updated every 4 h and synchronized with Baidu Map) as the source, a total of eight heatmaps were obtained (four maps on 6 April and four on 9 April; Figure 9).

The spatial proportion of population gathering areas is small in the morning but larger in the afternoon and evening. The significantly smaller area weights of high- and sub-heat zones compared to low-heat zones indicate relatively limited locations where large numbers of people concentrate. Population aggregation exhibits obvious fluctuating changes; this characteristic applies to high-heat, sub-heat, and low-heat zones alike. Overall activity within Shougang Park on weekdays is low. High-heat zones appear near transportation stations and office areas due to commuting, while heritage attractions serve primarily as background spaces, with their types and openness exerting a weak influence on population distribution. Combined with site research, the following occurred:

• At 8:00 a.m.: Commuters linger briefly; activity levels are low and dispersed. The main high-activity area is near the park’s subway station.
• At 12:00 p.m.: During lunch breaks, neighboring office workers enter commercial areas or Shougang Winter Olympic Plaza for short rests and meals, forming small sub-heat zones.
• From 16:00 to 20:00: A small number of post-work employees or residents engage in walking and leisure activities. Park activity levels rise slightly but remain significantly lower than on rest days, concentrated near the Six Workers’ Exchange and commercial/entertainment venues like pubs and bars.

On rest days, population activities are freer. Activity levels rise throughout the morning, contrasting sharply with the concentrated commuting patterns observed on weekdays. Activity also persists significantly longer, with notable evening extension, reflecting high levels of nighttime activity. Combined with site research:

• At 8:00 a.m.: Crowds in the park are sparse, with only a few people near entrance roads, likely morning runners or residents from surrounding communities.
• At 12:00 p.m.: Midday crowds begin gathering as tourists enter core attractions for sightseeing; activity levels gradually rise.
• At 16:00: Activity peaks, driven by industrial tourism and cultural activities, attract large crowds. High-activity areas concentrate around the No. 2 and No. 3 blast furnaces and viewing platforms on the aerial walkway.
• At 20:00: Crowd density begins decreasing. Activity near the Shangri-La Hotel and subway station rises due to tourist accommodation and return travel needs.

3.3.2. Relationship Between Heritage Openness and Crowd Viability

The influence of the degree of openness and the mode of conservation, utilization, and development of heritage space in Shougang Park on crowd distribution profoundly reflects the pattern of function-driven gathering and openness shaping spatial activity intensity. Completely open nodes—spaces accessible without payment (e.g., CHANG’AN MALL, Qunming Lake, and Xiuchi)—sustain baseline spatial activity through their low-threshold public accessibility. CHANG’AN MALL, as a commercial space, consistently functions as a high-activity zone in Shougang Park on both weekdays and rest days. In contrast, Qunming Lake and Xiuchi, limited by their singular functions as pure landscape or transportation spaces, only form scattered low-activity zones. Partially open nodes—spaces requiring partial payment or partially restricting visitor access (e.g., the three blast furnaces, the aerial walkway)—rely on functionally composite strategies (exhibition + viewing + activities) and visitor flow limitation to generate high-activity zones at specific times on rest days. Their activity intensity is constrained by the concentration of opening hours. Restricted-open nodes—spaces not yet publicly accessible or opened only for special festivals/activities (e.g., the four blast furnaces)—leverage the driving force of events such as special exhibitions to briefly activate high activity levels. Unopened nodes—spaces not yet utilized for protection and development (e.g., No. 1 Blast Furnace, Cooling Towers, and Shijingshan Park)—remain completely disconnected from crowd activity networks due to strict preservation protocols and lack of spatial functionality, becoming “heritage islands.”

This gradient pattern of layered openness reveals that excessive enclosure substantially reduces spatial vitality, while single-function design restricts the extent of activity zones. Conversely, dynamic openness and composite utilization bridge heritage value with crowd needs.

4. Discussion

The foregoing analysis employs multi-source big data to evaluate spatial vitality in post-industrial landscapes and examines its relationship with various assessment dimensions. The following discussion addresses the operational mechanisms of road network topology optimization, functional integration, and heritage planning to propose recommendations for future planning, construction, and management of post-industrial landscapes.

