The Comparative Study of the Evolutionary Characteristics of Spatial Forms and Cultural Differences in the Russian-Japanese Railway Residential Architecture Heritage in Jilin Province

Abstract

Residential architectural heritage in Jilin Province, particularly along the Chinese Eastern and South Manchuria Railways, represents the most abundant and widespread form of heritage in the region. This study investigates the distinctive features of these heritage buildings, which are often overlooked due to their technological simplicity, ubiquity, and atypical nature. We analyzed 28 floor plans of residential buildings from the Russian and Japanese colonial periods using GIS technology, space syntax, and Z-score normalization. The findings reveal that (1) Russian elite housing layouts are complex and circular, while Japanese layouts consist of multiple branching combinations; Russian common housing typically has single-node, multi-branch layouts, whereas Japanese housing features multi-node, fewer-branch designs; (2) Russian houses prioritize warmth and privacy, with enclosed layouts that emphasize defensiveness, while Japanese houses focus on space efficiency, rational movement, and multifunctional design, with open layouts highlighting spatial integration and functional differentiation; (3) cultural differences are reflected in factors such as the historical rivalry between Russian and Japanese powers, natural environments, family structures, and standardized construction techniques. This study offers new insights into the cultural characteristics and historical context of Jilin Province’s railway residential heritage and supports its protection and future research.

1. Introduction

1.1. Research Background

Railway heritage has become a growing focus internationally. In 1899, the World Heritage Committee held its first World Railway Heritage Congress, promoting railway heritage nominations by signatory countries. The next year, ICOMOS conducted research that led to the inclusion of railway heritage from countries like India, Switzerland, and Iran on the World Heritage List [1]. These countries have since gained valuable conservation experience, offering important theoretical and practical references. In Jilin Province, railway heritage includes both the Chinese Eastern Railway and the South Manchuria Railway. From the late 19th to the early 20th century, Russian and Japanese imperial powers expanded their influence, with the railway becoming a key link between Northeast China and the outside world. After the Second Sino-Japanese War, many modern buildings in Northeast China were repurposed by government institutions as functional assets. These buildings were valued for their utility, with little attention to their architectural heritage. Unfortunately, preservation efforts have mainly focused on typical structures, neglecting non-typical heritage such as residential buildings.

1.2. Literature Review

1.2.1. Planning Models of Russian and Japanese Settlements

After the Russo-Japanese War in 1904, the competition between Russia and Japan over control of settlements and colonies intensified. Russian military stations were built hastily to transport supplies, often without careful planning [2,3,4]. Japan adopted and refined these Russian designs for its own settlements. In Jilin Province, the highest-ranking Russian station was the third-class station in Dehui. Russian planning models were characterized by: (1) a central station aligned orthogonally with the church; (2) grid-like, narrow street layouts; (3) unclear land-use zoning; (4) a loose arrangement of buildings. In contrast, Japan’s planning model featured: (1) a station at the center with railways as the axis; (2) a grid-like, square street network; (3) distinct land-use zoning; (4) a more compact and dense layout [5].

1.2.2. Influence of Natural Environment on Residential Design

The natural environments of Russia and Japan significantly impacted residential design. In Russia, harsh climates led to designs focused on insulation, with small windows and large depths to retain heat, while stoves and fireplaces served as central heating [6]. In contrast, Japan’s humid climate led to designs emphasizing earthquake resistance and ventilation, with wooden, lightweight materials becoming common. While Russia’s environment is similar to Jilin’s, Japan’s differs, especially in climate. Initially, Japan used temporary wooden houses but later adopted brick construction with insulation techniques from Russia and Northeast China [7].

1.2.3. Internal Family Structure and Religious Culture

In terms of internal family structure, Russian homes featured living rooms, dining rooms, and kitchens as spaces for family gatherings, celebrations, and socializing, with various items displayed to create a warm atmosphere. The separation of spaces for servants and the household reflected a clear hierarchical relationship [8]. In Japan, the family structure gradually transitioned from a feudal system to a more modern structure, especially with the development of standardized housing, which eliminated traditional gender and social distinctions within the household. For example, the separation of spaces such as the living room and kitchen in Japanese homes reflected traditional gender roles but was modified through the design of “movement lines” to reduce domestic labor [9].

In religious culture, Russia constructed an Orthodox Church in Dehui to serve Russian employees of the Chinese Eastern Railway and their families, meeting their religious needs [10]. Japan’s religious architecture, such as the construction of the Seihon-ji Temple in Changchun, served primarily Japanese expatriates, military personnel, and colonists, spreading Japanese Buddhist sects and facilitating ideological infiltration during military expansion. In Japanese residential design, religious elements were reflected in the design of alcoves for altars and ancestor worship, as well as the veneration of nature spirits [11]. Japanese living rooms were typically located near these alcoves, and the seating arrangement often faced the alcove, with only elderly males allowed to sit with their backs to it. This spatial division highlighted gender and status distinctions within the household, reflecting Confucian influences on family hierarchies [12].

1.2.4. Standardized Construction of Residential Buildings

Russia: (1) Standardization of residential types was based on the residents’ social status. Single-family homes were typically occupied by military officers or stationmasters, while multi-family homes were used by single railway employees [8]. (2) Construction was standardized, with materials selected locally to reduce transport costs. Many of the builders were Chinese peasants who had fled famine in northern China, so the construction was not highly standardized. Single-family and duplex homes featured compact or inner-corridor layouts. Three-family homes had a “T” shaped floor plan, with “one large and two small” rooms, while four-family homes featured a “+” shaped layout. Six-family homes retained only kitchens and bedrooms as functional spaces [13].

