Culturally Sustainable Site Selection of Bazaars: A Spatial Analytics Approach in Ürümqi, Xinjiang

Abstract

This study develops a spatial-analytical framework that integrates commercial hierarchy theory with cultural sustainability principles to support the sustainable development of traditional cultural marketplaces. Using kernel density estimation and Ripley’s K function analysis of 160 bazaars and 83,127 POI data points in Ürümqi, we established a hierarchical business district classification system incorporating both cultural-demographic factors and commercial indicators. Our findings reveal that culturally attuned spatial planning generates synergistic effects between heritage conservation and contemporary development needs. The research contributes to sustainable urban theory by extending Central Place Theory through cultural dimensions while providing practical design strategies validated through 15 case studies. This framework offers urban planners an implementable revitalization approach that maintains cultural authenticity while achieving a balance between commercial vitality and social cohesion, thereby presenting an effective pathway for sustainable urban development.

1. Introduction

In an era marked by rapid urbanization and cultural homogenization, the preservation and sustainable revitalization of traditional cultural marketplaces have emerged as a pressing challenge for urban planners and policymakers worldwide. These spaces—exemplified by Xinjiang’s bazaars—function not only as economic hubs but also as vital repositories of intangible cultural heritage and social cohesion. Their decline threatens not only local economies but, more fundamentally, undermines the cultural sustainability of communities and the social resilience of urban environments. This challenge is particularly pronounced in rapidly developing regions, where the need for scientifically grounded planning methodologies has become an urgent priority in pursuit of balanced and sustainable urban growth.

Research on sustainable urban development has established a foundational understanding of the roles of commercial spaces. Studies have highlighted the importance of spatial design and ground-level integration in multi-commercial complexes for urban sustainability [1,2,3]. The composite nature of architectural space and functional mixing in high-rise complexes has been explored for enhancing spatial efficiency and creating multi-layered environments [4,5]. Furthermore, the social value of these spaces as semi-public hubs for community interaction and cohesion is widely acknowledged [6,7], alongside operational research focusing on risk management and smart technologies [8,9,10]. However, a significant gap persists in translating this broad recognition into practical, spatially explicit methodologies for culturally significant marketplaces. Seminal theories such as Central Place Theory [11,12] offer a foundational framework for understanding commercial hierarchies, yet their application to culturally specific commercial environments remains limited. This is particularly evident in the inadequate integration of cultural sustainability considerations with conventional economic models for site selection.

Furthermore, prevailing approaches to business district classification have predominantly relied on socioeconomic metrics—including floor area, commercial turnover, and tenant mix [13]. While useful for conventional retail analysis, these models frequently overlook critical socio-cultural drivers such as ethnic population distribution and intangible cultural practices, which fundamentally shape the viability of ethnic marketplaces. Although the cultural function of traditional markets is widely acknowledged in the literature, a robust quantitative methodology for integrating these cultural dimensions into spatial planning and site selection frameworks remains notably underdeveloped.

Within China’s unique development context, the “Cultural Enrichment” initiative provides an important policy framework for addressing these challenges. Preliminary research indicates that traditional cultural venues in Xinjiang face particular challenges, including accessibility limitations and outdated facilities [14]. This situation necessitates innovative approaches to cultural space planning that can bridge international theoretical frameworks with local implementation needs.

As a traditional commercial form prevalent in both urban and rural areas of Xinjiang [15,16], bazaars present distinctive advantages for sustainable cultural development, including their scale, accessibility, and socio-cultural embeddedness [17]. This study therefore aims to develop a hierarchical classification methodology for bazaar commercial circles that integrates international spatial theory with local cultural contexts. The research seeks to provide both theoretical foundations and practical references for optimizing commercial spaces while enhancing cultural vitality in traditional market settings [18,19].

Bazaars in Xinjiang represent distinctive socio-commercial entities that differ functionally from conventional marketplaces. Contemporary bazaars can be categorized into three primary types based on scale characteristics (Figure 1), fulfilling multiple social roles including commercial exchange, cultural transmission, and community interaction [16]. A comparative analysis revealing distinct functional differences between Xinjiang bazaars and various inland commercial formats (marketplaces, trade markets, and commercial complexes) is provided in

Table 1. Comparison of Xinjiang Bazaar, Market, Trade Market, and Commercial Complex.

	Similarities	Differences
Bazaars of all sizes	The basic functions are the same	Partially incorporates religious functions	Cultural activities prefer outdoor open spaces	Regional style design as the main focus	Operators are predominantly ethnic minorities
Municipal Trade Market Commercial Complex		rarely incorporates religious functions	Cultural activities prefer indoor open spaces	Modernist style design as the main theme	The operators are predominantly Han Chinese.

Source: Author’s own illustration.

. However, like many traditional commercial forms globally, they face significant challenges from e-commerce competition and urbanization pressures [20], with previous revitalization efforts achieving limited success [16,21].

The scientific site selection of bazaars represents a complex multidisciplinary challenge integrating urban planning, cultural economics, and spatial analysis. Its significance extends across four key dimensions:

(1)

Social Integration Value: Strategic positioning can enhance inter-ethnic interaction and community cohesion through improved accessibility and spatial arrangement [22].

(2)

Cultural Sustainability: Appropriate location selection strengthens the preservation and transmission of intangible cultural heritage within urban environments [23].

(3)

Economic Development: Scientifically chosen locations can stimulate regional economic activity, generating employment and supporting related industries [19,23].

(4)

Urban Planning Value: Integrated site selection contributes to improved urban functionality and governance efficiency [24].

While this study is grounded in the distinctive socio-spatial context of Xinjiang, it positions itself within three intersecting domains of international scholarly discourse. Methodologically, it advances spatial analytics through the development of a hierarchical classification framework that systematically integrates cultural-demographic variables with commercial clustering patterns—extending Central Place Theory into culturally complex urban settings. Theoretically, it engages with global literature on heritage marketplaces and informal economies by reconceptualizing bazaars not merely as vestiges of tradition but as dynamic socio-spatial entities shaped by both cultural practices and contemporary urban forces. Strategically, it contributes to urban design and cultural-economic policy by offering a scalable model for the placement and programming of culturally sensitive commercial spaces. By situating the empirical findings within these broader conversations, this research aims not only to address a regional planning challenge but also to offer a transferable conceptual lens through which similar marketplaces worldwide—from the bazaars of Istanbul to the souks of North Africa—can be re-examined and sustainably revitalized.

Despite this significance, research on site selection methodologies for Xinjiang’s bazaars remains notably underdeveloped [14]. This study therefore contributes to sustainable urban planning by developing a novel methodology that transcends conventional preservationist approaches. We propose a dynamic model that positions cultural marketplaces for sustained commercial viability while safeguarding their socio-cultural distinctiveness. The resulting framework offers a transferable approach for sustaining culturally significant marketplaces globally, demonstrating how strategic spatial planning can enable traditional economies to thrive within contemporary urban contexts—a crucial advancement toward genuinely sustainable urban development.

2. Materials and Methods

Bazaars in Xinjiang are typically situated within local business districts. In rural areas and small cities, these districts can be easily identified through basic observation. Bazaars in these contexts are predominantly community-level, characterized by their small scale and limited inclusion of cultural-commercial spaces. In larger cities, however, the positioning of a bazaar varies significantly based on the hierarchy of the business district in which it is located, making accurate assessment more challenging. Drawing on existing research and incorporating distinctive features of Xinjiang bazaars, this study proposes a methodology for their site selection and positioning [25,26,27,28].

This section details the data foundation and analytical framework of this study. First, we clarify the operational definitions of core concepts (Section 2.1). Second, we describe the sources and preprocessing procedures of the research data (Section 2.2). Finally, we systematically present the spatial analysis methods and the process of constructing the site selection model (Section 2.3). This coherent methodological chain ensures the reproducibility and scientific rigor of the research.

2.1. Conceptual Framework

2.1.1. Conceptual Definition: Bazaar Commercial District

To facilitate quantitative analysis, this study requires an operational definition of the core concept, the “Bazaar Commercial District.” International academic research on commercial center hierarchy and business districts is well-established. Christaller’s Central Place Theory laid the cornerstone for this field. Subsequent extensive studies have classified urban business district hierarchies based on indicators such as the quantity, variety, and formats of commercial facilities.

However, a fundamental distinction exists between conventional urban business districts and the Bazaar Commercial District. This study operationally defines the “Bazaar Commercial District” as: a geographical area delineated by overlaying the socio-cultural dimension of a relatively high concentration of ethnic minority populations (set in this study as >30%) onto the spatial extent of a traditional urban business district. This definition inherits the spatial-economic core of Central Place Theory while incorporating the essential characteristics of Xinjiang bazaars, where ethnic minority vendors dominate and serve specific cultural communities. It provides a clear conceptual boundary for the spatial analysis in this study [29,30,31].

2.1.2. Hierarchical Coupling Mechanisms

The “Hierarchical Coupling Mechanisms” refer to the integrated analytical framework that links different tiers of commercial districts with corresponding bazaar scales, functions, and service orientations. This mechanism operates on two interrelated levels:

Spatial-Economic Hierarchy: Derived from kernel density analysis and standard deviational ellipses, commercial districts are classified into tiers (e.g., Tier-1, Tier-2, Tier-3) based on their commercial intensity and spatial extent.

Socio-Functional Coupling: Each commercial tier is coupled with a recommended bazaar scale (e.g., city-level, regional-level, community-level) and a specific set of commercial and cultural functions. This coupling is informed by the ethnic population distribution and is designed to align the bazaar’s service orientation with the spatial and demographic characteristics of its host business district.