4.1. Road Network Topology Optimization Activates Spatial Potential in Post-Industrial Landscapes

Analysis of street accessibility and choice values in Shougang Park reveals that these indicators are associated with surrounding transportation and road structures, consistent with previous research findings on spatial connectivity. For example, a study of neighborhood parks in Riyadh discussed how community park road closure patterns can be integrated with the surrounding community environment and provide different solutions [55]. The spatial differentiation of normalized choice values essentially results from the synergy between urban functional organization and transportation network topology. High-choice vehicular corridors reflect efficiency-prioritized mobility, while high-choice pedestrian streets respond to human activity patterns. Their partial overlap (e.g., Qunminghu Avenue) demonstrates the resilience of multifunctional streets in stock regeneration—achieving compatibility between vehicular efficiency and pedestrian vitality through spatial reconfiguration. Actually, various optimization schemes have been summarized in studies on the spatial layout and vitality of roads. When designing street space renovations, priority should be given to compact street spaces, and cultural and recreational vitality is most adaptable to street space layouts [56]. Moreover, the built environment of streets is directly proportional to street vitality [57]. Based on this, Future planning should develop differentiated design strategies based on choice-value characteristics at various radii: enhance multimodal transfer nodes along motorways, while focusing on micro-space quality improvements within pedestrian networks.

In the context of post-industrial landscape transformation and urban regeneration, the optimization of the transportation network of post-industrial landscapes needs to take into account the needs of historical preservation and modern functions, without damaging the natural ecology. The addition of numerous main roads and side roads and the opening of broken roads aim to improve the system of combining post-industrial landscape with urban road networks, enhance spatial operational efficiency, increase the accessibility and coverage of the road network system in post-industrial landscape, and activate the potential of post-industrial landscape in urban space. In order to verify the rationality of the optimization measures, the existing closed roads and disconnected roads are optimized, and the optimized urban roads (Figure 10) are subjected to syntactic measurements. The findings suggest that the topology of the roads in the study area can be upgraded compared to the status quo, and the global integration values are distributed in the range of 0.10 to 0.87 (higher than 0.09–0.29 before optimization). The average integration is 0.21 (higher than 0.20 before optimization).

4.2. Functional Diversification as a Catalyst for Vibrancy in Post-Industrial Landscapes

As a paradigm for industrial heritage revitalization, Shougang Park demonstrates the potential and advantages of service-oriented development for post-industrial landscapes by evaluating the connections between its diverse functional layouts and surrounding communities. Yue’s research also found that service and facility POIs are strongly correlated with the vitality of park spaces [58]. People are more inclined to visit parks with convenient external services and facilities, which means that citizens pay more attention to the services provided by the surrounding environment when choosing which parks to visit. Public facilities within the park, such as sports venues and commercial services, are accessible to the community. This not only creates diverse cultural and consumption experiences for residents but also significantly enhances the regional quality of life. Concurrently, the park’s distinctive industrial culture interacts with the historical heritage of adjacent communities, fostering regional cultural exchange and establishing a model space for interaction through diverse cultural activities.

At the functional and industrial level, Shougang Park and its neighboring communities have developed a synergistic and tightly integrated relationship. Residents from surrounding areas constitute a key visitor base for the park, and their diverse needs actively drive the optimization and upgrading of its business formats. Conversely, community amenities such as education, healthcare, and commerce provide essential support for the park’s operations, creating a complementary service network. Regarding industrial synergy, Shougang Park cultivates clusters like “technology+” and “sports+”, driving the extension and upgrading of related industrial chains in nearby communities. Furthermore, hosting major sporting events and activities significantly enhances the region’s overall influence [59]. This discussion aligns with the assessment hypothesis based on user demands, namely that engagement with urban parks is influenced by people’s preferences for leisure, recreation, and diverse functional requirements [60].

Consequently, the future development of post-industrial landscapes should focus on integrating parks with cities as a core objective to achieve regional value enhancement. For urban planners and policymakers, it is important to pay attention to the integration of parks with their surrounding environments [61]. In terms of renewal strategies, emphasis should be placed on integrating post-industrial landscapes with other urban areas from multiple dimensions, including spatial integration, functional fusion, and industrial synergy. Additionally, improving infrastructure and public service facilities is crucial for transforming industrial zones into integral parts of the city, achieving resource sharing, and leveraging complementary advantages [21].