Japan: (1) Standardization of residential types was based on the employees’ income, divided into five types. Special-Grade, Grade A, and Grade B were single-family high-end homes, while Grade C and Grade D were four-family linked homes, with configurations based on family size; (2) Construction was standardized using prefabricated materials assembled onsite. Special-Grade, Grade A, and Grade B houses commonly featured inner-corridor layouts. The corridors and entrance halls of Japanese homes differentiated between “public” and “private” spaces, clearly delineating functional areas, with the living room being the largest space. In Grade C homes, the two bedrooms were adjacent, with an “L” shaped corridor that connected both ends of the house, providing adequate ventilation and lighting [14].

1.2.5. Quantitative Methods in Architectural Heritage and Urban Space

Space syntax, developed by Bill Hillier’s team at University College London, quantifies spatial layouts to reveal patterns of spatial usage, interaction, and social behavior. It has been widely applied in heritage and residential studies to explore the relationship between spatial features and cultural practices, such as the ceremonial layout of traditional courtyards [15], social spatial logic in Suzhou gardens [16], and the cultural continuity of vernacular architecture in the Yunnan-Tibet region [17]. Space syntax often incorporates statistical methods like hierarchical clustering, normalization, and regression analysis. It is commonly used in housing studies through Justified Plan Graph (JPG), applied to analyze layouts of Indonesian apartments [18], Edo period samurai houses [19], and Mexican social housing [20]. Additionally, space syntax combined with GIS has been used to examine issues like tourism, commercial organization [21], and the relationship between urban spatial hierarchy and crime rates [22]. This highlights GIS as a versatile and valuable spatial analysis tool.

1.2.6. Summary

In conclusion, Russian and Japanese residential layouts in Northeast China reveal significant differences, influenced by military needs, natural environments, family structures, and religious culture. A summary of the relevant literature is presented in Table 1. Russian planning was less rigorous, focusing on military needs, while Japan introduced more refined zoning and standardized construction systems. Space syntax provides valuable insights into these spatial patterns, offering tools for comparative analysis and future preservation efforts.

Table 1. Summary of the relevant literature.

Relevant Research Topics	Main Points	Scholars and Publication Time
Planning Models of Russian and Japanese Settlements	The planning of the Russian Empire was simple, focusing on military needs; Japan, based on this foundation, further optimized it, with clear land zoning.	Sun, R. (2013) [2]; Niu, X. (2014) [3]; Binko, M. (1995) [4]; Li, Y. (2010) [5]
Influence of Natural Environment on Residential Design	Russian design emphasized warmth and fire cores; Japanese design focused on earthquake resistance, ventilation, and lightweight materials.	Li, Z. (2025) [6]; Feng, L. (2025) [7]
Internal Family Structure and Religious Culture	Russian housing strongly emphasized space separation, with clear boundaries for family gathering areas; Japanese housing was more separated by function and cultural differences, especially in religious spaces.	Park, M. (2025) [9]; Baiz, Z.H. (2025) [8]; Feng, L. (2025) [10]; Zhang, S. (2025) [11]; Li, X. (2025) [12]
Standardized Construction of Residential Buildings	Russian housing standardized housing types but with simple layouts, leading to poor lighting; Japan, on the other hand, standardized housing based on income, using prefabricated materials for more efficient layouts.	Baiz, Z.H. (2025) [8]; Zhang, S. (2025) [13]; Feng, L. (2025) [14]
Quantitative Methods in Architectural Heritage and Urban Space	Space syntax was applied to housing standards, examining the relationship between space and cultural systems, with statistical methods applied in heritage and urban studies.	Mahdadi, N. (2025) [15]; Della Spina, L. (2025) [16]; Lei, Z. (2025) [17]; Ogunnusi, M. (2025) [18]; Rivera-Pabón, T. (2021) [19]; Acampa, G. (2025) [20]; Baiz, Z.H. (2025) [21]; Zhang, Y. (2022) [22]

1.3. Research Objectives

This study has three main objectives. First, it seeks to establish a spatial culture of Russian and Japanese railway residential architecture through comparative research, exploring the cultural factors behind their differences. Second, it systematically compiles and organizes floor plans of railway residential buildings from the Russian and Japanese periods in Jilin Province, offering detailed data support for the preservation and restoration of architectural heritage in the region. Finally, it examines the temporal and spatial distribution of Russian and Japanese railway residential buildings and investigates the spatial evolution and morphological features of these residential spaces during the Russian and Japanese colonial periods.

2. Materials and Methods

2.1. Research Scope

Jilin Province, located in Northeast China, houses railway heritage across the central plains. The region experiences long, cold winters and short, warm summers (see Figure 1). Within the railway network, stations range from first-class to fifth-class, serving distinct functions, with variations in station scale, architectural type, and building quantity. Residential buildings include both railway employee housing and military housing, which are distinctly differentiated in urban planning, although their floor plans do not always reflect these differences.

Figure 1. Geographic Location Map of Jilin Province.

The research samples were collected through a two-phase approach: a literature review and a field survey. The first phase involved gathering data from local archives at Jilin Provincial Library and the China National Knowledge Infrastructure (CNKI). The second phase consisted of field surveys, where mobile phones and cameras were used to photograph building exteriors. Building elevations and floor plans were measured using tape measures and laser rangefinders, and the data were digitized with AutoCAD 2025. The criteria for sample selection were: (1) buildings constructed during the Russian or Japanese occupation period, specifically between 1899 and 1931; (2) buildings clearly identified as Russian or Japanese constructions; (3) buildings in good condition, with minimal damage, for sampling; (4) representative building types, with a focus on single-story residential buildings. Ultimately, identifiable buildings were selected, and data were recorded based on the literature review and field surveys (see Table 2).

Table 2. Comparison of Russian and Japanese National Residential Cultures.