This dual mechanism ensures that bazaar site selection is not only based on commercial viability but also responsive to the socio-cultural fabric of the area, facilitating targeted and sustainable development.

In this study, “cultural attributes” are operationalized primarily through the spatial distribution of ethnic minority populations, using a threshold of >30% to define a “Bazaar Commercial District.” This threshold was informed by existing research on ethnic settlement patterns in the Ürümqi region and validated through preliminary case comparisons that confirmed its effectiveness in identifying commercial areas with distinctive cultural characteristics. We recognize that cultural vitality is a multidimensional construct encompassing—in addition to demographic composition—factors such as language use, festive activities, and consumption practices, which are not fully quantifiable. Nevertheless, within the constraints of available data, the proportion of ethnic minority residents serves as a meaningful proxy for cultural identity and consumer preference, effectively reflecting the spatial concentration of cultural communities and providing an operational criterion for identifying culturally significant commercial zones. Future studies could incorporate field surveys, interviews, or social media data to further refine the representation of cultural dimensions.

2.2. Data Sources and Preprocessing

2.2.1. Data Collection

The data foundation of this study stems from 83,127 commercial Points of Interest (POI) data points collected via the China map API for Ürümqi’s seven districts and one county in March 2022. Simultaneously, to define the bazaar commercial districts, we gathered data on the ethnic composition of various districts and counties from the Ürümqi Statistical Yearbook (2019).

2.2.2. Data Categorization and Processing

The raw POI data underwent rigorous preprocessing in the ArcGIS (10.8) platform, including coordinate correction, deduplication, and projection conversion (to WGS 1984 UTM Zone 44N). Subsequently, based on the national standard industry classification and the research objectives, all POIs were categorized into five major types for subsequent analysis: (1) Culinary Delights; (2) Shopping and Consumption; (3) Financial Institutions; (4) Hotel Accommodations; (5) Lifestyle Services

The categorization of POIs, particularly within the ‘Culinary Delights’ class, was based on the standard industrial classification system and prevailing market conventions in China. Key distinctions were made based on the primary business model—specifically, whether establishments primarily provide full sit-down meals (‘Chinese Cuisine’) versus quick-service, light meals and portable food items (‘Snacks and Fast Food’).

2.2.3. Data Quality Assessment and Limitations

While the POI data employed in this study provide a substantial foundation for commercial spatial analysis, we acknowledge potential biases inherent to the data source. POIs obtained via the Amap API primarily reflect registered business entities and platform-listed establishments, which may underrepresent informal or temporary vending activities, such as street vendors or seasonal markets. To assess the impact of such biases, we cross-referenced the POI data with the official business registry from the Ürümqi Statistical Yearbook and conducted field sampling surveys in areas with low POI coverage. Although the POI categorization follows the national standard industrial classification, certain nuances in business types may not be fully captured. To mitigate this, we introduced finer sub-categories within the “Culinary Delights” class, differentiating, for instance, between full-service restaurants and quick-service food vendors based on their primary business model. Despite these limitations, the POI dataset demonstrates strong representativeness in terms of spatial coverage, resolution, and classification consistency, thereby providing a robust basis for macro-scale identification of urban commercial structure.

2.3. Analytical Framework and Modeling

This study adopts a spatial analysis framework that progresses from macro to micro levels, sequentially identifying commercial agglomeration patterns, delineating business district boundaries, and finally constructing the bazaar site selection model.

2.3.1. Spatial Agglomeration Analysis

(1)

Kernel Density Analysis Estimation:

Kernel Density Analysis Estimation was employed to transform the discrete commercial POIs into a continuous density surface, thereby quantifying and visualizing their spatial agglomeration patterns. This method assigns an “area of influence” around each point, defined by a kernel function and a critical search radius. After testing radii from 200 m to 1200 m, a 300 m bandwidth was selected to optimally capture commercial clustering at the neighborhood scale. The resultant density surface functioned as a “commercial vitality detector,” clearly identifying urban hotspots. Subsequently, this surface was treated as a probability distribution, and the contours corresponding to one, two, and three standard deviations were extracted to scientifically delineate the boundaries of hierarchical commercial districts. This data-driven approach provided an objective and robust spatial foundation for the subsequent bazaar site selection model, moving beyond arbitrary administrative divisions or buffer zones.

Kernel Density Analysis (KDA) [32,33,34] estimates the spatial agglomeration of features by calculating density values from surrounding point data. The mathematical formulation is as follows:

$$K(s)={\sum }_{i=1}^n{\displaystyle \frac{1}{\pi r^2}}\Phi ({\displaystyle \frac{d_{\mathrm{is}}}{r}})$$

(1)

Here, K(s) denotes the kernel density value at point s; i represents the i-th point; n indicates the total number of samples; r signifies the kernel density search radius; and dis denotes the distance from the i-th point to points.

(2)

Ripley’s K Function:

A multi-distance spatial cluster analysis used to quantitatively test the statistical significance of POI clustering at different spatial scales. By comparing the difference (Diff K) between the Observed K and the Expected K values, we confirmed that the commercial POIs in Ürümqi exhibit a significant clustered pattern across all studied scales.

(3)

Data Standardization Methods:

The standard deviation method was applied to delineate commercial district boundaries from the kernel density surface. While spatial data rarely follows a strict normal distribution, this approach serves as an effective heuristic thresholding tool for identifying significant commercial agglomerations. The kernel density surface of Ürümqi’s commercial POIs exhibits a strong central tendency with predictable density decay, making standard deviation thresholds meaningful for capturing core commercial clusters. The robustness of this approach was verified through a supplementary quantile-based classification, which yielded consistent spatial patterns. Empirical validation from 15 case studies further confirms the practical relevance of the identified boundaries.

This process standardizes the measurement scales of all indicators, eliminating disparities in numerical magnitude to facilitate consistent comparison and evaluation under a unified benchmark. The specific formula is as follows:

$$X_n={\displaystyle \frac{x-x_{min}}{x_{max}-x_{min}}}$$

(2)

where Xn denotes the normalized value, Xmin represents the minimum value within the indicator category, and Xmax represents the maximum value.

(4)

Weighted Integrated Scoring Method:

By assigning different weights to various metrics, multiple indicators are synthesized into a composite score for comparing, ranking, or evaluating different objects. The specific formula is presented below:

$$SGi={\sum }_{j=1}^n\left(x_{ij}{w}_j\right)$$

(3)

where SGi denotes the composite indicator value of the i-th commercial district, Wj represents the weight assigned to the j-th indicator, i indicates the total number of commercial districts, and j refers to the number of indicators.

2.3.2. Delineation of Commercial District Boundaries

Drawing on the concept of the Standard Deviational Ellipse in statistics, we utilized the continuous surface generated by KDE and selected specific density thresholds to delineate commercial district boundaries. Specifically, we extracted the kernel density contours corresponding to one, two, and three standard deviations, using them as the spatial boundaries for different grades of commercial districts (see the Section 3). This method effectively captures the main direction and extent of commercial distribution.

2.3.3. Bazaar Site Selection Model

Based on the graded business districts derived from the above analysis and incorporating the ethnic minority population distribution data defined in Section 2.1, we finally propose a Hierarchical Site Selection Model. This model provides positioning and scale guidance for bazaar development within business districts of different tiers (

Table 2. Hierarchical Site Selection Model for Xinjiang Bazaars.

Business District Tier	Recommended Bazaar Scale and Positioning	Primary Service Orientation
Tier-1	“City-level”	Serves the entire urban population and even surrounding cities; integrates diverse cultural-commercial functions and activates the nighttime economy.
Tier-2	“Regional-level” (primary), with “City-level” as supplementary	Serves residents of a specific urban region; features sizable retail and entertainment formats.
Tier-3	“Regional-level” (primary), with “Community-level” as supplementary	Serves a specific urban region and surrounding communities; dominated by food, beverage, and retail formats.
Non-Bazaar District	“Community-level”	Serves immediate neighborhood residents with essential commercial activities and a limited mix of formats or community services.

Source: Author’s own illustration.

).

This study provides a detailed discussion of Table 2.

(1)

Site Selection in Tier-1 Bazaar Business Districts When a bazaar is situated within a city’s primary business district, it is recommended to prioritize a “city-level” scale and positioning. In addition to their large scale, these bazaars typically integrate diverse cultural-commercial spaces and public service functions. They play a significant role in activating the nighttime economy and hold distinctive appeal for residents. The service orientation of their cultural-commercial spaces targets the entire urban population, often extending to surrounding cities [35].

(2)

Site Selection in Tier-2 Bazaar Business Districts For bazaars located in secondary business districts, a “regional- level” scale is recommended as the primary positioning, supplemented by a “city-level” orientation. Retail and entertainment formats occupy considerable space and primarily attract residents from the surrounding area. The diversity of integrated formats is generally more limited compared to city-level bazaars. The cultural-commercial services are consequently oriented toward serving residents of a specific urban region.

(3)

Site Selection in Tier-3 or Township Bazaar Business Districts Bazaars in tertiary business districts or town ship commercial areas should adopt a positioning that is primarily “regional-level” with a secondary “community-level” focus. They mainly serve residents from a specific urban region and surrounding communities. Food, beverage, and retail formats dominate the tenancy and attract patronage from local residents [36,37,38]. These bazaars incorporate a wider variety of formats than purely community-level ones, while their cultural-commercial services are tailored principally to the adjacent communities.

