Introduction

The notion of “formats” in reference to retail and food service business management first appeared in Japan in the 1960s. Studies from the United States generally define retail formats as “types of retailers” or “types of retail establishments,” while recent studies have specifically used “retail formats.” As the economy becomes increasingly complex, a variety of businesses and entertainment providers are gradually being included under this category. Consequently, the scope of the concept has expanded from the retail industry to envelop the service industry. As such, the term “format” is now used to express the extant or realized form of retail and service industry activities.

There is a plethora of studies on urban centers from different perspectives. Shi and Yang (2015), for example, have examined the central Xinjiekou District of Nanjing City, with a specific focus on the scale, distribution, and pattern of mixed land use. Relying on urban morphology, Dhar and Khirfan (2017) have proposed a theoretical framework to influence the resilience of the urban form, particularly in response to climate change. Ogle et al. (2017) have used clustering, compactness, connectivity, and density as variables in an algorithmic analysis of various aspects of the city on a decadal basis. Wang et al. (2011) have examined consumer behavior in Beijing’s Wangfujing Street and Shanghai’s Nanjing East Road, establishing mathematical models that investigate the effects of commercial spatial factors on consumer behaviors and exploring how business layout models affect consumer behavior. Banzhaf et al. (2017) have proposed that changes in land use are influenced by the relationships between environmental, social, and economic factors. Pili et al. (2017) examined long-term form–function relationships in compact and dispersed cities, offering an empirical base from which to identify sustainable urban forms. Moreover, while Valls et al. (2018) have determined a means of identifying the different uses and architectural requirements of an urban space; they note that such data and information can be challenging to retrieve, structure, analyze, and visualize.

The aforementioned studies predominantly focus on urban centers from the perspectives of spatial and traffic structure and shape, as well as industrial structure and morphology. The construction of different spatial theory models involving human geography, the industrial economy, transportation, construction, and other disciplines, has enriched and perfected the related theoretical research on urban center structures. While the development and evolution of the function of the urban center district is affected by many factors, those related to economic development have a greater impact on the functions of the central district. Given the limitations of science and technology, however, it is difficult to conduct a comprehensive and in-depth study of urban centers from an economic perspective. However, the arrival of the age of Big Data makes data acquisition easier—including that of format data. The functions of formats within the central district are a direct representation of the level of urban economic development and the central district of the city. The size of the space also has a decisive influence. With the development of big cities, the industrial structure of urban centers is becoming increasingly complex. In terms of the perspective of space, it is clear that the extant research on the central district is relatively one sided. As such, the introduction of formats into urban center analysis will address the deficiencies of current research methods.

Existing research on formats has focused on the format itself. Brown (2010), for example, has applied modern financial portfolio theory (MPT) to the managing portfolios of retail formats in order to explore different returns in retail form. Meanwhile, Fowler and Bridges (2010) have found that the majority of consumers expect high performance in all dimensions—from employee performance to process design, to shopping experience reliability—and dislike surprising or “risky” stores. Teller et al. (2012) have empirically evaluated the relevance of shopping-related logistics for consumers on store-based retail formats, concluding that while shopper logistics are relevant for grocery store consumers, this relevance varies between different shopper groups. Davisa and Hodges (2012) conceptualize consumer shopping value as having two components: a shopping trip value and an in-store shopping value. In their study, Han et al. (2017) empirically measured the potential spillover effects of the arrival of big-box retail on the productivity of incumbent retailers in the various entry regions, investigating how the impact differed from one situation to another. In doing so, they provided a quantitative discussion of the spatial form and structure of urban space from the interdisciplinary perspective, rather than from the economic perspective. While the aforementioned studies discuss format from the perspective of consumer behavior and economics, they do not explore the relationship between format, urban spatial structure, and urban form.

As such, this paper seeks to address the gap in this research by evaluating the industrial form of the urban central area—especially that of megacities. More specifically, it examines two central districts of Shanghai—the People’s Square and Lujiazui District—by identifying, unpacking, and examining their format characteristics. The remainder of this paper is structured as follows. The next section explains why this study focuses its analysis on Shanghai and introduces the research methods. The third section describes the basic characteristics of the spatial distribution of formats in central areas, while the fourth summarizes the basic rules of format spatial distribution in central districts and concludes the main findings of this study.