ID	Typology	Period	Site	Latitude and Longitude	Current Status	Source
1	Russian	1899–1904	Changchun	g125.31420605, 43.92543070	vacant and unused	[8] p. 122
2	Russian	1899–1904	Buhai	g125.62650864, 44.38209943	retain residential use	[8] p. 122
3	Russian	1899–1904	Deihui	g125.71958515, 44.53618121	retain residential use	[8] p. 122
4	Russian	1899–1904	Deihui	g125.71951479, 44.53283380	retain residential use	[8] p. 122
5	Russian	1899–1904	Buhai	g125.62664827, 44.38234485	retain residential use	[8] p. 130
6	Russian	1899–1904	Taolaizhao	g125.94172224, 44.82593167	retain residential use	[8] p. 130
7	Russian	1899–1904	Gongzhuling	g124.81032850, 43.51121032	retain residential use	[23] p. 141
8	Russian	1899–1904	Taolaizhao	g125.94104011, 44.82636826	retain residential use	[23] p. 141
9	Russian	1899–1904	Gongzhuling	g124.81481009, 43.51326825	retain residential use	[23] p. 142
10	Russian	1899–1904	Gongzhuling	g124.80748779, 43.50994097	retain residential use	[23] p. 142
11	Russian	1899–1904	Mishazi	g125.47893380, 44.14756413	retain residential use	[23] p. 142
12	Russian	1899–1904	Gongzhuling	g124.80996443, 43.51132956	retain residential use	[23] p. 143
13	Russian	1899–1904	Gongzhuling	g124.80983588, 43.51100496	retain residential use	[23] p. 143
14	Russian	1899–1904	Gongzhuling	g124.81525715, 43.51346585	retain residential use	[23] p. 143
15	Japanese	1905	Taojiatun	g124.98681022, 43.64326276	retain residential use	investigation
16	Japanese	1905	Caijia	g124.66588835, 43.40797949	retain residential use	investigation
17	Japanese	1905	Caijia	g124.66610631, 43.40860343	retain residential use	investigation
18	Japanese	1905	Caijia	g124.66599333, 43.40851164	retain residential use	investigation
19	Japanese	1906	Changchun	Lilac Street District or Comrade Street District	vacant and unused	[24] p. 4
20	Japanese	1906	Changchun		vacant and unused	[14] p. 52
21	Japanese	1906	Changchun		vacant and unused	[14] p. 52
22	Japanese	1906	Changchun		vacant and unused	[14] p. 52
23	Japanese	1907	Changchun		vacant and unused	[25] p. 18
24	Japanese	1907	Changchun		vacant and unused	[25] p. 55
25	Japanese	1907	Changchun		vacant and unused	[25] p. 55
26	Japanese	1915–1931	Gongzhuling	g124.82175133, 43.51701918	disused	investigation
27	Japanese	1915–1931	Gongzhuling	g124.82122669, 43.51679454	disused	investigation
28	Japanese	1915–1931	Gongzhuling	g124.81894591, 43.51479159	disused	investigation

The buildings’ identity was determined from three sources:

(1)

Archival records: Two census reports from the local history section of Jilin Provincial Library detailed the builders, housing types, materials, and functions of buildings along the railway stations.

(2)

Plan provenance: Floor plans were sourced from books and academic papers.

(3)

Field evidence: Architectural style and exterior decoration helped confirm the builder’s identity. For example, Japanese buildings feature elements like hipped roofs, straight chimneys, and linear decorations, while Russian buildings include corner stones, shadow play, and arches.

2.2. Research Subjects

2.2.1. Data Processing

Representative samples were extracted from the collected data, and spatial names were mapped using abbreviations (see Abbreviation). A total of 28 floor plans were studied, including 14 Russian and 14 Japanese residential layouts.

2.2.2. Russian Residential Floor Plans

Russian residential floor plans tend to be regular, typically in rectangular or “L” shapes (see Figure 2). These layouts are relatively enclosed, with clear functional zones. Living rooms, dining rooms, and bedrooms are generally separated to ensure privacy.

2.2.3. Japanese Residential Floor Plans

Japanese residential floor plans are often irregular, typically in “凸” or “凹” shapes (see Figure 3). These spaces are highly adaptable, emphasizing multifunctionality. In traditional Japanese homes, sliding doors were often used to merge rooms into larger spaces for various activities. Even after Westernization, Japanese homes prioritized space efficiency, with open layouts that made the interiors feel more expansive.

2.3. Research Methods

2.3.1. GIS Technology

During the field survey, the geographical coordinates (latitude and longitude) of the buildings were determined and marked on OMAP. In ArcGIS 10.8, the coordinates were converted to WGS 1984, and spatial features related to heritage changes were recorded. Kernel density analysis was used to generate a continuous density surface, aiding the analysis of the spatial clustering of railway residential heritage (see Table 3). The bandwidth was set to 12 km, with adjustments based on local town spacing. Projections were made using the CGCS2000 Gauss-Krüger Zone 22 to correct for distance distortions. The grid resolution was set to 600 m to match the region’s urban scale. These settings ensured scientific rigor and effective visualization of the analysis.

Table 3. Kernel Density Indicators.

Index	Formula	Formula and Parameter Description
Kernel Density Analysis (KDA)	$f\left(x\right)={\displaystyle \frac{1}{nh}}{\displaystyle \sum _{i=1}^n}K\left({\displaystyle \frac{x-x_i}{h}}\right)$	Where $n$ is the number of points in the dataset, $h$ is the bandwidth parameter, $x_i$ is the position of the data point, $K$ is the kernel function, and ${\displaystyle \frac{x-x_i}{h}}$ is the normalized distance between the data point and the target location.

Figure 2. Floor Plans and Layouts of Russian Imperial Residential Architecture.

Figure 2. Floor Plans and Layouts of Russian Imperial Residential Architecture.

Figure 3. Floor Plans and Layouts of Japanese Residential Architecture.

Figure 3. Floor Plans and Layouts of Japanese Residential Architecture.