(4)

Site Selection Outside Formal Bazaar Business Districts For locations situated outside established business districts—characterized by a weaker commercial atmosphere and considerable distance from core commercial areas—a “community-level” scale is advised. These bazaars primarily serve immediate neighborhood residents and focus on essential commercial activities, offering a limited mix of formats or community services. The cultural-commercial function, consequently, is oriented almost exclusively toward the local community [39,40].

To more convincingly substantiate the proposed link between cultural vitality and commercial performance, this study complemented quantitative analysis with field investigations of 15 bazaar cases. These investigations included merchant interviews, behavioral observations of patrons, and assessments of spatial function usage, serving to validate the congruence between the business district tiers identified by the quantitative model and the actual cultural-commercial functional match. For instance, in Tier-2 business districts, we documented the actual composition of retail and entertainment formats and gathered merchant insights regarding customer demographics and culturally oriented services. This mixed-methods approach—combining quantitative identification with qualitative validation—not only strengthens the explanatory power of the hierarchical coupling mechanism but also provides contextualized, real-world validity for the model outputs.

3. Results

3.1. Research on the Structural Characteristics of Commercial District Business Formats

Analysis of bazaar site selection in Xinjiang revealed that accurately classifying business districts—particularly into “city-level,” “regional-level,” and “community-level” categories—becomes challenging when they are numerous and unevenly distributed. To address this, Ürümqi was selected as a case study to delineate the spatial extent of urban bazaar business districts.

A total of 83,127 commercial POI (Points of Interest) data points [31,41,42] were collected from seven districts and counties of Ürümqi in March 2022 using Amap. These data were then processed through projection transformation in ArcGIS software and categorized into five primary types, as summarized in

Table 3. Statistics on Five Categories of POIs in Ürümqi City.

POI Category	Subclass	Number of POIs	Ratio
culinary delights	tea lounge, cake and dessert shop, coffee, international cuisine, snacks and fast food, Chinese cuisine	20,550	24.72%
shopping and consumption	department stores, convenience stores, supermarkets, shopping malls, pet shops, electronics and appliances, home furnishings and building materials, duty-free shops, commercial streets, daily necessities, stationery and sports goods, markets, fitness and sports	35,407	42.59%
financial institutions	ATM, insurance, investment and wealth management, banking	1882	2.26%
hotel accommodations	budget chain hotels, inns, hostels, three-star, four-star, five-star hotels	1890	2.27%
lifestyle services	KTV, lottery sales, telecom retail stores, cinemas, retirement resorts, public utilities, bars, beauty salons and hairdressers, farm stay tourism, card and board game rooms, automotive services, photo printing, internet cafes, logistics, laundry, bath and massage, information services, pharmaceutical sales, post offices, amusement parks, agencies, other	23,397	28.15%
Total		83,126	100.00%

Source: Author’s own illustration. Note: Classification criteria for culinary sub-categories: (a) ‘Chinese Cuisine’: establishments primarily offering full, sit-down meals based on traditional Chinese culinary traditions; (b) ‘Snacks and Fast Food’: establishments specializing in quick-service, portable food items and light meals; (c) ‘Tea Lounge’: establishments whose primary business is the sale of non-alcoholic beverages (including milk tea) for immediate consumption.

.

In the study of commercial POI screening and spatial distribution in Ürümqi, the selected POI data were first geographically projected along with the city’s administrative boundaries. As shown in Figure 2, these commercial POIs exhibit varying degrees of spatial agglomeration, characterized by a multi-nodal distribution pattern.

To further quantify this clustering, a nearest neighbor index analysis was conducted on all POI data using ArcGIS, with results illustrated in Figure 3. The analysis yielded a nearest neighbor ratio of 0.103, which is below the threshold value of 1. Additionally, the Z-score of −494.599 is well below the critical value of −2.58, and the p-value approximates 0. Together, these results confirm a statistically significant clustered pattern of commercial POIs in Ürümqi, rejecting the null hypothesis of random distribution.

Research on Structural Characteristics of Commercial Space Formats Across Different Types

To examine the internal composition of the five commercial categories, detailed statistics were compiled for their subcategories, as organized in

Table A1. Subcategory Commercial Facility POI Statistics.

Business Major	Commercial Subcategory	Quantity	Ratio	Summary
Culinary Delights	Tea Lounge	165	0.80%	20,550
	Cake and Dessert Shop	1195	5.82%
	Coffee	142	0.69%
	Foreign Cuisine	214	1.04%
	Snacks and Fast Food	2613	12.72%
	Chinese Cuisine	16,221	78.93%
Financial Institutions	ATM	629	33.42%	1882
	Insurance	134	7.12%
	Investment And Financial Management	369	19.61%
	Bank	753	40.01%
Shopping And Consumption	Department Store	104	0.29%	35,407
	Convenience Store	5739	16.21%
	Supermarket	922	2.60%
	Shopping Center	48	0.14%
	Flowers, birds, fish, and insects	438	1.24%
	Home Appliances and Electronics	1518	4.29%
	Home Furnishings and Building Materials	6949	19.63%
	Duty-Free Shop	13	0.04%
	Commercial Street	71	0.20%
	Daily Necessities	13,682	38.64%
	Sports and stationery supplies	696	1.97%
	Market	4218	11.91%
	Exercise and Fitness	1009	2.85%
Hotel Accommodations	Budget hotel chain	81	4.29%	1890
	Hotel	1657	87.67%
	Youth Hostel	4	0.21%
	Three-star hotel	76	4.02%
	Four-star hotel	52	2.75%
	Five-star/Luxury hotel	20	1.06%
Lifestyle Services	KTV	58	0.25%	23,397
	Lottery Sales	543	2.32%
	Telecom Service Center	479	2.05%
	Cinema	49	0.21%
	Retirement Living	111	0.47%
	Public Utilities	42	0.18%
	Bar	159	0.68%
	Beauty Salon	3659	15.64%
	Farm Stay	699	2.99%
	Others	3793	16.21%
	Card and Board Game Room	173	0.74%
	Automotive-related	6935	29.64%
	Photo Printing	679	2.90%
	Internet Cafe	119	0.51%
	Logistics	986	4.21%
	Laundry	364	1.56%
	Bath and Massage	824	3.52%
	Information Consultation Center	18	0.08%
	Pharmaceutical Sales	2285	9.77%
	Post Office	175	0.75%
	Amusement Park	55	0.24%
	Agency	1192	5.09%

Note: The star-rating classification for hotels in this study was directly sourced from the commercial POI attributes obtained via the Amap (AutoNavi) API. This classification reflects the hotel’s self-reported or platform-verified market positioning at the time of data crawling (March 2022), and is based on the Chinese national standard (GB/T 14308-2010) [43] for tourist hotel star ratings. It is important to note that this commercially sourced data may, at times, differ from the official, static list maintained by tourism authorities, as the latter can have a different update cycle and certification process.

. Pie charts were further generated to visualize the proportional distribution of subcategories within each commercial type. This graphical representation allows for clear identification of the dominant subcategories in each facility type.

Specifically, snack bars and Chinese restaurants emerged as the dominant subcategories within the food and beverage sector, reflecting both diverse consumer demands and local culinary characteristics. In the lodging category, hotels predominated due to their widespread distribution and broad consumer appeal.

When projected into the WGS1984 UTM Zone 44N coordinate system, all five commercial POI categories exhibited significant spatial clustering and a pronounced north–south disparity, as illustrated in Figure 4.

3.2. Identification and Demarcation of Bazaar Commercial Zone Boundaries

A kernel density analysis was conducted on commercial POIs in Ürümqi using a cell size of 100 m × 100 m and search radii ranging from 200 m to 1200 m. The results indicate that larger radii produce smoother density surfaces but coarser spatial details, as illustrated in Figure 5.

Spatial cluster analysis was conducted using the multi-distance spatial cluster analysis tool Ripley’s K function. As shown in Figure 6 and

Table 4. Commercial Ripley’s K-Chart.

OBJECTID	ExpectedK	ObservedK	DiffK
1	200	3161.517189	2961.517189
2	250	3719.492922	3469.492922
3	300	4249.512465	3949.512465
4	350	4761.089132	4411.089132
5	400	5258.365377	4858.365377
6	450	5743.804796	5293.804796
7	500	6218.964006	5718.964006
8	550	6688.426251	6138.426251
9	600	7153.738643	6553.738643
10	650	7612.675103	6962.675103
11	700	8066.070798	7366.070798
12	750	8515.216424	7765.216424
13	800	8959.809914	8159.809914
14	850	9400.173351	8550.173351
15	900	9837.661634	8937.661634
16	950	10,272.16684	9322.166843
17	1000	10,702.86948	9702.869477
18	1050	11,128.38804	10,078.38804
19	1100	11,547.78695	10,447.78695
20	1150	11,963.7555	10,813.7555
21	1200	12,379.19353	11,179.19353
22	1250	12,793.24566	11,543.24566
23	1300	13,205.15475	11,905.15475
24	1350	13,615.9597	12,265.9597
25	1400	14,024.76547	12,624.76547

, across all bandwidths (200 m–1200 m), the observed K value for commercial POIs consistently exceeded the expected K value. This strongly indicates a statistically significant clustered spatial pattern throughout the examined scales. Furthermore, the increasing Diff K value with larger bandwidths confirms an intensification of the clustering phenomenon.

Detailed analysis reveals that the commercial distribution in Ürümqi exhibits a pronounced multi-nucleated cluster structure. While the degree of spatial clustering was observed to increase with larger bandwidths within the 200 m–1200 m range, a bandwidth of 300 m was selected as the standard for this study. This scale was chosen to align with the focus on central business district characteristics and to ensure the precision and granularity of the analysis.