4.3. Heritage Space Revitalization Enhances the Spatial Value of Post-Industrial Landscapes

Analysis of crowd density patterns within Shougang Park indicates that the openness and usage patterns of heritage spaces have a certain impact on the space. The conservation and development strategies adopted for heritage sites directly impact the diversity and functional layout of pedestrian pathways within the area, leading to uneven spatial distribution of people. This aligns with previous issues identified regarding the quality of post-industrial landscape spaces. Within parks, important heritage spaces often have low development intensity, lack distinctive features, exhibit low spatial diversity, and have underutilized spaces with low utilization rates [21], all of which contribute to reduced spatial appeal. Field surveys assessing the accessibility levels and development models of heritage spaces reveal distinct patterns: fully accessible nodes sustain baseline crowd density; partially accessible nodes drive high-density zones through functional diversification and curated activities; restricted-access nodes rely on special events for activation; and closed nodes exhibit near-zero density.

To promote the sustainable development of heritage in post-industrial landscapes, accessible heritage spaces should leverage their inherent spatial characteristics, actively open to the public, and integrate diverse functions while continuously improving supporting facilities. Unlike the Ruhr Valley in Germany, which transformed blast furnaces into performance venues through cultural activities, the restricted areas of Shougang currently lack embedded cultural projects [62]. Shougang Park has designed a solution: transforming the former coal transportation workshop (with a ceiling height of 30 m) into an artificial intelligence metaverse experience center, leveraging the scale advantages of industrial spaces to adapt to new digital forms. Within the office area, silos are being converted from storage units into smart office clusters, creating a new working environment. The regeneration project of the Qunming Lake industrial water system involves transforming the original circulation pool into an artificial wetland, where water-purifying plants form an ecological filtration pool. Through such cultural and ecological development models, the value of post-industrial landscape spaces is enhanced during the revitalisation of heritage areas.

4.4. Limitation

The current work uses the North District of Shougang Park as a case example to propose a framework based on multi-source data for assessing spatial vitality in post-industrial landscapes across different dimensions. However, the crowd dynamics distribution based on Baidu heatmaps can only provide relative visitor density within the park. We cannot estimate the exact number of people. Additionally, the Baidu heatmap data was collected every two hours over a week in spring, and thus cannot reflect crowd dynamics distribution in other seasons or more detailed park dynamics.

Further research can build upon the methods and results of the findings to compare the relationship between crowd distribution in post-industrial landscapes across different seasons and the openness of heritage landscapes. Differences between weekdays and holidays also warrant exploration. Additionally, other potential influencing factors can be incorporated into the research framework. Adopting similar methods can help researchers better understand people’s actual behaviors and needs, ultimately driving urban planning and post-industrial landscape design.

5. Conclusions

The present research proposes a multi-dimensional framework for assessing spatial vitality in post-industrial landscapes by integrating spatial syntax, POI data, and Baidu heat maps, with Beijing’s Shougang Park as a case study. The findings reveal a clear spatial mismatch between historical structures and present-day functional demands, marked by low legibility, moderate facility mixing, and uneven crowd distribution. Crucially, the level of openness and functional integration of heritage nodes strongly influences spatial vitality. Optimization simulations demonstrate that targeted improvements in street network connectivity can enhance spatial integration, while functional diversification and heritage reuse foster user engagement and sustained vibrancy. Large-scale post-industrial landscape projects of significant influence within China’s landscape design practice have begun to garner preliminary international recognition. Particularly following the Winter Olympics, landscape architects from the United States, Germany, and Italy have started to take notice of the monumental Shougang Park project. Beyond its local context, this study offers globally relevant insights and a replicable research framework for the regeneration of post-industrial sites. The challenge of integrating industrial heritage into contemporary urban life is a shared concern for cities worldwide undergoing post-industrial landscape transformation. The proposed framework offers scalable and replicable tools for diagnosing vitality constraints and guiding spatial optimization strategies, applicable across multicultural geographical settings. It contributes to the international discourse on sustainable urban renewal while providing practical guidance for planners and policymakers globally engaged in post-industrial landscape transformation.