2.3.2. Space Syntax

(1) Justified Plan Graph (JPG)

In space syntax, Justified Plan Graph (JPG) is used to describe and analyze building floor layouts (see Figure 4) [26,27]. It aids in understanding space organization and functional distribution by converting architectural spaces into a graph structure. The core of JPG is geometric connectivity, which reveals the relative relationships between spaces, their organization, and their interactions within the overall structure. In this study, we established JPG rules, including methods for dividing spaces (see Table 4), processing spaces (see Table 5), and inclusion/exclusion criteria (see Table 6) [28,29]. The AGRAPH software, developed by Delft University of Technology, was used to support data validation and graph generation (version 3.0) [30,31].

Figure 4. Schematic Diagram of JPG (Justified Plan Graph).

Table 4. Different Spatial Processing Methods.

Spatial Name	Spatial Processing Method
Door	The door itself is not regarded as a node; it only serves as a basis for judging whether two convex spaces are “directly connected”. If there is a passable door/opening between two convex spaces (regardless of the door’s size or type), the corresponding nodes are connected with a “line” in the graph.
Corridor	If the corridor is “straight” or “regular (without depressions)” and meets the definition of a convex space, it can be treated as a single node and connected with the rooms at both ends of the corridor by lines. If the non-convex corridor has a clear “functional integrity” (e.g., serving as a single passageway with no secondary spatial division), some studies will also ignore its non-convex shape and treat it as a single node. This study follows this approach.
Foyer	If the foyer is an independent convex space (e.g., entry foyer, lobby), it is directly treated as a single node and connected with all spaces (e.g., outdoor area, corridor, rooms) that it links to. If the foyer and adjacent spaces together form a convex space with no obvious physical separation, they are counted as two nodes (not merged). This is because the Japanese foyers in this study have epochal symbolism and functionality.

Table 5. Principles of Inclusion Criteria.

Inclusion Criteria	Meaning Explanation
Passability	Areas that provide actual passage or staying functions for human activities, including rooms, corridors, foyers (split by convexity), and outdoor public spaces (e.g., the stayable area outside the building entrance, regarded as the “external node ⊕”).
Convexity Adaptation	Conforms to the definition of a convex space (or can be split into convex spaces) and can serve as an independent topological unit to participate in connectivity calculation.
Functional Independence	Even if the space is extremely small (e.g., a small storage room), it must be included as a node as long as it meets the requirements of passability and convexity. This situation only applies to residential houses in this study.

Table 6. Principles of Exclusion Criteria.

Exclusion Criteria	Meaning Explanation
Non-spatial Elements	Physical obstacles such as walls, columns, furniture, and door/window frames are not regarded as nodes.
Impassable Areas	Regions that cannot be accessed by humans, such as the interior of fixed storage cabinets, wall interlayers, and pipe shafts.
Duplicate or Redundant Spaces	Sub-areas within the same convex space (with no physical separation and belonging to the same convex space) are not split into separate nodes. For example, in this study, the integrated dining and living area in the Shabo residence is counted as one node.
Externally Unrelated Spaces	Only the “outdoor public space directly accessible from the building entrance” is included as the “external node ⊕”; other non-public areas outside the building are excluded.

(2) Relevant Indicators

The software used for space syntax is DepthmapX 0.8.0, with clearly defined indicators (see Table 7).

Table 7. The main indicators in space syntax.

Index	Formula and Parameter Description	Formula and Parameter Description
Total Depth (TD)	$TD=\left(0\times n_x\right)+\left(1\times n_x\right)+\left(2\times n_x\right)+\cdots +\left(x\times n_x\right)$	The sum of the number of connections between a particular node (the carrier or root) and every other node in the set, weighted by depth (L). $n_x$ is the number of nodes at a specific level, and L is the level number (0, 1, 2, …).
Mean Depth (MD)	$MD={\displaystyle \frac{TD}{(K-1)}}$	The average degree of depth of a node in a justified plan graph. TD is the total depth, and K is the total number of nodes (excluding the node itself).
Relative Asymmetry (RA)	$RA={\displaystyle \frac{2\left(MD-1\right)}{K-2}}$	A normalized measure of how deep a system is (for a given carrier) relative to a symmetrical or balanced model of the same system, with a value range of 0.0–1.0.
Integration Value (i)	$i={\displaystyle \frac{1}{RA}}$	A measure of the degree of integration (relative centrality of spaces) in a system, which is the reciprocal of Relative Asymmetry (RA).
Control Value (CV)	$CV\left(a\right)={\displaystyle \sum _{D\left(a,b\right)=1}}{\displaystyle \frac{1}{Va\left(lb\right)}}$	A measure of the degree of influence each node has in a system. $D\left(a,b\right)=1$ means node a is directly connected to node b, and $Va\left(lb\right)$ is the number of connections of node b $\left(N{C}_n\right)$.
Connectivity (C) or Node Connection number $(N{C}_n$)	——	The number of spaces directly connected to a node (no specific formula, it is a counting index).

By combining the Connectivity and Mean Depth indicators, we analyzed the building layout in terms of spatial flow and hierarchy [28]. This combination reveals aspects of spatial layout such as hierarchical relationships, functional distribution, usage efficiency, and social interaction, highlighting the space’s uniqueness.

Using Integration (Zi) and Control Value (CV), we examined spatial layout changes from centrality and control perspectives [28]. This approach shows how space adjusts its role and control within the layout as social and functional factors evolve. Integration helps us understand spatial flow and connectivity, while Control Value assesses the influence of space and its functional dominance. Together, these indicators provide a comprehensive view of how residential spaces evolve with social and functional changes.

Space syntax data were manually extracted and organized in Excel after software calculations. Due to the large number of spatial units, directly exporting the data from the software resulted in a chaotic dataset. Therefore, the data were manually entered into Excel, classified, and then imported into OriginPro 2025 for graphing.