Case selection and methodology

The People’s Square and Lujiazui District in Shanghai were chosen as the research objects of this study. In terms of methodology, this study relies on the format-type integrated approach, format classification methods, and kernel density analytical method based on GIS software. An important foundation for the study of formats in central districts, the format-type integrated approach classifies all formats according to the format type and the different characteristics of the “National Economic Sector” (GB/T 4754-2002), linking urban planning with economics. The second method is the format classification method, which classifies formats hierarchically by observing the proportion of various types of business and their degree of influence in the central area. Finally, the kernel density analytical method is used to describe the spatial distribution characteristics of the central district. This method studies the spatial distribution of point density in regular regions and examines the distribution characteristics of these points. The density of a point is generally estimated by measuring the number of events per unit area in the study area.

Case selection

Following a comprehensive analysis of 13 cities and 20 typical central urban areas, the decision was made to focus this research on the People’s Square and Lujiazui District in Shanghai—an economically developed city and the center of Asia’s economy, as well as its trade and shipping industries. By the end of 2013, the urban population of Shanghai’s municipal districts was approximately 13.63 million. Moreover, the 2013 GDP of the municipal districts was about CNY 2.132 billion.

These two central areas of Shanghai—the People’s Square and Lujiazui District—each have their own development level and structures. As the main centers of a Chinese megacity, these two districts have some of the most concentrated construction scales and land use of any central urban areas in China. As such, focusing this research on the format characteristics in Shanghai provides a broader dataset than any other city and allows for the extrapolation of these findings to estimate the composition of other central urban areas in China.

Data acquisition method

This study used programming software to identify the position of a spatial point from the Baidu map website and an anthropic investigation method to correct it, thereby obtaining an accurate business function point. An ArcGIS software POI spatial digital map information point was then used; this overlapped with the previous map to create a “format spatial point position” map (Fig. 1). This map makes the business function more precise and accurate and is specific to the internal business functions of each building, thereby making the research deeper and more micro. The database includes 49,172 format organizations in central urban areas, and provides the names, addresses, and format types, in addition to other information.

Fig. 1
figure 1

Spatial distribution of X industry in Shanghai’s central areas (data source author plotted). The “format spatial point position” map (Fig. 1). The database includes 49,172 format organizations in central urban areas; it provides names, addresses, and format types, in addition to other information. The graph illustrates the distribution of various business types in Shanghai

Full size image

In order to facilitate an in-depth study of the formats of the two identified central areas, all the formats were sorted and classified according to the different characteristics of the format types and The Classification of National Economic Industries (GB/T 4754-2002) in China. In doing so, the different formats were divided into 15 categories (Tables 1, 2).

Table 1 Integration map of the service industry and industry classifications
Full size table
Table 2 Integration map of the service industry and industry classifications
Full size table

This study found that the central area of the People’s Square comprises a total of 35,5055 businesses, while Lujiazui District comprises 14,117. The spatial distribution features are shown in Fig. 2.

Fig. 2
figure 2

Total format proportions in Shanghai’s central areas (data source Author plotted)

Full size image

Format grading method

In order to equalize the distribution of the number of formats at each level and to appropriately reflect the degree of influence of each format on the central district, the format of the industry was divided into three levels according to their respective proportions: the first level denotes a format accounting for 10% or more, the second level reflects those accounting for 1–10%, while formats accounting for 1% are categorized as third level. As such, the first level indicates the main type of format in the central area and reflects its core characteristics. The second level is an important format of the central district and reflects the characteristics of the central district to a more limited extent. The third level is a component of the central area format; while it has less impact on the characteristics of the central district, it can increase the diversity of central functions. By grading formats in this manner, the composition and characteristics of formats in the central district can be discerned more clearly.

Kernel density analysis

There are two ways of analyzing point patterns for format spatial research: the distance-based method, which uses dispersion as the baseline (i.e., Ripley’s K function); and the density-based method, which uses aggregation as the baseline (i.e., sample analysis, kernel density estimation, hotspot analysis). Format can be regarded as a set of points in continuous space. While it is difficult to express the distribution characteristics of a large number of format spatial points based on the distance method, the density-based method can better express this distribution and has superior visualization. Therefore, this paper uses the kernel density analysis method to analyze the spatial distribution of industrial space in Shanghai’s central areas, employing GIS software for analysis.

Results

This section describes the spatial distribution of different formats in Shanghai’s large central areas. This study found that there are various types of formats in central areas and that they are large in scale. These data were examined closely to determine any differences in the number of business types, and how each business type was distributed in the central districts.

We categorized the formats into three levels in terms of their numbers depending on their prominence. The first level contained retail services, business offices, and catering services. The second included entertainment and recreation, finance and insurance, administrative management, culture and art, accommodation services, education and scientific research, as well as healthcare services. The third level encompassed sports services, market services, other services, traffic services, and social welfare (Table 3).

