地球信息科学学报 ›› 2020, Vol. 22 ›› Issue (4): 792-804.doi: 10.12082/dqxxkx.2020.190628

• • 上一篇    下一篇

城市交通与土地利用一体化模型的核心算法进展及技术创新

赵鹏军*(), 万婕   

  1. 北京大学城市与环境学院城市规划与交通研究中心,北京 100871
  • 收稿日期:2019-10-25 修回日期:2019-11-21 出版日期:2020-04-25 发布日期:2020-06-10
  • 通讯作者: 赵鹏军 E-mail:pengjun.zhao@pku.edu.cn
  • 作者简介:赵鹏军(1975— ),男,陕西延安人,教授,博士生导师,主要从事交通与空间规划研究。
  • 基金资助:
    国家自然科学基金项目(41925003);英国研究理事会全球挑战基金项目(R48843)

The Development and Innovation of the Core Algorithm of the Integrated Model of Urban Transport and Land Use

ZHAO Pengjun*(), WAN Jie   

  1. The Centre for Urban Planning and Transport Studies, College of Urban and Environmental Sciences, Peking University, Beijing 100871, China
  • Received:2019-10-25 Revised:2019-11-21 Online:2020-04-25 Published:2020-06-10
  • Contact: ZHAO Pengjun E-mail:pengjun.zhao@pku.edu.cn
  • Supported by:
    National Natural Science Foundation of China(41925003);Global Challenges Research Fund of the Research Councils United Kingdom(R48843)

摘要:

交通与土地利用的协调发展是缓解城市交通拥堵、提升交通水平的关键。城市交通与土地利用一体化模型则是分析和模拟城市交通与土地利用相互作用过程的重要科学支撑。多年来不同国家学者已研发出一些操作模型,并用于城市空间政策的制定研究,但这些模型的核心算法仍需进一步升级完善。本文首先梳理了当前已有主流模型的理论特征,解析了这些模型的核心算法发展及其优缺点;然后针对关键算法的不足,提出新的综合均衡模型,就区位可达性、用地混合度与强度、出行成本3个关键变量的核心算法展开讨论;最后,进一步形成模型各子模块的算法创新,包括考虑增量离散选择过程的居住与就业区位决策模块、细分房地产类型并将土地开发结果动态反馈至城市用地演变的房地产开发模块、采用改进的阻抗函数、小汽车拥有情况和动态出行成本的交通综合模型。本文对于定量化分析模拟城市复杂系统、研发我国城市交通与土地利用一体化模型具有一定的参考价值。

关键词: 城市交通, 土地利用, 一体化模型, 算法特征, 关键变量创新, 区位决策, 房地产开发, 模型模拟

Abstract:

The coordinated development of urban transport and land use is of great practical significance to optimize urban spatial structure, reduce traffic congestion, and improve transport level. The integration of land use and transport systems is important for smart growth and sustainable development of cities. The integrated model of urban transport and land use is a key scientific support to analyze and simulate the interaction between urban transport and land use. Over the years, scholars in different countries have developed operational models that can be used for urban spatial policy formulation. However, the core algorithm of these models still need to be further improved. In this paper, the theoretical characteristics of existing mainstream models were analyzed. Based on the theoretical basis and the core algorithms, the advantages and disadvantages of six models were discussed, including spatial interaction model, urban economics and mathematical planning model, spatial input-output model, discrete choice model, micro-simulation model, and cellular automata model. A new comprehensive equilibrium model was proposed to overcome the shortcomings of the existing models. On the one hand, endogenous variable processing and dynamic feedback of the key algorithms need to be improved. On the other hand, the existing models do not fully consider the impacts of exogenous variables such as technological innovation, urban management, and urban planning policies. Therefore, new thinking was put forward for the core algorithm incorporating three key variables: Accessibility, land use characteristics, and travel cost. The equilibrium model adopts land use intensity and diversity to represent land use characteristics, and uses the improved algorithm of accessibility that takes into account the repulsive force caused by housing price and the opportunity scale of land use characteristics. In the generalized travel cost, attributes at the individual level and characteristics at the urban environment level are considered comprehensively. New algorithms were also proposed for the three modules of the integrated model, which include the residence and employment location decision module that considers incremental discrete selection process, the land supply and real estate development module that subdivides real estate types and dynamically feeds back land development results to urban land use evolution, and the comprehensive transport model that adopts improved impedance function, dynamic travel cost, and car ownership. It has important theoretical value for quantitative simulation of urban spatial evolution, prediction of trip distribution, and evaluation of urban management and planning policies. At the same time, it is of practical significance to develop the integrated model of urban transport and land use in China, so as to optimize the distribution of population and trip demand, and ease urban traffic congestion.

Key words: urban transport, land use, integrated model, algorithm features, key variable innovation, location choice, real estate development, model simulation