The specific process for space syntax operations was as follows: the CAD format was exported as dxf, and a New Map was created using the Convex Map option. Tools were selected as Axial/Convex/Pesh, and “Run Graph Analysis…” was chosen. The Radius was set to “n” for global analysis. The options “Include local measures,” “Include RA, RRA, and total depth,” and “Include weighted measures” were checked.

2.3.3. Z-Score Normalization

Z-score normalization was applied to compare the Integration (Zi) and Control Value (CV) indicators, eliminating scale differences between the two metrics. This approach ensures consistent comparison of both indicators (see Table 8) [32]. It reveals distinct spatial features and trends while allowing a comprehensive examination of spatial changes in terms of centrality and control. It also helps reveal how space adjusts its role and control as social and functional shifts occur.

Table 8. Z-score Indicators.

Index	Formula	Parameter Description
Z-score	$Z={\displaystyle \frac{X-\mu }{\sigma }}$	In this context, $X$ is the raw data value, $\mu$ is the mean of the raw data, and $\sigma$ is the standard deviation of the raw data.

2.4. Research Framework

The research framework is illustrated in Figure 5, which outlines the integration of GIS and space syntax for analyzing the spatial characteristics and cultural implications of Russian and Japanese railway residential heritage.

Figure 5. Research framework.

3. Results

3.1. Residential Distribution Analysis

We analyzed the temporal and spatial distribution characteristics of the residential architectural heritage (see Figure 6) to identify the time periods of each building. Russian residential construction began after the Russian Empire laid the railway tracks in Jilin Province in 1899, continuing until the Russo-Japanese War in 1905, when Russia withdrew to the north of Changchun. Therefore, the period from 1899 to 1904 corresponds to Russian residential construction, while the period from 1905 to 1931 corresponds to Japanese residential construction. Russian residences were more widely scattered along the railway, with stations positioned nearly equidistant from one another. Notable stations such as Dehui, Kuan Chengzi, and Gongzhuling demonstrate clear residential clustering. Residential distribution is relatively sparse, with concentration mainly around key transportation nodes, forming a linear pattern. In contrast, Japanese residences, which expanded upon the existing Russian infrastructure, are more concentrated and clustered at specific stations like Kuan Chengzi and Guojia Dian.

Figure 6. (a) Distribution Map of Russian Imperial Residential Architecture; (b) Distribution Map of Japanese Residential Architecture.

3.2. Spatial Structure Analysis

We converted the spatial layouts of 28 residential buildings into visual spatial maps to explore the relationship between their spatial nodes, line connections, structure, and social effects (see Figure 7). Since all residential samples are single-story buildings, they feature shallow tree-like or circular layouts.

Figure 7. JPG Diagrams of Russian Imperial and Japanese Residential Architecture.

Single-family homes were typically inhabited by individuals of higher status, such as military officers or stationmasters. Duplex, triplex, and fourplex homes were occupied by ordinary railway workers or soldiers, with fourplex homes mainly housing single individuals. A common feature across all housing types is that single-family homes tend to have deeper spatial layers, more spatial nodes, and a broader range of functions, while multi-family homes generally have fewer nodes and simpler, more essential functions. However, there are notable differences between Russian and Japanese homes:

(1)

Circular Layouts: Russian homes with circular layouts (Residences 1, 2, and 3) feature multiple “d” spaces. For example, Residence 2 includes a dining room (DR) and foyer (F), while Residence 3 includes a corridor (C), living room (LR), and master bedroom (MB). These spaces exhibit greater fluidity and serve as key transition and connection points. The “c” spaces are primarily kids’ rooms (KR), secondary bedrooms (SBR), and study rooms (SRO), which surround the “d” spaces and provide secondary functions. Japanese homes, however, do not have the “d” space. For instance, Residence 19 has only “c” spaces, including a sunroom (S), living room (LR), tea room (TR), and corridor (C), while Residence 24 includes a master bedroom (MB), kitchen (K), entry hall (EH), and primary storage room (PSR). The difference in layout may be due to Residence 24’s smaller area and later construction, leading to a more space-efficient design. Residence 24 also has a secondary foyer (SF) at the back, connecting key functional spaces, whereas Japanese homes feature clearer functional zoning.

(2)

Non-circular Independent Homes: A comparison between Residences 4 and 25 shows that in Russian homes, all space nodes are connected through the corridor (C). The single functional space has a small area but is supplemented with auxiliary spaces, such as a cloakroom (CR) connected to the bathroom (SR), and a toilet (T) linked to secondary storage (SSR). In Japanese homes, the living room (LR) distinguishes two distinct functional layouts. The kitchen (K) acts as the secondary hub, connecting the main spaces, while the corridor (C) leads to more private spaces. Residences with dual functional zones, such as Residence 1, do not exhibit a clear distinction in layout functions. These comparisons suggest that Japanese homes, even in compact layouts, have clear functional zones centered around the living room (LR) and kitchen (K), while Russian homes focus more on connectivity, with major spaces flowing seamlessly but lacking clear functional groupings.

(3)

Tree-like Layouts: In duplex homes, the corridor (C) serves as the central hub. However, differences arise between the two countries:

• Only the hybrid Japanese homes built in 1905 (Residences 16-A, 17, 18) have a simple, single-line tree-like structure with no branches. These homes feature poor privacy and compact layouts, and were not built to standardized designs. Despite a shorter occupation period, Russian homes adhered to a unified, standardized style.
• In Russian homes, bathrooms (SR) and cloakrooms (CR) may have separate entrances and are not directly connected to the main living areas. In contrast, Japanese homes often do not directly connect the bathroom (B) and kitchen (K) to the main living areas.
• When comparing the complexity of residential functions, Russian homes with more functions use the corridor (C) as the circulation hub, while simpler homes rely on the foyer (F) as the center. This is not the case in Japan, where homes evolved across different periods: in 1905, single-line tree-like layouts dominated, followed by multi-branch designs in 1906, introducing greater flexibility. By 1907, a combination of tree-like and branching layouts emerged, with more defined functional zoning. From 1915 to 1921, standardized Japanese homes became more sophisticated, integrating foyers (F) and corridors (C) and enhancing privacy and efficiency.

