路网形态对城市酒店业空间布局的影响研究:广州案例

Guangzhou's Spatial Distribution under the Effect of the Urban Road Network on the Hotel Industry

酒店行业的空间分布模式、分布的影响机制、区位选择等内容正日益成为旅游地理学等学科的研究热点。文章运用空间句法模型,结合GIS技术,选取2014年广州市中心区酒店业POI数据,从"区域酒店业空间分布格局分析"、"城市路网形态分析"、"城市路网形态对酒店业分布规律的影响"3个角度入手,探究了酒店业空间布局规律。研究发现:(1)广州市中心区酒店业呈现多核心聚集式分布,分布密度由城市中心地向外围呈圈层式递减,分布方向总体呈"东-西"向模式。(2)各档次酒店的聚集程度和分布核心具有差异性,酒店档次越高,聚集程度越弱,其分布中心越靠近东部城市新区。(3)不同尺度下路网形态在交通穿行性方面体现出一定的布局相似性与连续性,而在城市整合力方面却具有较大差异性。(4)相同尺度下路网形态的整合中心相比交通穿行中心表现出内聚性的特征,推动了酒店产业集聚分布特征的形成。(5)路网形态对酒店业布局产生较大影响,酒店档次越高,所受路网形态的影响越小;酒店业提供到达性服务大于穿行性服务,随着酒店档次的升高,所提供到达性服务相对于穿行性服务的比重增加;局部性的路网形态对中低档酒店的影响更大,而全局性的路网形态对高档酒店影响更大。

The urban road network is a significant concept within space syntax, a method that examines the relationship between buildings and areas in an urban space. The road network forms the core of space syntax, and space syntax is an important factor influencing the economic and social activities in cities, including the hotel industry. There is an increased focus on the hotel industry＇ s spatial distribution mode （the effect of distribution and the choice of location） in tourism geography among other subjects. The development of the hotel industry in Canton （the capital of Guangdong province, South China） started during China＇ s reform and opening in the 1980s and it experienced significant growth in the subsequent 30 years. To determine the distribution pattern of Canton＇ s hotel industry, this study uses big data including information on the position, rank and name of all of the hotels （from low class to high） in the city. Big data is a broad term for data sets that are so large or complex; therefore, traditional data processing applications are inadequate. The challenges of big data include analysis, capture, curation, search, sharing, storage, transfer, visualization, and information privacy. The term often refers simply to the use of predictive analytics or other advanced methods to extract value from data, and seldom to a particular data set size. Canton＇ s hotel industry is taken as an example and is examined based on space syntax in combination with geographic information system （GIS） kernel density estimation, shared nearest neighbor clustering, standard deviation ellipse and other space analytic methods. The hotel industry data in central Canton are selected from Baidu （2014） and are analyzed from three aspects： the distribution pattern of the regional hotel industry, the urban road network pattern and the effect of the urban road network on the distributional regulations of the hotel industry. The article reports on the distributional regulations of Canton＇ s hotel industry and provides advice toward future urban tourism planning. It provides a new perspective on exploring the distribution pattern of hotels. The research findings are as follows. （1） The hotel industry in central Canton presents a multi-core cluster distribution. The distributional density decreases from the city center to the outskirts in a ring pattern and the distribution tendency follows an ＇ east-west＇ pattern that reflects the urban expansion and economic development in the past 30 years; （2） There is a discrepancy between the aggregation degree and the distributional core. In other words, the more prestigious the hotel, the weaker the aggregation degree and the closer its location to the east-side; （3） The road network presents distributional similarity and continuity from the perspective of NA choice; however, it shows a comparable difference in urban integration; （4） The integration center of the urban network shows more cohesion than NA choice does, and this promotes the level of aggregation development in the hotel industry; （5） The urban road network pattern has a substantial effect on the hotel industry distribution. The more prestigious the hotel, the less the road network pattern influence. Considering the road network pattern influence, the arrival services provided by hotels are more than NA choice; that means, the arrival service amounts enjoy high NA choice levels when the hotel rank is high. A partial road network pattern has a significant effect on the mid-to-low class hotels, while a global road network pattern has a significant effect on the higher class hotels.