The study of multi-level modeling of accessibility in urban space

大多数区位分析都是建立在“线性”距离的基础上，无法对复杂的城市空间进行可达性建模。本文以GIS为工具，采用尺度相关的多层次建模方法，提出了可达性的概念，并探讨了这一概念在真实的城市空间中的适用性。探索的结果意味着，重要的是要概念化的可达性依赖于每个规模和定义在本地空间，而不是线性成比例的距离。其次，他从类比和修辞的角度考察了艾伦·威尔逊的“熵最大化模型”。他认为，模型是有问题的线性关系之间的宏观，中观和微观层面的聚合，一方面，另一方面，封闭系统建模。第三，通过比较MDS和Q分析两种方法，提出了一个更加动态的城市系统时空收敛演化模型。他的结论是，Q-分析可以更相关的治疗一个定性的系统变化，如衔接，依赖，conurbation和隔离在国家城市空间，相比MDS的定量转换方法。

Most of locational analysis are not relevant to model the accessibility in some complicated urban space because they have been based on the concept of "linear" distance. The author aims to make another concept of accessibility concerning to the connectivity among diverse transportation and communication systems, using the multi-level modeling approach of scale-dependent viewpoint with GIS, and furthermore, to look into the availability of the concept in the real urban space.Firstly, the author explored how the ideas of fractal, chaos, and self-organization can be applied to understand the becoming urban systems. The results of the exploration implies that it is important to conceptualize the accessibility dependent on each scale and defined in local space, rather than that is linearly proportionate to distance. Secondly, he examined the "entropy-maximizing models" originated by Alan Wilson from a viewpoint of analogy and rhetoric. He suggests that the models are problematic for the linear relationship among macro-, meso-, and micro-level aggregation on one hand, and for the closed system modeling on the other hand. Thirdly, the author compared two similar methodology of MDS and Q-analysis to make a more dynamic and evolutionary model of urban systems concerning to time-space convergence. He concludes that Q-analysis can be more relevant to treat a qualitative system change such as articulation, dependence, conurbation, and isolation in national urban space, compared to the MDS's quantitative transformation method.