This study adapted the standard deviation concept of the normal distribution to delineate urban commercial district boundaries. As illustrated in Figure 7, within a normal distribution, one, two, and three standard deviations from the mean encompass approximately 68.2%, 95.4%, and 99.6% of the data, respectively. It is noted that a mean minus standard deviation yielding a negative value was deemed statistically insignificant for the purpose of boundary demarcation.

The three standard deviation surfaces exhibit distinct spatial characteristics. The one-standard-deviation surface covers 1.33% of Ürümqi’s area, contains 94.83% of the commercial POIs, and has a density of 413.98 per km², which is 71.23 times the city’s average. The two-standard-deviation surface, covering a smaller area (0.99%), contains 90.79% of POIs at a higher density of 534.94 per km² (91.52 times the average). The three-standard-deviation surface covers only 0.78% of the area yet contains 86.24% of POIs at a significantly elevated density of 641.28 per km², equivalent to 110.34 times the average density. These results demonstrate the efficacy of the standard deviation method for delineating commercial districts. The one and two-standard-deviation surfaces, being more extensive and continuous with moderately high POI density, represent the broader central commercial boundary. In contrast, the three-standard-deviation surface, characterized by its exceptional POI density and coherent spatial pattern, was identified as the optimal boundary for Ürümqi’s core central business district (

Table 5. Statistical Summary of Standard Deviation Curves for Commercial POI Recognition in Ürümqi.

Region	Area (km2)	POI Quantity	POI Density	Area Ratio	Ratio of Quantities
Three-Standard-Deviation Area	111.78	71,682	641.2775	0.78%	86.24%
Area of two standard deviations	141.86	75,461	531.9399	0.99%	90.79%
One standard deviation area	190.41	78,825	413.9751	1.33%	94.83%
Urumqi City	14,301.26	83,120	5.812075	100.00%	100.00%

Source: Author’s own illustration.

and Figure 8).

Analysis based on the three-standard-deviation surface successfully identified 172 commercial clusters within Ürümqi (Figure 9). These clusters are predominantly concentrated in the Saybag, Tianshan, Shuimogou, Xinshi, and Midong districts, demonstrating a clear pattern of high density. A limited number of clusters are sporadically distributed in other areas, while no commercial clusters were identified in Dabancheng District or Ürümqi County.

Analysis of the spatial distribution reveals well-developed commercial clusters in the old urban core, exhibiting a distinctive “large agglomeration with multiple small clusters” pattern. This configuration features primary agglomerations concentrated in the central city district, complemented by numerous smaller clusters dispersed throughout peripheral areas. These elements interact dynamically to form the cohesive spatial structure of Ürümqi’s commercial network.

Each of the 172 commercial districts was evaluated and rated across three primary dimensions: commercial scale, commercial quality, and external factors. Commercial scale was assessed based on the district’s land area, commercial quality was measured by the number of large shopping centers and the diversity of commercial types, and external factors included road network density and public transportation stop density. These metrics were validated through field investigations, and the results are presented in

Table A2. Statistical Table of Key Indicators for Ürümqi Commercial Districts.

Code Name	Number of Bus Stops	Large Shopping Mall	Number of Commercial Outlets	Types	Commercial Area (km2)	Road Network Density
1	0	0	13	3	0.01	10,002.81
2	0	0	49	3	0.1	7030.219
3	0	0	21	3	0.05	9378.248
4	0	0	31	3	0.07	6076.22
5	0	0	23	4	0.06	0
6	0	0	35	3	0.1	4004.827
7	1	0	52	3	0.14	6699.731
8	3	0	153	3	0.32	4033.144
9	0	0	98	1	0.22	6635.766
10	0	0	2	2	0.01	20,062.06
11	0	0	15	4	0.02	24,141.9
12	0	0	39	5	0.09	4435.902
13	3	1	197	5	0.46	7992.997
14	7	0	294	3	0.56	6167.645
15	0	0	20	3	0.04	5013.364
16	0	0	27	2	0.05	3945.49
17	0	0	12	3	0.03	5922.301
18	1	0	77	4	0.19	2639.236
19	2	0	94	3	0.24	11,464.2
20	0	0	7	2	0.01	6687.315
21	0	0	27	3	0.06	6646.724
22	0	0	19	3	0.03	12,497.29
23	3	0	62	4	0.14	7006.885
24	0	0	80	3	0.17	9633.737
25	2	0	87	3	0.19	8598.28
26	5	0	220	3	0.52	0
27	0	0	3	1	0.01	213,176
28	48	7	3710	5	6.88	6615.145
29	0	0	13	4	0.02	5013.934
30	3	0	83	4	0.19	7837.607
31	0	1	68	3	0.13	4142.54
32	0	0	9	3	0.02	10,010.01
33	0	0	19	3	0.04	3508.338
34	0	0	55	3	0.14	12,075.84
35	0	0	120	4	0.29	5129.035
36	1	0	37	3	0.08	11,284.58
37	0	0	89	3	0.21	7629.972
38	5	0	159	4	0.35	7215.884
39	13	2	128	5	0.32	11,753.5
40	2	1	131	3	0.25	811.6033
41	10	1	353	4	0.43	7807.151
42	0	0	36	3	0.08	3236.85
43	2	0	585	4	0.7	3702.952
44	28	2	4830	5	4.64	5186.993
45	28	1	935	4	1.57	6030.729
46	0	0	5	2	0.01	0
47	7	0	65	4	0.15	16,188.12
48	12	0	27	3	0.06	12,754.45
49	0	0	162	3	0.33	3540.737
50	2	0	54	3	0.13	4093.555
51	0	0	203	4	0.32	6157.512
52	0	0	46	3	0.11	2160.019
53	4	0	212	4	0.47	8984.155
54	3	0	221	4	0.51	10,659.65
55	2	0	57	3	0.16	2255.367
56	0	0	55	4	0.15	12,245.76
57	2	0	21	2	0.04	5133.852
58	9	0	283	4	0.62	9192.963
59	0	0	102	5	0.23	6133.657
60	2	2	124	3	0.33	9263.566
61	0	0	16	2	0.02	6338.113
62	16	1	281	5	0.51	10,736.45
63	5	0	115	4	0.33	9076.223
64	1	0	91	5	0.22	13,617.99
65	1	0	32	3	0.09	6205.643
66	2	0	17	3	0.03	3958.834
67	14	0	1066	5	1.85	6195.358
68	0	0	8	2	0.02	8744.065
69	0	0	124	3	0.23	9996.211
70	1	0	42	3	0.09	3978.691
71	3	0	54	4	0.13	4344.255
72	3	0	166	4	0.39	7930.483
73	0	0	33	2	0.07	0
74	11	0	100	5	0.25	8629.358
75	8	0	189	4	0.34	8941.877
76	3	0	281	4	0.44	9925.58
77	0	0	61	3	0.14	2508.618
78	9	0	271	4	0.46	10,916.79
79	0	0	42	3	0.08	7786.537
80	0	0	36	3	0.09	6682.585
81	1	0	25	2	0.05	4003.685
82	51	1	665	5	1.3	11,722.59
83	1	0	37	3	0.08	8209.109
84	6	0	261	5	0.39	6363.445
85	0	0	10	3	0.01	1452.754
86	0	0	21	3	0.03	15,798.75
87	0	0	12	2	0.02	95.50135
88	4	1	459	3	0.61	6803.498
89	1	0	28	3	0.06	9621.84
90	0	0	30	2	0.07	2868.577
91	0	0	380	5	0.73	6273.33
92	9	0	226	5	0.45	14,732.17
93	4	2	263	5	0.5	6835.767
94	5	0	73	3	0.16	6870.481
95	2	1	67	3	0.13	17,117.09
96	10	0	203	4	0.5	12,225.49
97	4	0	55	3	0.11	12,135.89
98	10	0	182	4	0.38	8007.826
99	0	0	27	3	0.06	24,103.98
100	0	0	34	3	0.09	6157.684
101	4	0	43	4	0.09	7852.053
102	5	0	30	3	0.07	28,036.64
103	15	3	1309	5	0.93	11,849.29
104	1	1	119	4	0.25	9484.119
105	0	0	28	4	0.06	8352.634
106	0	0	10	3	0.01	10,179.83
107	1	0	54	3	0.11	8241.276
108	0	0	40	3	0.09	3931.09
109	14	0	460	5	0.87	8873.35
110	2	0	15	3	0.05	15,276.08
111	8	0	61	4	0.16	6181.829
112	0	1	104	4	0.18	4319.237
113	0	0	37	3	0.08	17,205.65
114	0	0	262	2	0.46	12,405.82
115	47	2	898	5	1.12	14,724.56
116	1	0	31	4	0.07	13,300.46
117	2	0	37	4	0.07	7798.386
118	2	0	142	3	0.24	4299.45
119	10	0	39	5	0.1	26,894.5
120	0	0	90	3	0.23	11,519.68
121	4	0	11	3	0.03	11,412.98
122	6	0	71	3	0.16	19,281
123	4	0	111	4	0.27	7186.834
124	0	0	14	3	0.02	5136.521
125	3	0	36	4	0.07	5728.426
126	2	0	259	3	0.42	6579.779
127	4	1	77	3	0.18	12,317.91
128	0	0	21	3	0.04	11,141.93
129	0	0	22	3	0.04	8673.574
130	2	0	20	3	0.04	15,732.21
131	52	0	645	5	1.22	13,965.93
132	2	0	28	3	0.07	3893.127
133	0	0	24	4	0.04	12,660.85
134	4	0	161	5	0.31	2841.795
135	2	0	46	4	0.11	17,265.15
136	4	0	58	3	0.14	6463.699
137	3	0	65	4	0.13	13,569.26
138	1	0	46	3	0.09	0
139	2	0	94	3	0.18	2840.986
140	0	0	24	4	0.04	7546.298
141	0	0	56	3	0.13	10,628.52
142	8	0	457	5	1.08	5795.175
143	32	1	722	5	1.46	10,514.52
144	3	0	50	3	0.11	10,056.56
145	0	0	15	1	0.02	10,000.31
146	2127	97	41,732	5	60.25	11,645.4
147	4	0	59	3	0.12	19,658.42
148	25	6	334	5	0.71	9734.6
149	0	0	137	3	0.21	2302.422
150	17	0	194	5	0.38	14,705.68
151	2	0	57	4	0.14	14,108.75
152	1	0	118	3	0.2	6130.129
153	5	0	11	3	0.02	10,245.57
154	5	0	23	3	0.03	21,905.78
155	0	0	20	3	0.03	7506.487
156	0	0	16	3	0.02	9982.14
157	12	0	94	4	0.24	6368.833
158	2	0	61	3	0.15	4632.193
159	3	0	31	3	0.07	12,818.5
160	0	0	21	4	0.06	3080.419
161	0	0	14	3	0.02	5905.266
162	0	0	12	3	0.03	7348.327
163	0	0	12	1	0.03	0
164	0	0	47	4	0.13	2047.319
165	7	0	166	4	0.4	7539.799
166	0	0	93	5	0.18	12,324.54
167	10	0	86	4	0.21	10,205.93
168	0	0	52	4	0.1	6177.964
169	1	0	65	4	0.16	18,651.49
170	1	1	163	5	0.28	6577.65
171	4	0	107	5	0.26	8877.142
172	8	1	178	5	0.35	5250.789

