零售不動產透過聚集經濟效益將多樣性的業種廠商聚集在一起,為使利潤極大化,零售不動產必須使商場空間達到最有效率的使用,本研究嘗試從業種多樣性角度,分析零售不動產的空間配置。研究標的以臺灣的新光三越A4館、新光三越A8館、新光三越A9館、新光三越台南西門店、美麗華與101,美國的Four Season與Puentehills Mall,新加坡的SunTech City等九個購物商場,共五十四個樓層平面為研究樣本,利用GIS地理資訊系統的空間資訊處理功能,分析並計算櫃位業種部門化程度,將商場空間、業種多樣性與部門化程度三者之間的關係,利用線性迴歸模型作分析。實證結果發現:(1)當總樓層數越高,每樓層的多樣性越低,部門化程度越高;較高的樓層有低樓層沒有的業種且以業種部門化的方式作為零售吸引力,使距離較遠的高樓層具有獨特性的吸引力,促使人潮有向上遊逛的動力。(2)當毛營運面積越大,則多樣性越高,業種部門化越低;為使廣大的樓層平面面積裡分散零售吸引力,使人潮能夠完成完整的迴流。(3)當空間複雜度越複雜時,越以業種部門化的方式以避免遊逛空間的複雜使消費者耗費更多商品搜尋成本。接著利用二元Logistic迴歸分析出當樓層數越高,垂直距離開始稀釋了零售吸引力時,建議當零售商場樓層數超過4樓時,業種狀態以「部門化」的方式去配置。最後利用鑑別分析得到國內外的商場在業種「部門化程度」上的確有明顯的差異;國內的商場大多以單層面積較小、樓層數較高且以部門化的方式作為業種的配置型態;而國外的商場大多以單層面積較為廣大、樓層數少且以非部門化的方式配置其業種。
Retail tenants in a large-scale multi-unit retail property are pursuing retail agglomeration economies. Compatible tenants generate spillover effects through locational cluster. Hence, the efficiency of the allocation of the precious resource, ie, floorspace, is one of the significant factors that affect the performance of the property. The objectives for these geometrical patterns are to guide the shoppers’ circulation to pass all/most of the tenants through the simplest way. Spatially departmentalized retail categories become necessary when floorspace with more complex layout. This paper compares nine large-scale retail properties on a GIS-based cases study, so as to analysis the patterns of the agglomeration for various retail categories. Analylize the relationship within space of retail property、retail variety and degree of departmentalization with simple linear regression model、logistic regression model and discriminate analysis. In this research suggested that floor plan should transform the non-purposive shopping route into a guided purposive shopping pattern by departmentalized allocation when retail property were extended vertically and high complexity. The results also reveal that centres with lower total floor levels and larger space tends to dispersively locating the same retail categories. However, for retail centres with higher levels, higher complexity and smaller space are more likely to spatially departmentalized their retail categories. Then logistic regression analysis result suggested that when floor number of retail properties surpasses 4, the same type of tenants tends to be departmentalized . Finally, discriminate analysis obtains that the degree of departmentalization has the obvious difference between the domestic and foreign retail properties.