Japanese homes also became more space-efficient, with greater functional grouping. Field research photos (see Figure 8) reveal that although space remains narrow, storage and movement lines are more efficiently arranged.

Figure 8. Interior Space Atlas of Japanese Residential Architecture.

Russian homes included three- and four-family units, while Japanese homes had only three-family units with a single design, preserving the most critical functional spaces (see Figure 9). Key differences include:

Figure 9. Interior Space Atlas of Russian Imperial Residential Architecture.

(1)

Space Division: Russian homes separate the kitchen (K) from other spaces, with bedrooms located at the far end. Japanese homes feature independent spaces, without connections between them.

(2)

Foyer Usage: Russian homes eliminate the corridor (C), with all spaces connected via the foyer (F). In Japanese homes, the foyer (F) is minimized, only existing at the entrance, connecting directly to the corridor (C). This contrasts with the prominence of the foyer (F) in the Edo period and earlier.

In conclusion, through comparative analysis of Russian and Japanese residential layouts, the differences in spatial organization, functional division, and movement flow are clearly revealed. Russian homes emphasize flow and connectivity, with circular layouts and simpler functional separations. In contrast, Japanese homes prioritize space efficiency and functional grouping, especially between 1905 and 1931, with clearer zoning and integration of the foyer (F) and corridor (C) to enhance privacy and efficiency. These differences reflect the cultural and historical development of residential design in both countries.

3.3. Spatial Accessibility Analysis

Connectivity directly reflects the number or degree of connections between a spatial unit and other spatial units. It influences accessibility, which refers to the ease of reaching other spatial units from a given unit. We analyzed the connectivity of spatial nodes in both Russian and Japanese homes by creating heatmaps to examine spatial layout patterns and accessibility (see Figure 10). We then compared the average connectivity values of each spatial unit (see Table 9) to explore differences in spatial layout.

Figure 10. Connectivity Heatmap of Residential Spaces in Russian Imperial and Japanese Architecture.

Table 9. Average connectivity ranking of residence.

Object	Average Connectivity Ranking
Tsarist Russian residence	C (2.24) > F (2) > MB (1.53) > K (1.18) > LR (0.88) > LD (0.76) > DR = T (0.53) > SRr (0.47) > SR = MR (0.29) > PSR = SRO (0.24) > KR = CR = S (0.18) > SF = EH = TR (0)
Japanese residence	C (2.67) > K (1.73) > LR (1.6) > MB (1.27) > F(1) > SBr (0.53) > EH = B = SR = CR (0.33) > PSR (0.27) > SF = DR = T (0.2) > MR = TR = S (0.13) > LD = KR = SRO (0)

From the heatmap, a common feature is that the space with the highest heat value in both Russian and Japanese homes is the corridor (C) in single-family homes, as this space serves as the key connection point between various rooms. However, differences emerge as follows:

(1)

In Russian homes, the key spaces with high connectivity are the foyer (F), master bedroom (MB), and kitchen (K), reflecting high accessibility. In contrast, Japanese homes show that connectivity more closely reflects the frequency and proximity of activity, with the kitchen (K), living room (LR), and bedroom (MB) having the highest connectivity, while the foyer (F) is lower in comparison.

(2)

Unique spaces in Russian homes include the open-plan living and dining area (LD), children’s room (KR), and study room (SRO). Unique spaces in Japanese homes include the secondary foyer (SF), the rear entrance, as well as the hall (EH) and tea room (TR). These unique spaces are present in higher-tier homes, where the hall (EH) serves as a transitional space and reflects differing cultural and lifestyle practices, with children and women living in the spaces indicated.

We used chord diagrams to show the connections between pairs of spatial nodes (see Figure 11). The main differences between Russian and Japanese homes are:

Figure 11. Direct Connectivity Diagram of Spatial Nodes.

(1)

In Russian homes, the spaces with the most directional color indicators are the dining room (DR), toilet (T), children’s room (KR), and master bedroom (MB), each having four colors. In Japanese homes, the kitchen (K) is connected to five other spaces, with the master bedroom (MB) connected to four. This suggests that dining rooms are less common in Japanese homes, as the kitchen (K) serves dual functions, and other spaces often have multifunctional roles. On the other hand, Russian homes often feature adjacent spaces with more distinct functional divisions.

(2)

In Russian homes, it is common to separate the toilet (T) and bathroom (SR), while in Japanese homes, these are combined into a single bathroom (B).

(3)

In Russian homes, lower-tier homes often combine the dining room (DR) and living room (LR), whereas in Japanese homes, the kitchen (K) encompasses all household labor functions.

(4)

In Russian homes, there are often maid’s rooms (MR), some of which have independent entrances, separated from the residential spaces. In contrast, Japanese homes have only one independent home with such a room, and it is located within the residence itself.

By analyzing the average depth (MD) of the spaces (see Figure 12), we compared the differences between Russian and Japanese homes: The range distribution of average depth (MD) in Russian homes shows strong hierarchical differences. Core functional areas such as the living room (LR) and master bedroom (MB) are often found in low MD ranges (1.790 < MD ≤ 2.148), while service spaces like the maid’s room (MR) are concentrated in higher MD ranges (2.506 < MD ≤ 2.864), with isolated “V(MD) = 0” areas. This suggests that Russian homes use topological distances in MD to strictly layer the accessibility of master-servant spaces, reinforcing the spatial expression of the hierarchical system. In contrast, the distribution of average depth (MD) in Japanese homes is more homogeneous. Core spaces such as the living room (LR), kitchen (K), and associated spaces like the master bedroom (MB) are concentrated in the MD ranges of 1.376 < V(MD) ≤ 1.720 and 1.720 < V(MD) ≤ 2.064, without distinct high MD isolation zones. Even though functional zones exist, the topological accessibility differences between spaces are smaller, reflecting the layout logic in Japanese homes that emphasizes functional integration.