Source: Author’s own illustration.

.

This study employed a systematic evaluation framework to assess commercial districts based on three dimensions: scale, quality, and external factors. Utilizing the Analytic Hierarchy Process (AHP), we integrated input from a panel of 20 domain experts to determine the relative weighting of each evaluation factor. The resulting weights and comprehensive scores are presented in

Table 6. The proportion of different factor weights and the specific results.

Major Factors	Level 1 Weight	Subcategory Factors	Secondary Weight
Commercial scale	59.538	Commercial district land area	0.837
		Number of commercial outlets	0.173
Commercial-grade quality	12.827	Number of large shopping malls	0.862
		Business Categories	0.148
External business factors	27.635	Road network density	0.516
		Bus frequency	0.494

Source: Author’s own illustration.

. These findings provide a robust empirical foundation and theoretical basis for understanding and optimizing the hierarchical classification of commercial districts.

Given the multitude of indicators and their significant variations in magnitude, a normalization procedure was applied to ensure comparability and precision in the quantitative analysis across commercial districts.

The normalized indicator values were then aggregated using weighted summation to compute a composite score for each commercial district. These composite scores subsequently served as the basis for hierarchical classification.

The seven indicators corresponding to the 172 identified commercial districts were incorporated into the aforementioned model to calculate their respective composite indicator values. These values were then normalized and classified into hierarchical clusters based on their magnitude. The normalization process involved multiplying each indicator by its corresponding primary and secondary weights, followed by summation. Detailed results are provided in

Table A3. Urumqi Commercial District Rating Statistics Table.

