執行都市計畫，規劃者需蒐集分析資料，訂定都市規劃與發展準則，然而都市發展的耗時長、影響因素繁雜，由於發展過程中充滿諸多不可預測的因素，造成規劃內容無法落實於地方的現象。若能有效的運用事前評估，將有助於減少都市計畫內容與實際發展不符合的問題。因此，本研究擬藉由空間型構(Space Syntax)參數探討商業活動分佈的預測，以檢討包含商店數量與商業使用強度等與空間型構參數的關聯性。期望能推衍出一套可模擬預測商業分佈趨勢之分析模型，使其作為未來都市計畫規劃過程及劃定商業分佈區域之參考依據。本研究對象為台南市中西區，利用實地調查的方式，記錄研究區內之商業分佈與使用程度，並針對該地區的路網，進行空間型構參數便捷值與選擇值的分析，以探索不同參數與商業使用強度、商店數量間的關聯性。最後使用統計方法中，皮爾森 (Pearson) 相關係數及線性迴歸之統計分析方法，解析最可能用以預測商業活動分佈的參數，藉以檢測空間型構理論用於城市商業活動模擬推衍之可能性。研究發現，商業會沿著主要幹道發展，而商業使用的強度，會隨著相對於主要幹道之路線轉折的次數增加而逐漸降低，其中關於商業使用強度及商店數量的分佈傾向，運用空間型構參數選擇值R3(Choice R3) 最能詮釋商業使用現況的分佈，約有68%的解釋能力。後續更以台南市北區及台中市西屯區為實證研究的案例，在初步檢視Choice R3的預測能力後得知，對於商業聚集的程度與商業使用強度皆約有60%的解釋能力。且本研究對比空間型構與都市計畫兩者對於商業分佈的預測能力，結果顯示空間型構分析方法的預測準確性高於都市計畫的分析方法。故本研究認為，為改善既有商業區劃設僅能仰賴規劃者粗略判斷的問題，規劃者在進行分區劃設時，若能事先運用空間型構參數Choice R3作為規劃時的分析及預測工具，使商業使用分區劃設在未來區域內最多選擇穿越性人潮與車流的地方，將有助於商業活動的形成，使都市規劃內容能更有效地落實。 During the implementation of urban planning, planners need to collect and analyze data to formulate rules for urban planning and development. However, conducting urban development is very time-consuming, and the influencing factors are complicated. This makes the planning content extremely hard to be implemented in the local area. An effective pre-evaluation process can help to improve urban planning and also reduce the possibility of urban planning content not matching with actual development. Therefore, this study is to predict the distribution of business activities, by analyzing the correlation between the number of stores and the intensity of commercial use. Data of units of stores and the intensity of commercial use are gathered by conducting field surveys in the research object, which is the current situation of commercial activities in the mid-west district of Tainan city. All the streets in the research area are analyzed by the choice and integration of space syntax to find the correlation between commercial use intensity and the number of stores. Lastly, Pearson product-moment correlation coefficient and linear regression statistical analysis methods are used to find the parameters that can best predict the distribution of business activities when applying Space Syntax theory.This study found that commercial activities mainly develop along major roads, and the intensity of commercial activities will gradually decrease as the number of routes turns increases. The results of this research show that choice R3 can best predict the distribution of business activities, commercial use intensity and the number of stores using the space syntax model. In the case of the north district of Tainan city and Xitun district of Taichung city, around 60% of the results using choice R3 can be explained. Therefore, this study shows that to effectively implement urban planning content, the planner can find the best land use by using Space Syntax choice R3 to predict the distribution of business activity in a certain area.
