Planning in the context of urban sustainability is very challenging as it requires simultaneous integration of a large number of conflicting objectives pertaining to economic, environmental and social dimensions. It also involves multiple stakeholders with opposing views, and a high level of uncertainty. In addition, the process should be transparent, participatory and understandable by all participants. Existing literature has called for exploring new analytical methods to support decision-making for urban sustainability. This research introduces a structured decision support framework underpinned by Multi-objective Decision-Making (MODM) and Multi-attribute Decision Making (MADM). The framework provides systematic guidance to decision-making starting from problem structuring to generating a wide range of alternatives until the selection of the final solution. The developed framework is tested in two different decision-making situations pertaining to real urban problems. In the first case study, the framework is examined in the situation where the decision-maker is available to interact with the planner at the design stage. The framework is used to find a sustainable distribution of healthcare centers taking into consideration conflicting objectives including cost, efficiency of service and accessibility. The optimal solution is reached through an interactive process with stakeholders. In the second case study, the framework is examined in the situation where the decision-maker is unavailable for interaction at the design stage. The framework is used to develop planning scenarios for transit-oriented development (TOD) around a transit station. The optimal intensification of land use and land use mix is achieved taking into consideration the conflicting objectives of various stakeholders. Large number of non-dominated alternative solutions has been generated. An interactive tool has been developed by which the stakeholders can identify the alternative that best reflects their preference. The quality of the outputs for both case studies has shown that the developed planning methodology outperforms conventional approaches. The developed framework has proved to be a flexible and practical approach to assist decision-making in the context of urban sustainability / Dissertation / Doctor of Philosophy (PhD) / Achieving sustainable urban development is a very challenging task for planners. It requires simultaneous integration of a large number of conflicting objectives pertaining to economic, environmental and social dimensions. It also involves multiple stakeholders with opposing views, and a high level of uncertainty. In addition, the decision-making process should be transparent, participatory and understandable by all participants. This research introduces a decision-support framework, underpinned by mathematical modeling techniques, to assist in achieving sustainable solutions for urban development problems. The framework is examined through two case studies. First, it is used to find a sustainable distribution of healthcare centers taking into consideration conflicting objectives including cost, efficiency of service and accessibility. Second, the framework is used to develop planning scenarios for transit-oriented development (TOD) around a transit station. The developed framework has proved to be a flexible and practical approach to assist decision-making in the context of urban sustainability

本論文之要在探討多重準則決策（Multiple Criterion Decision Making）理論模式 及其相關應用。 ＭＣＤＭ問題可概括分為兩種類型：一是多屬性決策（Multiple Attribute Decisi- on Making ），一是多目標決策（Multiple Objective Decision Making）。 ＭＡＤＭ一般用於選項間之選擇；而ＭＯＤＭ則在設計構想中的最佳選項（Alterna- tive），一般ＭＣＤＭ法則須由下列三者組成：ヾ量化的目標函數；ゝ一組限制式； ゞ已知資訊的量化過程。線性規劃問題（Ｌ．Ｐ．）即是ＭＯＤＭ的一個例子。 ＭＡＤＭ問題通常包含了個數不多且事先決定的選項，此問題旨在經由屬性（Attri- bute）（或稱準則）之間的比較，對一群特定的選項做一個選擇。此類問題在各學科 都發展出特有的法則；例如：決策理論中的最大最小法則、事前機率、及效用理論； 經濟學中的Pareto最適性、社會福利函數；統計學中的多變量分析、因子分析；另外 在心理學方面也有應用。 本文偏重ＭＡＤＭ之剖析，對社會選擇論也有研討，主要研究動機在於探討各學者所 建立之理論模式，如何經由偏好現象尋取一個團體的共識；利用公設化系統誘導出審 慎的（Prudent ）結果，使得結果理性化（Rational）；我們的目的就在於尋求一個 能得到理性結果的法則。 本文結構計分四個部分： 第一部分：效用理論，偏好關係，投票行為。第二部分：Electre 法則，Hierarchy 法則，LINMAP法則以及其它相關法則概介。第三部分：多數決法則（Majority Meth- od），Borda Count ，Condocet法則排名公設、Kohler法則及Arrow-Raynaud's 法則 。第四部分：教育投資方面應用的討論及結論。