Code	NORMALIZED AREA	Bus Stop Density	Number of Large Shopping Malls	Commercial Outlets	Types	Road Network Density	Total Score	Total Score × 1000
1	0.0000	0.0000	0.0000	0.0003	0.5000	0.0469	0.0214	21.3910
2	0.0015	0.0000	0.0000	0.0011	0.5000	0.0330	0.0203	20.2513
3	0.0007	0.0000	0.0000	0.0005	0.5000	0.0440	0.0213	21.3256
4	0.0010	0.0000	0.0000	0.0007	0.5000	0.0285	0.0193	19.3308
5	0.0008	0.0000	0.0000	0.0005	0.7500	0.0000	0.0226	22.5877
6	0.0015	0.0000	0.0000	0.0008	0.5000	0.0188	0.0182	18.2171
7	0.0022	0.0005	0.0000	0.0012	0.5000	0.0314	0.0204	20.4319
8	0.0051	0.0014	0.0000	0.0036	0.5000	0.0189	0.0205	20.5178
9	0.0035	0.0000	0.0000	0.0023	0.0000	0.0311	0.0063	6.3427
10	0.0000	0.0000	0.0000	0.0000	0.2500	0.0941	0.0206	20.6393
11	0.0002	0.0000	0.0000	0.0003	0.7500	0.1132	0.0382	38.2012
12	0.0013	0.0000	0.0000	0.0009	1.0000	0.0208	0.0332	33.1808
13	0.0075	0.0014	0.0103	0.0047	1.0000	0.0375	0.0401	40.1470
14	0.0091	0.0033	0.0000	0.0070	0.5000	0.0289	0.0245	24.4944
15	0.0005	0.0000	0.0000	0.0004	0.5000	0.0235	0.0184	18.3553
16	0.0007	0.0000	0.0000	0.0006	0.2500	0.0185	0.0104	10.3728
17	0.0003	0.0000	0.0000	0.0002	0.5000	0.0278	0.0189	18.8548
18	0.0030	0.0005	0.0000	0.0018	0.7500	0.0124	0.0256	25.5936
19	0.0038	0.0009	0.0000	0.0022	0.5000	0.0538	0.0246	24.5677
20	0.0000	0.0000	0.0000	0.0001	0.2500	0.0314	0.0118	11.8089
21	0.0008	0.0000	0.0000	0.0006	0.5000	0.0312	0.0196	19.6162
22	0.0003	0.0000	0.0000	0.0004	0.5000	0.0586	0.0232	23.2188
23	0.0022	0.0014	0.0000	0.0014	0.7500	0.0329	0.0282	28.1620
24	0.0027	0.0000	0.0000	0.0019	0.5000	0.0452	0.0226	22.6220
25	0.0030	0.0009	0.0000	0.0020	0.5000	0.0403	0.0222	22.2458
26	0.0085	0.0024	0.0000	0.0052	0.5000	0.0000	0.0198	19.7818
27	0.0000	0.0000	0.0000	0.0000	0.0000	1.0000	0.1409	140.9409
28	0.1140	0.0226	0.0722	0.0889	1.0000	0.0310	0.1093	109.3166
29	0.0002	0.0000	0.0000	0.0003	0.7500	0.0235	0.0256	25.5502
30	0.0030	0.0014	0.0000	0.0019	0.7500	0.0368	0.0292	29.1723
31	0.0020	0.0000	0.0103	0.0016	0.5000	0.0194	0.0196	19.6393
32	0.0002	0.0000	0.0000	0.0002	0.5000	0.0470	0.0215	21.4680
33	0.0005	0.0000	0.0000	0.0004	0.5000	0.0165	0.0174	17.3579
34	0.0022	0.0000	0.0000	0.0013	0.5000	0.0566	0.0239	23.9298
35	0.0046	0.0000	0.0000	0.0028	0.7500	0.0241	0.0281	28.1007
36	0.0012	0.0005	0.0000	0.0008	0.5000	0.0529	0.0229	22.9345
37	0.0033	0.0000	0.0000	0.0021	0.5000	0.0358	0.0216	21.6472
38	0.0056	0.0024	0.0000	0.0038	0.7500	0.0338	0.0304	30.3855
39	0.0051	0.0061	0.0206	0.0030	1.0000	0.0551	0.0430	42.9590
40	0.0040	0.0009	0.0103	0.0031	0.5000	0.0038	0.0187	18.7016
41	0.0070	0.0047	0.0103	0.0084	0.7500	0.0366	0.0332	33.2265
42	0.0012	0.0000	0.0000	0.0008	0.5000	0.0152	0.0175	17.5477
43	0.0115	0.0009	0.0000	0.0140	0.7500	0.0174	0.0318	31.7764
44	0.0769	0.0132	0.0206	0.1157	1.0000	0.0243	0.0864	86.4153
45	0.0259	0.0132	0.0103	0.0224	0.7500	0.0283	0.0440	43.9614
46	0.0000	0.0000	0.0000	0.0001	0.2500	0.0000	0.0074	7.3828
47	0.0023	0.0033	0.0000	0.0015	0.7500	0.0759	0.0346	34.5760
48	0.0008	0.0056	0.0000	0.0006	0.5000	0.0598	0.0244	24.4182
49	0.0053	0.0000	0.0000	0.0038	0.5000	0.0166	0.0201	20.1051
50	0.0020	0.0009	0.0000	0.0012	0.5000	0.0192	0.0187	18.6953
51	0.0051	0.0000	0.0000	0.0048	0.7500	0.0289	0.0292	29.2281
52	0.0017	0.0000	0.0000	0.0011	0.5000	0.0101	0.0171	17.1062
53	0.0076	0.0019	0.0000	0.0050	0.7500	0.0421	0.0326	32.6038
54	0.0083	0.0014	0.0000	0.0052	0.7500	0.0500	0.0340	33.9979
55	0.0025	0.0009	0.0000	0.0013	0.5000	0.0106	0.0177	17.7334
56	0.0023	0.0000	0.0000	0.0013	0.7500	0.0574	0.0315	31.4997
57	0.0005	0.0009	0.0000	0.0005	0.2500	0.0241	0.0112	11.1892
58	0.0101	0.0042	0.0000	0.0067	0.7500	0.0431	0.0345	34.4629
59	0.0037	0.0000	0.0000	0.0024	1.0000	0.0288	0.0356	35.6046
60	0.0053	0.0009	0.0206	0.0029	0.5000	0.0435	0.0259	25.9338
61	0.0002	0.0000	0.0000	0.0003	0.2500	0.0297	0.0117	11.6819
62	0.0083	0.0075	0.0103	0.0067	1.0000	0.0504	0.0434	43.4023
63	0.0053	0.0024	0.0000	0.0027	0.7500	0.0426	0.0313	31.3446
64	0.0035	0.0005	0.0000	0.0021	1.0000	0.0639	0.0405	40.5077
65	0.0013	0.0005	0.0000	0.0007	0.5000	0.0291	0.0196	19.6465
66	0.0003	0.0009	0.0000	0.0004	0.5000	0.0186	0.0177	17.6962
67	0.0305	0.0066	0.0000	0.0255	1.0000	0.0291	0.0522	52.1641
68	0.0002	0.0000	0.0000	0.0001	0.2500	0.0410	0.0133	13.2531
69	0.0037	0.0000	0.0000	0.0029	0.5000	0.0469	0.0235	23.4605
70	0.0013	0.0005	0.0000	0.0010	0.5000	0.0187	0.0182	18.1985
71	0.0020	0.0014	0.0000	0.0012	0.7500	0.0204	0.0263	26.3002
72	0.0063	0.0014	0.0000	0.0039	0.7500	0.0372	0.0311	31.0757
73	0.0010	0.0000	0.0000	0.0007	0.2500	0.0000	0.0079	7.9429
74	0.0040	0.0052	0.0000	0.0023	1.0000	0.0405	0.0381	38.1141
75	0.0055	0.0038	0.0000	0.0045	0.7500	0.0419	0.0317	31.7083
76	0.0071	0.0014	0.0000	0.0067	0.7500	0.0466	0.0331	33.0838
77	0.0022	0.0000	0.0000	0.0014	0.5000	0.0118	0.0176	17.6191
78	0.0075	0.0042	0.0000	0.0064	0.7500	0.0512	0.0343	34.2610
79	0.0012	0.0000	0.0000	0.0010	0.5000	0.0365	0.0206	20.5703
80	0.0013	0.0000	0.0000	0.0008	0.5000	0.0313	0.0199	19.9079
81	0.0007	0.0005	0.0000	0.0006	0.2500	0.0188	0.0105	10.4701
82	0.0214	0.0240	0.0103	0.0159	1.0000	0.0550	0.0537	53.6945
83	0.0012	0.0005	0.0000	0.0008	0.5000	0.0385	0.0209	20.9012
84	0.0063	0.0028	0.0000	0.0062	1.0000	0.0299	0.0378	37.8366
85	0.0000	0.0000	0.0000	0.0002	0.5000	0.0068	0.0157	15.7309
86	0.0003	0.0000	0.0000	0.0005	0.5000	0.0741	0.0254	25.4063
87	0.0002	0.0000	0.0000	0.0002	0.2500	0.0004	0.0075	7.5450
88	0.0100	0.0019	0.0103	0.0110	0.5000	0.0319	0.0265	26.5391
89	0.0008	0.0005	0.0000	0.0006	0.5000	0.0451	0.0216	21.6493
90	0.0010	0.0000	0.0000	0.0007	0.2500	0.0135	0.0098	9.8322
91	0.0120	0.0000	0.0000	0.0091	1.0000	0.0294	0.0405	40.4728
92	0.0073	0.0042	0.0000	0.0054	1.0000	0.0691	0.0440	43.9678
93	0.0081	0.0019	0.0206	0.0063	1.0000	0.0321	0.0409	40.9388
94	0.0025	0.0024	0.0000	0.0017	0.5000	0.0322	0.0210	21.0144
95	0.0020	0.0009	0.0103	0.0016	0.5000	0.0803	0.0283	28.3422
96	0.0081	0.0047	0.0000	0.0048	0.7500	0.0573	0.0354	35.3531
97	0.0017	0.0019	0.0000	0.0013	0.5000	0.0569	0.0240	23.9781
98	0.0061	0.0047	0.0000	0.0043	0.7500	0.0376	0.0315	31.5293
99	0.0008	0.0000	0.0000	0.0006	0.5000	0.1131	0.0312	31.1579
100	0.0013	0.0000	0.0000	0.0008	0.5000	0.0289	0.0196	19.5560
101	0.0013	0.0019	0.0000	0.0010	0.7500	0.0368	0.0283	28.3282
102	0.0010	0.0024	0.0000	0.0007	0.5000	0.1315	0.0342	34.1655
103	0.0153	0.0071	0.0309	0.0313	1.0000	0.0556	0.0520	52.0232
104	0.0040	0.0005	0.0103	0.0028	0.7500	0.0445	0.0317	31.7181
105	0.0008	0.0000	0.0000	0.0006	0.7500	0.0392	0.0281	28.1220
106	0.0000	0.0000	0.0000	0.0002	0.5000	0.0478	0.0215	21.5007
107	0.0017	0.0005	0.0000	0.0012	0.5000	0.0387	0.0212	21.2098
108	0.0013	0.0000	0.0000	0.0009	0.5000	0.0184	0.0181	18.0985
109	0.0143	0.0066	0.0000	0.0110	1.0000	0.0416	0.0444	44.4256
110	0.0007	0.0009	0.0000	0.0003	0.5000	0.0717	0.0253	25.3376
111	0.0025	0.0038	0.0000	0.0014	0.7500	0.0290	0.0281	28.0965
112	0.0028	0.0000	0.0103	0.0024	0.7500	0.0203	0.0276	27.6291
113	0.0012	0.0000	0.0000	0.0008	0.5000	0.0807	0.0268	26.7855
114	0.0075	0.0000	0.0000	0.0062	0.2500	0.0582	0.0199	19.8996
115	0.0184	0.0221	0.0206	0.0215	1.0000	0.0691	0.0555	55.5181
116	0.0010	0.0005	0.0000	0.0007	0.7500	0.0624	0.0315	31.5462
117	0.0010	0.0009	0.0000	0.0008	0.7500	0.0366	0.0280	27.9868
118	0.0038	0.0009	0.0000	0.0034	0.5000	0.0202	0.0199	19.9472
119	0.0015	0.0047	0.0000	0.0009	1.0000	0.1262	0.0487	48.7477
120	0.0037	0.0000	0.0000	0.0021	0.5000	0.0540	0.0244	24.3853
121	0.0003	0.0019	0.0000	0.0002	0.5000	0.0535	0.0227	22.7371
122	0.0025	0.0028	0.0000	0.0017	0.5000	0.0904	0.0293	29.2783
123	0.0043	0.0019	0.0000	0.0026	0.7500	0.0337	0.0295	29.5299
124	0.0002	0.0000	0.0000	0.0003	0.5000	0.0241	0.0183	18.2581
125	0.0010	0.0014	0.0000	0.0008	0.7500	0.0269	0.0267	26.6795
126	0.0068	0.0009	0.0000	0.0062	0.5000	0.0309	0.0232	23.2152
127	0.0028	0.0019	0.0103	0.0018	0.5000	0.0578	0.0257	25.7310
128	0.0005	0.0000	0.0000	0.0005	0.5000	0.0523	0.0224	22.4096
129	0.0005	0.0000	0.0000	0.0005	0.5000	0.0407	0.0208	20.7801
130	0.0005	0.0009	0.0000	0.0004	0.5000	0.0738	0.0256	25.5693
131	0.0201	0.0244	0.0000	0.0154	1.0000	0.0655	0.0535	53.5315
132	0.0010	0.0009	0.0000	0.0006	0.5000	0.0183	0.0180	18.0075
133	0.0005	0.0000	0.0000	0.0005	0.7500	0.0594	0.0308	30.7966
134	0.0050	0.0019	0.0000	0.0038	1.0000	0.0133	0.0345	34.4822
135	0.0017	0.0009	0.0000	0.0011	0.7500	0.0810	0.0346	34.5956
136	0.0022	0.0019	0.0000	0.0013	0.5000	0.0303	0.0205	20.4813
137	0.0020	0.0014	0.0000	0.0015	0.7500	0.0637	0.0324	32.4259
138	0.0013	0.0005	0.0000	0.0011	0.5000	0.0000	0.0156	15.5777
139	0.0028	0.0009	0.0000	0.0022	0.5000	0.0133	0.0184	18.3744
140	0.0005	0.0000	0.0000	0.0005	0.7500	0.0354	0.0274	27.4152
141	0.0020	0.0000	0.0000	0.0013	0.5000	0.0499	0.0229	22.8933
142	0.0178	0.0038	0.0000	0.0109	1.0000	0.0272	0.0437	43.7239
143	0.0241	0.0150	0.0103	0.0173	1.0000	0.0493	0.0531	53.1369
144	0.0017	0.0014	0.0000	0.0012	0.5000	0.0472	0.0225	22.5276
145	0.0002	0.0000	0.0000	0.0003	0.0000	0.0469	0.0067	6.7251
146	1.0000	1.0000	1.0000	1.0000	1.0000	0.0546	0.8655	865.4780
147	0.0018	0.0019	0.0000	0.0014	0.5000	0.0922	0.0290	29.0432
148	0.0116	0.0118	0.0619	0.0080	1.0000	0.0457	0.0501	50.1071
149	0.0033	0.0000	0.0000	0.0032	0.5000	0.0108	0.0182	18.2414
150	0.0061	0.0080	0.0000	0.0046	1.0000	0.0690	0.0438	43.8077
151	0.0022	0.0009	0.0000	0.0013	0.7500	0.0662	0.0328	32.7815
152	0.0032	0.0005	0.0000	0.0028	0.5000	0.0288	0.0207	20.7075
153	0.0002	0.0024	0.0000	0.0002	0.5000	0.0481	0.0219	21.9470
154	0.0003	0.0024	0.0000	0.0005	0.5000	0.1028	0.0298	29.7671
155	0.0003	0.0000	0.0000	0.0004	0.5000	0.0352	0.0199	19.9216
156	0.0002	0.0000	0.0000	0.0003	0.5000	0.0468	0.0215	21.4666
157	0.0038	0.0056	0.0000	0.0022	0.7500	0.0299	0.0292	29.2111
158	0.0023	0.0009	0.0000	0.0014	0.5000	0.0217	0.0192	19.2325
159	0.0010	0.0014	0.0000	0.0007	0.5000	0.0601	0.0240	23.9794
160	0.0008	0.0000	0.0000	0.0005	0.7500	0.0145	0.0246	24.6194
161	0.0002	0.0000	0.0000	0.0003	0.5000	0.0277	0.0188	18.7664
162	0.0003	0.0000	0.0000	0.0002	0.5000	0.0345	0.0198	19.7976
163	0.0003	0.0000	0.0000	0.0002	0.0000	0.0000	0.0002	0.1883
164	0.0020	0.0000	0.0000	0.0011	0.7500	0.0096	0.0246	24.5737
165	0.0065	0.0033	0.0000	0.0039	0.7500	0.0354	0.0312	31.1541
166	0.0028	0.0000	0.0000	0.0022	1.0000	0.0578	0.0393	39.2656
167	0.0033	0.0047	0.0000	0.0020	0.7500	0.0479	0.0314	31.3551
168	0.0015	0.0000	0.0000	0.0012	0.7500	0.0290	0.0271	27.0706
169	0.0025	0.0005	0.0000	0.0015	0.7500	0.0875	0.0359	35.9047
170	0.0045	0.0005	0.0103	0.0039	1.0000	0.0309	0.0375	37.5249
171	0.0042	0.0019	0.0000	0.0025	1.0000	0.0416	0.0379	37.9312
172	0.0056	0.0038	0.0103	0.0042	1.0000	0.0246	0.0377	37.7039

