都市計畫草圖替選方案之分析模式

A MODEL TO GENERATE ALTERNATIVE SKETCH MAPS FOR URBAN PLANNING

都市計畫配置草稿替選方案的產生過程，一直被視為機關內部行政作業，致產生兩個問題：一是方案數目稀少，很難、涵蓋到最佳的配置；二是配置內容為規劃者的主觀想法，忽略居民的客觀感受。為解決這些問題，有必要構建一個能很快地產生多量、客觀且完整替選方案的分析工具，以減輕規劃人員負擔並提高規劃品質。因此本文構建一個能同時配置運輸路網與土地使用的多目標0-1非線性規劃模式，並利用基因演算法求解假想簡例的近似非劣解集，以驗證模式的可操作性並進行一些測試。

The generation of alternative sketch maps for urban planning has long been viewed as an administrative process. There were two basic issues: few alternatives and subjective results. The ideal concept is that the more sketch maps we generate, the more chances we can get the best layout. In other words, if sketch maps could be generated rapidly, we may enhance planning efficiency and reduce planner's burden. We build a multiobjective 0-1 nonlinear programming model for land use-network design, and utilize a genetic algorithm to solve such complex problem. The operational feasibility and some tests of this model is examined and discussed by a numerical example.