Table 3 The formats of the central region of Shanghai
Full size table

Data statistical results

Formats in the first group—retail services, business offices, and catering services—are predominantly operation oriented and account for more than 70% of the total formats. Indeed, they are the main type of formats found in Shanghai’s central areas; the vast majority of businesses in Shanghai’s two districts are one of the three aforementioned formats (Fig. 3). However, the two districts vary slightly in terms of their format composition: the proportion of retail services is the highest in the People’s Square (37.1% of total formats), while business offices are the most popular (30.4% of total) in Lujiazui District. It makes sense that these three formats are the most popular. Indeed, they not only provide people living in the central area with the ability to maintain their daily consumption expenditure, but also provide others from surrounding areas with places to work and spend in the central area (Fig. 4).

Fig. 3
figure 3

Kernel density analysis of the format spatial distribution of Shanghai. Both the distribution relation and the degree of clustering in different lots are examined through kernel density analysis

Full size image
Fig. 4
figure 4figure 4

Total format distributions in central areas of Shanghai. Businesses are represented according to the distribution of their functions in order to illustrate the utility of different locations

Full size image

The second group contains entertainment and recreation, finance and insurance, administrative management, culture and art, accommodation services, education and scientific research, as well as healthcare services. The format types are also important in Shanghai’s central areas, comprising about 20–30% of the total. Formats in this second level are very stable. The financial and insurance businesses are highest in the Lujiazui area (about 8.4%), while the remaining formats of both Lujiazui District and the People’s Square are less than 8.4%. The numbers of all formats are roughly the same in both districts.

The third group includes sports, market, traffic, and other services, as well as social welfare. Of these, social welfare is provided by the Chinese government, which is a function of the non-operation mode, while the other four are business functions. The percentage of each format is less than 1% and the total format proportions are below 10%. Formats at the third level are often diversified and have little effect on the central areas except for increasing the diversity of their functions.

Spatial statistical results

While all three of the formats in the first group—retail services, business offices, and catering services—are popular, their spatial distribution varies greatly. In general, high-value retail areas show a point distribution, while median-value areas show a linear distribution along the main axes and diversification in other places. Median- and high-value business offices are mostly located on the east side of Beijing East Road, Nanjing East–West Road, Yanan East Road, Zhangyang Road, and Jinzhou Street. High-value business offices show a general point distribution, while median-value areas follow a planar distribution on the east side of the People’s Square and a diversified trend in other places. The distribution of median- and high-value retail areas is different from that of business offices. High-value retail services are mainly located in the People’s Square and clustered on the Nanjing East–West, Fuzhou, and Fuyou roads. High-value catering service areas show a general point of distribution, while median-value areas evidence a continuous distribution on the east side of the People’s Square and a diversified trend in other places.

The second group contains entertainment and recreation, finance and insurance, administrative management, culture and art, accommodation services, education and scientific research, and healthcare services. Given the nature of these industries, administrative management, education and scientific research, and healthcare service facilities are mainly scattered throughout the central areas. The high-value administrative management areas are distributed in a star-like pattern near several roads, including the Hankou Road, Beijing West Road, Hangdu Road, Yanan Middle Road, and Dagu Road. The high-value education and scientific research areas also show a point distribution near several roads, including the Beijing East Road, Huaihail Road, Fuzhou Road, and Shangcheng Road. Meanwhile, the distribution of high-value healthcare areas is more diffuse, showing a point distribution near Nanjing West Road, Sichuan Middle Road, Kowloon Road, Shangqiu Road, and Laoshan Road.

Moreover, we found that the five formats—sports services, market services, other services, traffic services, and social welfare—mentioned above were uniformly dispersed in both areas and show a uniform scattered point distribution. While these functions rarely appear in high-value areas, they have their own distributional characteristics. For example, they often appear in a top-down, governmentally dictated layout that places sports, traffic, and other services close to residents.

We also found that social welfare and market service functions require specific spatial characteristics. Since market service functions must account for logistics and transportation, they are often grouped together with other functions of the same type. For instance, social welfare is predominantly operational, and its placement is driven by economic factors. It was not often found in high-value areas, but in both the shadow region and other central matrix surrounding the two study areas.

Discussion and conclusion

This study identified two main characteristics of format composition: retail and business offices are advantageous, and the structure of format composition is stable. It also discerned three modes of spatial distribution in the central urban areas examined: (1) high dispersion and high concentration, (2) moderate dispersion and high concentration, and (3) even distribution and moderate concentration.

Structural characteristics

The following two conclusions were drawn with regard to the spatial distribution of the different formats in the central areas of megacities. First, retail and business offices are advantageous formats and account for the majority of land use in central urban areas. Under the strong influence of market demand, a large number of retail and business offices is necessary in central areas to provide places for accommodation and consumption in order for them to maintain their vitality.