Source: Author’s own illustration.

.

The results indicate that District 146 (Figure 10) exhibited notably higher scores, as represented by the blue areas, attributable to its central urban location. This finding also reflects the developmental disparity between urban and rural commercial areas. Consequently, a kernel density analysis was reapplied to District 146 using a 200 m search radius and a three-standard-deviation threshold to delineate its core commercial area more precisely.

Using District 146 as the base map—which contains 41,732 POIs and covers an area of 60.25 square kilometers—the kernel density analysis was performed, yielding a three-standard-deviation surface as shown in Figure 11. This resulted in a simplified delineation of the core commercial area, illustrated in Figure 12.

This methodology enabled the precise delineation of commercial district boundaries and their hierarchical classification in Ürümqi. The final demarcation results are presented in the figure below. By integrating the boundaries identified in both the initial and secondary delineation phases, we classified the commercial districts into three tiers using the natural breaks method: primary, secondary, and tertiary. These tiers represent the most commercially active areas in Ürümqi and provide critical reference data for subsequent analysis of bazaar commercial districts, as illustrated in Figure 13.

Building upon the identified commercial districts in Ürümqi, we delineated the spatial extent of bazaar commercial areas. According to data from the Ürümqi Statistical Yearbook (2019), districts with relatively high ethnic minority populations include Dabancheng, Ürümqi County, and Tianshan, while other districts all contain less than 30% ethnic minority residents (

Table 7. Ethnic Composition by District in Ürümqi.

Various Areas of Ürümqi	Han Chinese	Ethnic Minorities	Total Number of People	Percentage of Ethnic Minorities (%)
Tianshan District	274,064	186,430	460,494	40.48
Shaibak District	358,250	108,859	467,109	23.30
High-Tech Zone	412,617	92,696	505,313	18.34
Shuimogou District	180,740	51,804	232,544	22.28
Tou Tunhe District	147,477	48,700	196,177	24.82
Dabancheng District	10,515	22,003	32,518	67.66
Midong District	193,887	81,753	275,640	29.66
Urumqi County	18,206	34,557	52,763	65.49

Source: Author’s own illustration.

). As established previously, conventional commercial districts are predominantly concentrated in Saybag, Tianshan, Shuimogou, Xinshi, and Midong districts. Although Dabancheng and Ürümqi County exhibit higher ethnic minority proportions, they contain no significant commercial districts and were therefore excluded from consideration. Consequently, Tianshan District was identified as the primary area for Xinjiang bazaar commercial activity in Ürümqi, as illustrated in Figure 14.

4. Discussion

The proposed relationship between location selection criteria and hierarchical classification of bazaar commercial districts was applied to the delineation of Ürümqi’s bazaar districts. This framework was used to evaluate the positioning of 15 existing Xinjiang bazaars in Ürümqi. The analysis revealed 10 cases demonstrating strong concordance with the proposed model (exhibiting excellent operational status in 1 case and good status in 9 cases), while 5 cases showed discordance (showing fair operational status in 1 case and poor status in 4 cases), as detailed in

Table A4. Statistical Table on the Relationship Between Site Selection and Commercial Districts for the Xinjiang Bazaar in Ürümqi.

Number	Names of Bazaar in Xinjiang	Number of Retail Units	Scale Level	Commercial District Tier	Is the Site Selection Relationship Appropriate	Types of Cultural Commercial Spaces	Actual Operational Status
1	Western Regions International Trade City	953	City-level	Level 3	No	Less	Generally
2	Border International Trade City	830	City-level	Level 3	No	Less	Poor
3	Erdaoqiao Grand Bazaar	282	Regional level	Level 1	Yes	More	Better
4	Celestial Realm City	550	City-level	Level 1	Yes	More	Better
5	Xinjiang Min Street, Shanxi Lane	820	City-level	Level 3	No	More	v
6	International Bazaar	350	City-level	Level 1	Yes	More	Well
7	Yongfeng Town Market	150	Community-level	Non-commercial district	Yes	Less	Better
8	Shuimogou Village Reemployment Market	46	Community-level	Non-commercial district	Yes	None	Better
9	Urumqi County Harmony Shopping Center	50	Community-level	Non-commercial district	Yes	None	Better
10	Urumqi County Snow Town Specialty Commercial Street	100	Regional level	Non-commercial district	No	Less	Poor
11	Dabancheng Farmers’ Market	74	Community-level	Non-commercial district	Yes	None	Better
12	Hualong Meite Commercial Street	305	Community-level	Level 3	Yes	Less	Better
13	Hongshan Dried Fruit Market	420	Regional level	Level 2	No	Less	Poor
14	Shuixigou Permanent Shop Street	50	Community-level	Non-commercial district	Yes	None	Better
15	Reed Village Job Fair	57	Community-level	Non-commercial district	Yes	None	Better

Source: Author’s own illustration.

.

4.1. Analysis of Coupling Mechanisms Between Bazaar Location and Performance

The analysis of five non-conforming cases (

Table 8. Root Cause Analysis of Non-Conforming Bazaar Cases.

Business District Tier	Recommended Bazaar Scale and Positioning	Primary Service Orientation
Case Study	Primary Reason for Non-Conformance	Underlying Failure Factors
Xiyu International Trade City	Specialized Operational Model	Wholesale-centric business model negates need for high-density retail location Targets specific commercial clients rather than general public Strategic focus on logistics and parking over pedestrian accessibility
Bianjiang International Trade City	Physical and Competitive Obsolescence	Severe physical constraints (parking, space quality) Outcompeted by newer, superior facility (Xiyu International) Inability to adapt due to large size and tertiary zone location
Xinjiang Minjie Shanxi Alley	Strategic Overreach and Locational Disadvantage	Unfavorable location away from main thoroughfares with single street frontage Over-ambitious “city-level” positioning inconsistent with actual tertiary commercial zone location Over-investment in esthetics without developing core operational competencies Severe homogenization with other commercial offerings
Ürümqi County Ice-Snow Town	Scale-to-Demand Mismatch	Regional-scale development in area lacking fundamental urban commercial district Critical overdependence on seasonal skiing tourism Inadequate year-round operational viability

Source: Author’s own illustration.

) provides particularly valuable insights into the boundary conditions and complex interactions within these coupling mechanisms:

(1)

Xiyu International Trade City. Although classified as a “city-level” bazaar, it is situated within a tertiary-level commercial district—a locational mismatch. Despite this, it maintains relatively successful operations. Field investigation revealed its customer base consists primarily of domestic and international merchants, with wholesale being the predominant sales model. Consequently, its location strategy prioritizes access for major commercial clients rather than local residents, emphasizing transportation access and parking availability. This unique operational focus makes it a special case where conventional proximity to urban commercial centers becomes less relevant.

(2)

Bianjiang International Trade City. This bazaar thrived two decades ago but has experienced operational decline in recent years due to space constraints, inadequate parking, poor spatial quality, and competition from the newer, larger Xiyu International Trade City. Its location within a tertiary commercial zone, combined with its substantial size, presents significant challenges for functional adaptation or revitalization.