Figure 12. (a) Mean Depth Heatmap of Russian Imperial Residential Spaces; (b) Mean Depth Heatmap of Japanese Residential Spaces.

Therefore, by analyzing the connectivity and average depth of Russian and Japanese homes, significant differences in spatial layout and accessibility were revealed. Russian homes exhibit a clear hierarchical spatial layout, with considerable accessibility differences between core functional areas and service spaces. In contrast, Japanese homes focus more on functional integration, with smaller differences in accessibility between spaces, resulting in a more balanced layout.

3.4. Spatial Centrality and Flow Control Analysis

Integration reflects the extensive connectivity and accessibility of a space. A space with high integration occupies an important position in the spatial network, indicating that it is a high-traffic area. Control value, on the other hand, provides a perspective on how a space controls or limits flow, describing the power structure of certain spaces within the layout. For example, a space with a high control value, even with a low integration value, may still be an important functional or social activity center. We compared the proportions of integration and control value in each residential space (see Figure 13).

Figure 13. (a) Proportional Chart of Spatial Integration in Residential Architecture of Both Countries; (b) Proportional Chart of Spatial Control Values in Residential Architecture of Both Countries.

By conducting a vertical comparison, i.e., comparing each residential unit, Russian homes with a high integration proportion but low control value include Residences 10-A, 4, and 6, while Japanese homes include Residences 19, 20, 24, 25, 26, 27, and 28. Regardless of the housing type or builder, these homes in the JPG diagrams have multiple branching nodes at the control center, showing basic floor layouts. The greater the number of spaces connected to a single node, the larger the discrepancy between the two values, as seen in Residences 6 and 19. Overall, Japanese homes tend to have more control nodes, indicating higher spatial functional correlation and more distinct grouping.

When comparing the proportions of each spatial node, the largest proportions are seen in the foyer (F), corridor (C), living room (LR), and kitchen (K). We selected the values for these spaces, performed Z-score calculations, and made comparisons (see Figure 14).

Figure 14. Comparison of Integration and Control Values for the Entrance (F), Corridor (C), Living Room (LR), and Kitchen (K) in Residential Architecture of Both Countries.

Low integration and low control values indicate spaces on the periphery, auxiliary functions, or spaces with a high degree of privacy. We focused on analyzing spatial nodes with low integration but high control values. The details are as follows:

(1)

In the foyer (F): Residence 14 in Russian homes shows extreme values (Vi = 3.14, Vcv = −1.63). The JPG diagram is similar to Residence 9-B, but the values differ significantly. This is mainly because Russian homes saved space by having one foyer shared by two units.

(2)

In the corridor (C): High connectivity naturally leads to high control values. A special case is Residence 22, where the bathroom (B) and kitchen (K) are located at the two ends of the residence, extending the usable space.

(3)

In the living room (LR): Residence 22 shows a similar situation. Additionally, in Residence 25, the living room (LR) has a low control value, but the integration value is unique, as the kitchen (K) is the main hub, and the living room (LR) serves as a secondary space. In Residence 2, the living room (LR) is located in a non-central position within the circular layout but has strong flow, also showing the low integration and high control value scenario.

(4)

In the kitchen (K): In Residence 2, the kitchen (K) is primarily used by the maid, not connected to the main living spaces but linked to the maid’s room (MR), primary storage room (PSR), and secondary storage room (SSR). In contrast, Russian single-family homes typically have high integration in the kitchen (K) but low control values, while the remaining duplex, triplex, and fourplex homes generally have both low integration and low control values, indicating that the kitchen (K) is not a central functional space. In Japanese homes, however, the kitchen (K) almost always has high integration and low control values, suggesting that other spaces have clearly defined functional divisions, and the kitchen (K) serves multiple purposes, including dining and storage.

From the analysis, we can draw several conclusions. Whether in Russian or Japanese homes, the corridor (C) serves as the central passage, connecting various rooms. In duplex and triplex homes, Russian designs save space by directly connecting the foyer (F) to the rooms, eliminating the need for a corridor (C). In contrast, Japanese homes reduce the foyer (F) size and merge it with the corridor (C), distinguishing them only in terms of layout. The living room (LR) is primarily used for daily activities in Russian homes, whereas in Japanese homes, there is a clear functional distinction between the kitchen (K) and living room (LR).

4. Discussion

4.1. Residential Distribution Characteristics Based on GIS

GIS technology was used to reveal the spatiotemporal distribution characteristics of Russian and Japanese homes. Russian residential areas demonstrate a multi-core clustering pattern, while Japanese homes show a denser, point-like distribution. This difference likely results from Russia’s use of standardized planning and Japan’s emphasis on compact, adaptive design. Stations like Dehui and Gongzhuling, both third-class stations, are located in high-density areas. Russian railway stations were spaced at regular intervals, with higher-class stations featuring more residential buildings. These homes were built according to standardized designs, with ample space between them. In contrast, Japan expanded residential areas around Russian stations, following the original designs, which optimized both cost and land-use.

4.2. Spatial Layout Patterns Based on JPG

The JPG diagrams illustrate the spatial layout patterns. Russian homes emphasize shared spaces with clear functional divisions, while Japanese homes focus on clear movement lines and space efficiency. High-end Russian residences, such as Residences 2 and 3, feature circular layouts that facilitate space sharing but result in some functional overlap, reflecting a focus on centralizing functions due to the harsh climate. Common residences like Residences 4 and 6 utilize multiple branching nodes, suggesting a balance between functional flow and privacy. This design could be influenced by Russia’s vast land area and cold climate, which favored larger, enclosed layouts with centralized heating systems.

