高齡者住宅社區之區位規劃模式

A Location-allocation Model for Senior Citizens Communities Planning

本研究目的在建立適合高齡者住宅社區之區位規劃模式，並應用於台北市萬華區，以驗證模式之適當性與實用性，分析結果可具體提供台北市政府都市規劃單位參考。本研究使用土地使用適宜性分析與數學規劃方法將模式設計為多目標混合整數規劃問題型態；考量不同關係者對規劃結果之需求，故設計外部環境適宜性最大化、配置規模最大化以及開發成本最小化三個目標；而考量實質環境所能提供規劃高齡者住宅社區所需之資源，故設計基地開發規模、住宅社區容量、居住需求以及值域宣告四種限制條件；最後則根據模式本身特性選擇以ε-限制法求解。 This study aimed at developing a location-allocation model for senior citizens communities planning. The developed model was applied to the case study of Wanhua District, Taipei City, to verify the applicability and reasonableness of model. The planning results of case study offer concrete recommendations to Taipei City Government. This study employed land-use suitability analysis and mathematical programming approach to design the model as a multi-objective mixed integer programming problem. The model considers three objectives for different stakeholders: maximizing environmental suitability, maximizing developed dwelling units and minimizing development costs, with the following constraints: upper and lower bounds of development scale for each site, dwelling unit demands and value ranges of decision variables. According to the model characteristics, this study applied ε-constraint method to solve the developed model.

This study aimed at developing a location-allocation model for senior citizens communities planning. The developed model was applied to the case study of Wanhua District, Taipei City, to verify the applicability and reasonableness of model. The planning results of case study offer concrete recommendations to Taipei City Government. This study employed land-use suitability analysis and mathematical programming approach to design the model as a multi-objective mixed integer programming problem. The model considers three objectives for different stakeholders: maximizing environmental suitability, maximizing developed dwelling units and minimizing development costs, with the following constraints: upper and lower bounds of development scale for each site, dwelling unit demands and value ranges of decision variables. According to the model characteristics, this study applied ε-constraint method to solve the developed model.