Second, the composition of different formats is stable. This is true for both the three-format levels as a whole and for each individual format, especially for those in the second group. This is because most areas have developed into relatively mature and large-scale central areas through a long period of development and self-adjustment. Over time, the formats came to exist in a relatively stable proportion.

Distribution characteristics

The spatial distributions of retail and catering services are “highly dispersed and highly concentrated.” This is demonstrated by the relatively concentrated spatial point distributions of formats in numerous and large “hot” areas. However, the spatial point distributions outside these hot areas are more balanced. For example, while retail services are clustered near the People’s Square, they are distributed far more evenly in other parts of this area along the main roads and their branches. This distribution combines the characteristics of uniform distribution with those of aggregated distribution. It is a relatively mature distribution for commercial development.

Finance and insurance, as well as traffic services, are moderately dispersed and highly concentrated. This is evidenced in the fact that these formats form relatively isolated hotspots. The number of these spots outside hot areas is rather small, however, with the exception of finance and insurance and market services, which are not in a constant state of weak dispersion and strong aggregation. Finance and insurance industries in main accumulation areas—as represented by Lujiazui District and the Bund—diverge outward along the axes. While the high aggregation phenomenon of some service spaces can improve the overall competitiveness of the space, it also reveals that its spatial distribution is not suitable for a decentralized state.

As indicated by the uniform distribution of their formats along spatial points, the spatial distributions of business offices, as well as culture and art, are “evenly distributed and moderately concentrated.” While there is more than one hot spot, they take up less space. There is no evidence of strong attachment to the road network. Of these formats, administrative management is the most widely distributed. However, this format does form a localized aggregation in the Jiujiang Road area. Moreover, while the culture and art industries are also evenly distributed, their distribution density is higher near the People’s Square than in other regions.

Limitations

This study’s data indicate that its method is valid, reliable, and universal. However, this research only focuses on Asia—where cities have particularly high density. There are significant differences between Asian cities and European/American ones due to several factors including urban design, geography, and culture. Future research must address these questions in European and American countries in order to see whether the results of this study hold true.

Conclusions

To sum up, every service organization located in central areas becomes denser to produce a better scale effect as a result of the influence of the agglomeration effect. At the same time, agglomeration results in competition between similar formats, manifested in the struggle for advantageous lots. This leads to the formation of different types of format agglomeration, as well as the replacement of low-end or unnecessary formats. As such, aggregation and diffusion occur at the same time. Moreover, the existence of the dispersed spatial distribution of formats is relative: partial aggregation still exists in overall dispersion and vice versa. Future research must examine the reasons behind this phenomenon in order to explore the formation of integrated and complex spaces in central areas.

This study provides a valuable policy basis for urban planning managers and can be summarized in three points.

First, the layout of the format location is not isolated, but influenced by the layout of the other formats around it. For example, the Lujiazui area—where business offices are most concentrated—often brings a lot of communication and people flow to other cities or regions, and many people are unable to travel back and forth on the same day. Consequently, the accommodation service function emerged. Moreover, gathering in the vicinity of business office functions can generate a larger consumer market. Therefore, the correlations and linkages between formats cannot be ignored in the actual planning of a project.

Second, there is a relatively stable numerical relationship between the proportions of various formats in the central region. This should be reflected in planning and design. For example, there is a correlation between financial and insurance functions and the number of business office functions. The financial and insurance products provided by the financial and insurance functions are usually those that serve the business functions of the commercial office and provide some guarantee for its capital turnover and currency security. At the same time, however, both supply and demand constantly promote the development of other forms. That is, the industrial chain expands as a result of accumulation of financial insurance, which provides services for other business office functions—such as the securing of the information technology service function through the supply of financial guarantees, for example. As such, both supply and demand promote the development of the other forms, thereby producing a stable numerical relationship between the number of formats.

Finally, the spatial distribution of different business forms is characterized by differentiation, and thus needs to be adjusted and controlled in planning and management. For example, the distribution of public service types—such as sports service, market services, and transportation services, as well as social welfare—in the region is often influenced more by the administrative instructions of the planning management department than the laws of the market economy. Nonetheless, administrative agencies tend to place sports, transportation, and other services closer to residents on the premise of balance. Of course, the impact of large public activities on the surrounding living environment cannot be ignored. For example, the environmental noise generated by a large number of scattered people brought to an area by sports events will disturb the surrounding residential areas. This will also have negative repercussions on both medical and health functions and education scientific research functions, as well as other areas with higher environmental quality. Therefore, the balance should be considered when making differentiation-based adjustments for different facilities in the planning of sports, market, transportation, and other services, as well as social welfare facilities.