Japanese high-end residences, such as Residences 22 and 25, incorporate multi-branch layouts, with functional spaces grouped into zones and movement lines clearly defined. Common homes, such as Residences 17 and 18, feature linear, tree-like layouts, initially open to maximize space but lacking privacy. Later, the foyer (F) and corridor (C) were combined to save space while maintaining privacy (see Residences 26 and 27). This reflects Japan’s strong emphasis on orderly living, privacy, and space utilization. Given Japan’s small island size and frequent natural disasters, residential designs evolved post-Westernization, continuously improving functionality.

4.3. Spatial Functional Uniqueness Based on Connectivity and Mean Depth

The combination of connectivity and average depth provides insight into the uniqueness of spatial functions. Russian homes prioritize privacy with distinct functional spaces, while Japanese homes offer multifunctional spaces, with the kitchen serving as the central hub to support efficient, pragmatic living.

Russian homes include unique spaces like the children’s room (KR) and study room (SRO), while the living and dining rooms (LD) serve as important gathering areas. The toilet and bathroom are separated, and living and dining rooms are often combined to save space. Maid’s rooms (MR) have independent entrances (as seen in Residence 2), highlighting the social structure’s focus on hierarchy and family privacy.

In contrast, Japanese homes feature spaces like the tea room (TR) and hall (EH). The living room (LR) and kitchen (K) are key to daily life. The bathroom is not divided, and there is no separate dining room; the kitchen (K) integrates most household functions. Maid’s rooms (MR) are incorporated into the residential space (as seen in Residence 19), emphasizing multifunctionality and family interaction. The multifunctional use of the kitchen (K) indirectly reflects post-Russo-Japanese War social changes, including rising democratic ideals and improving women’s status, which contributed to more efficient and pragmatic family dynamics.

4.4. Spatial Control Relations Based on Integration and Control Value

The combination of integration and control value analyzes spatial control dynamics. Russian homes feature clear functional divisions with fluid movement, while Japanese homes exhibit functional groupings that are rationally linked.

Russian homes typically have fewer control nodes, with most spaces controlled by a single node (as seen in Residences 4 and 6), and a corridor-centered layout. In lower-class homes, the foyer (F) serves as the main connection point. This design is consistent with Russia’s colonization approach, which emphasized rapid construction and occupation.

In contrast, Japanese homes have more control nodes, with each node managing several rooms (as seen in Residences 22, 23, 24, and 25), ensuring efficient use of small spaces while maintaining a rational and interconnected layout. These homes adapted and improved upon the standardized construction methods from Russia. Over time, Japanese homes evolved towards modernist, standardized designs, with layouts centering around corridors (as seen in Residences 26, 27, and 28), allowing for more rational and efficient space use.

5. Conclusions

This study, based on a review of previous research on the planning of Russo-Japanese station areas, environmental adaptation, family life organization, and differences in standardized construction, examines 28 railway residential floor plans from Jilin Province. A comparison was made using GIS, spatial syntax, and Z-score normalization. The quantifiable results reveal stable patterns of differences at both the macro-distribution and micro-organization levels. The conclusions are as follows:

(1)

Differences in Distribution Patterns: Russian residential areas along the railway lines tend to exhibit multi-core clustering, whereas Japanese residential areas show fewer clusters but higher density. Russian layouts are more focused on linear expansion and node-based settlements, while Japanese designs prioritize concentration and integration based on existing infrastructure, aiming to reduce costs.

(2)

Differences in Spatial Structure Types: Higher-level Russian residences more commonly feature complex, ring-shaped layouts, whereas ordinary residences tend to have single-node, multi-branch designs dominated by corridors. In contrast, Japanese residential areas predominantly adopt tree-like and branching combinations, progressing in stages from a single, compact line towards multiple branches and more distinct zoning. Russian designs emphasize shared spaces and functional concentration, while Japanese layouts prioritize movement lines and spatial economy.

(3)

Differences in Accessibility and Hierarchy: Russian residences exhibit a stronger gradient in Mean Depth, with clearer isolation and hierarchical separation between core and service spaces. Japanese residences demonstrate more balanced depth distribution and stable connectivity. Russian designs emphasize functional independence and hierarchical privacy, while Japanese designs are characterized by functional integration, efficiency, and pragmatism.

(4)

Differences in Centrality and Control Models: Russian designs tend to centralize control in a few key nodes, while Japanese designs distribute control across multiple nodes, which then interact post-zoning. Z-score comparisons reveal that kitchens in Japanese layouts often become key organizational nodes. Russian standardization is more focused on typology and centralized connectivity, facilitating rapid construction and clear spatial isolation. Japanese standardization, on the other hand, leans towards seriality and modularity, promoting clearer functional groupings and node divisions within compact spaces, while positioning the kitchen at the heart of daily organization.

Overall, the literature review summarizes four cultural factors: planning models, natural environments, family structures, and standardized construction. The differences in core and node structures identified by GIS and the variations in spatial syntax concerning structure, accessibility, and control are discussed. These findings provide visual and quantifiable evidence, transforming the differences in residential spatial forms from mere descriptions into more concrete hypotheses or spatial indicators and patterns.

Although this study sampled 28 floor plans, residential buildings along the Chinese Eastern Railway and South Manchurian Railway extend beyond Jilin Province to include areas in Heilongjiang and Liaoning provinces, meaning that not all residential spatial characteristics were captured. Future studies could expand the sample size and types, incorporating thermal environment simulations, construction lineage, and residential use surveys to further enhance the explanatory power and generalizability of the findings.