(3)

Ürümqi County Ice-Snow Town Commercial Street. As established previously, Ürümqi County lacks conventional urban commercial districts, making it suitable only for community-level bazaar development. This project’s regional-scale capacity significantly exceeds local demand patterns. Consequently, it experiences seasonal viability dependent solely on winter skiing tourism, with markedly poor performance during other seasons.

(4)

Hongshan Dried Fruit Market. Although situated within a secondary urban commercial district, this case was excluded from our defined bazaar commercial zones due to the low ethnic minority population (23.30%) in its vicinity, falling below the threshold for culturally significant commercial clustering. The failure of this market stems from a fundamental product-market mismatch. Its core operational model focused exclusively on selling Xinjiang specialty dried fruits and local products—goods that are widely available throughout Ürümqi and lack a unique value proposition to differentiate this market as a distinctive destination for tourists. Simultaneously, this product offering did not align with the daily consumption needs of the local residential community, creating a disconnect with both potential customer bases. Compounding this issue, the market’s operational format—resembling a conventional wholesale-retail outlet—failed to provide the vibrant, experiential atmosphere characteristic of successful bazaars, where cultural interaction and sensory engagement are key attractions. Consequently, the market experienced overall operational failure, with only peripheral street-front food and beverage establishments currently maintaining viable operations due to their alignment with neighborhood service demands.

(5)

Xinjiang Minjie Shanxi Alley. This bazaar experienced initial success but subsequently declined due to a combination of locational disadvantages and strategic missteps. Its primary constraint is an unfavorable geographical position: situated away from principal urban thoroughfares with limited street frontage, significantly reducing its visibility and pedestrian accessibility. Compounding this issue was a pronounced strategic overreach. The project was ambitiously positioned as a “city-level” attraction intended to draw domestic and international tourists, an aspiration that proved misaligned with its actual location outside Ürümqi’s primary bazaar commercial zone. Development resources were disproportionately allocated to architectural esthetics and interior finishes, rather than cultivating a unique operational core competency. This led to a severe lack of differentiation from other commercial offerings and an absence of competitive vitality. Consequently, the bazaar suffers from extremely low tenancy and footfall; currently, only a fraction of the ground-floor units remain operational, while the upper levels are largely vacant and closed to the public.

The spatial-commercial coupling manifests differently across cases. Xiyu International Trade City demonstrates that specialized business models (wholesale commerce) can decouple success from conventional location hierarchies, creating an exception to the general rule. Conversely, Xinjiang Minjie Shanxi Alley illustrates the severe consequences of strategic-locational decoupling, where ambitious “city-level” positioning fundamentally conflicts with its actual tertiary-zone location.

City-level bazaars should be conceived as culturally dedicated destinations, incorporating large public squares, iconic architectural elements, and mixed-use layouts that accommodate festivals, performances, and extended evening economies. Their governance should involve public–private partnerships to coordinate large-scale events and cross-regional marketing.

Regional-level bazaars perform better as community-integrated anchors, with a focus on pedestrian-oriented design, shaded walkways, and a balanced mix of retail, casual dining, and public seating. Management should emphasize local vendor curation and frequent cultural programming, such as weekend markets or craft workshops, to sustain repeat visitation.

Community-level bazaars ought to prioritize accessibility and essential service provision, often taking the form of compact, modular structures or open-air settings with flexible stalls. These spaces should support micro-vendors and everyday needs, governed through community cooperatives or municipal oversight to ensure affordability and local

The scale-context coupling emerges as crucial in cases like Ürümqi County Ice-Snow Town, where regional-scale development in an area lacking basic urban commercial infrastructure creates unsustainable operational models dependent solely on seasonal tourism.

The cultural-commercial coupling proves essential in ethnic marketplaces. Hongshan Dried Fruit Market’s failure demonstrates how low ethnic population density (23.30%), combined with product-market mismatch and lack of experiential retail atmosphere, disrupts the cultural sustainability necessary for bazaar success.

4.2. Design Strategies and Planning Implications

Based on these identified coupling mechanisms, we propose specific design and planning strategies tailored to different contexts:

For Locational-Strategic Alignment:

• Implement hierarchical zoning regulations that match bazaar scale and function with appropriate commercial district tiers.
• Develop transitional development guidelines for areas between different commercial tiers to prevent strategic overreach.
• Establish accessibility-based site selection criteria that prioritize pedestrian connectivity and public transportation access.

For Cultural-Commercial Integration:

• Create cultural vitality assessment tools incorporating ethnic population thresholds and intangible cultural heritage factors.
• Design experiential retail frameworks that emphasize cultural interaction, sensory engagement, and authentic atmosphere.
• Develop product-market matching mechanisms that align local specialties with both tourist and residential consumer needs.

For Scale-Context Optimization:

• Formulate demand-based scaling guidelines that prevent oversized development in commercially underdeveloped areas.
• Implement seasonal adaptability strategies for tourism-dependent locations to ensure year-round viability.
• Establish functional conversion protocols for oversized or poorly located bazaars to facilitate adaptive reuse.

4.3. Validation and Theoretical Contribution

To situate our findings within a broader international context, we draw instructive comparisons between Xinjiang bazaars and other traditional marketplace typologies globally. While Middle Eastern bazaars and North African souks similarly function as socio-cultural hubs, their spatial organization often reflects deeper historical layering and religious influences, with central mosques typically anchoring the commercial fabric. In contrast, Xinjiang bazaars—particularly those in Ürümqi—exhibit a more fluid integration into modern urban hierarchies, shaped by regional ethnic distributions and contemporary planning interventions. European covered markets, such as those in Barcelona or Budapest, share functional hybridity but often prioritize tourism and gourmet retail over everyday cultural practice. What distinguishes the Xinjiang bazaar is its explicit “hierarchical coupling mechanism”—a spatially embedded system that aligns commercial tier, cultural function, and service scope in a manner not systematically observed in other regional models. This framework offers a replicable approach for culturally grounded commercial regeneration in multi-ethnic cities worldwide, particularly where rapid urbanization threatens intangible cultural heritage.

The overall pattern of 10 conforming cases against 5 non-conforming ones provides strong validation for our location assessment methodology while revealing its limitations. The framework demonstrates particular strength in predicting success within conventional retail environments while identifying specific conditions—specialized business models, extreme scale-context mismatches, and cultural-commercial disconnects—that require additional considerations.

These findings contribute to sustainable urban planning theory by demonstrating how spatial strategies can mediate between cultural preservation and economic development. The identified coupling mechanisms offer a transferable framework for other regions facing similar challenges of traditional marketplace revitalization, while the design strategies provide actionable guidance for policymakers and urban planners seeking to balance commercial viability with cultural sustainability.

The exceptional case of Xiyu International Trade City further enriches our understanding by demonstrating that specialized commercial functions with unique customer bases can successfully operate outside conventional location hierarchies, suggesting avenues for future research on niche market adaptations within traditional commercial formats.

5. Conclusions

This study developed and empirically validated a spatial-analytical framework that integrates commercial hierarchy theory with principles of cultural sustainability to guide the site selection of bazaars in Xinjiang. Through multi-scale spatial analysis—including kernel density estimation and Ripley’s K-function analysis—of commercial agglomeration patterns in Ürümqi, we established a reproducible methodology for delineating and classifying bazaar commercial districts based on both economic vitality and socio-cultural criteria, particularly the distribution of ethnic minority populations. The proposed hierarchical model was tested against 15 real-world bazaar cases, with 10 demonstrating strong concordance between predicted and observed outcomes, confirming its practical applicability and predictive robustness.

Theoretical contributions of this work are threefold. First, it provides a theoretical extension of Central Place Theory by systematically incorporating cultural-demographic dimensions, thereby uncovering the socio-cultural mechanisms underpinning the spatial organization of commercial structures in multi-ethnic regions. Second, the introduction of the “hierarchical coupling mechanism” offers a conceptual framework that clarifies the intrinsic relationship and functional alignment between commercial spatial tiers and the socio-cultural roles of bazaars, enriching the theoretical discourse in sustainable commercial geography. Third, the research represents a methodological innovation through the integration of GIS-based spatial analysis, quantitative modeling, and cultural geography, establishing a rigorous and transferable paradigm for the empirical study of traditional marketplace spaces.

From an applied perspective, the hierarchical site selection model provides urban planners and policymakers with actionable guidance. By aligning bazaar scale, business mix, and cultural service functions with the hierarchical level of host commercial districts, the framework facilitates the synergistic development of cultural vitality and commercial resilience. These insights resonate strongly with the strategic objectives of Xinjiang’s “Cultural Enrichment” initiative, reinforcing the role of bazaars as multifunctional spaces for cultural transmission, social integration, and localized economic development.

These findings carry concrete implications for urban design and bazaar programming: at the city level, bazaars should incorporate symbolic architecture and multi-functional plazas capable of hosting large-scale cultural events; at the regional level, the emphasis shifts to creating pedestrian-friendly environments with a balanced retail-food mix that serves both local and visiting populations; while community-level bazaars benefit from modular, flexible layouts that adapt to daily needs and micro-enterprises. Each tier necessitates a distinct governance approach—from city-led cultural branding to community-based vendor management—ensuring that physical planning and operational models align with the specific cultural-commercial role each bazaar is expected to fulfill.

It is important to acknowledge the limitations related to the study’s geographical focus on Ürümqi and its reliance on cross-sectional data. Future research should aim to apply this framework to other socio-cultural contexts—such as Southern Xinjiang—to evaluate its generalizability and enable comparative analysis. The incorporation of longitudinal data would also allow for dynamic tracking of the co-evolution of commercial districts and bazaars, further deepening the understanding of their interactive mechanisms.