新质生产力视角下多代理行为模拟赋能城市设计全路径分析
Multi-Agent Behavior Simulation Empowering Full-Path Urban Design Analysis from the Perspective of New Quality Productive Forces
杨春侠
同济大学建筑与城市规划学院 副教授，博士
詹 鸣（通信作者）
同济大学建筑与城市规划学院 博士研究生，zhanming@tongji.edu.cn
摘要: 依赖静态蓝图与经验判断的传统城市设计难以应对城市发展的动态变化与多元需求，亟需引入以技术驱动、数据支撑 与高效协同为特征的新质生产力。多代理行为模拟通过刻画个体与环境的动态交互，构建涵盖基础数据采集、交互机 制获取、仿真模型建构、动态模拟输出和设计优化决策5个环节的方案优化框架。该方法促进了城市设计研产学的应用 协同：在科研层面，呈现“空间—行为”动态耦合关系并拓展应用场景；在实践层面，通过“行为模拟—问题诊断—优 化预判—反馈验证”的闭环流程提高方案决策效率；在教学层面，构建“经验＋实践＋仿真＋验证”的循证式教学新 路径。其通过智能仿真辅助动态规划、综合诊断提升智慧管控、精准预判支持科学决策，推动城市设计由经验驱动向数 据驱动、由静态蓝图向动态仿真转型，为城市空间的精细化与数智化治理提供重要方法支撑。
Abstract: Traditional urban design, relying on static blueprints and experiential judgment, faces challenges in addressing the dynamic transformations and diverse demands of contemporary urban development. This necessitates the introduction of new quality productive forces, characterized by technology-driven approaches, data support, and efficient collaboration, to enable scientific decision-making and digitally intelligent governance. Multi-agent behavior simulation (MABS) captures the dynamic interactions between individuals and their environments and establishes a five-stage optimization framework, comprising data acquisition, interaction mechanism extraction, simulation modeling, dynamic simulation output, and design optimization. This method promotes the integration of research, practice, and education in urban design. At the research level, it reveals the dynamic coupling between space and behavior while expanding application scenarios; at the practical level, it enhances decisionmaking efficiency through a closed-loop process of “simulation–diagnosis–prediction–verification”; and at the educational level, it establishes an evidence-based pedagogical framework integrating “experience + practice + simulation + verification”. Overall, by enabling intelligent simulation, comprehensive diagnosis, and precise prediction, MABS facilitates the transformation of urban design from experience-driven to data-driven approaches and from static blueprints to dynamic simulation, providing essential methodological support for the refined and digitally intelligent governance of urban spaces.
关键词：多代理行为模拟；城市设计；新质生产力；数智化治理
Keyword: multi-agent behavior simulation; urban design; new quality productive forces; digital-intelligent governance
中图分类号：TU984
文献标识码: A
新华社. 中共中央 国务院关于推动城市高质量
发展的意见[N]. 人民日报，2025-08-29（001）.
Xinhua News Agency. Opinions on promoting
high-quality urban development by CPC Central
Committee and State Council[N]. People’s Daily,
2025-08-29(001).
王松，徐政，江小鹏. 数字时代的新质生产力：
“十五五”时期数据要素化与产业范式重构战略研
究——学习贯彻党的二十届四中全会精神[J/OL]. 重
庆大学学报（社会科学版），2026，32（1）：29-40.
WANG Song, XU Zheng, JIANG Xiaopeng. New
quality productivity in the digital age: strategic
research on data factorization and industrial
paradigm reconstruction during the 15th Five-Year
Plan period: studying and implementing the spirit
of the Fourth Plenary Session of the 20th Central
Committee of the Communist Party of China[J/
OL]. Journal of Chongqing University (Social
Sciences Edition), 2026, 32(1): 29-40.
SEYED M G, JAMSHID M, THEO A. A multiagent assisted approach for spatial group decision
support systems: a case study of disaster management practice[J]. International Journal of Disaster
Risk Reduction, 2019(38): 1-13.
JOSEPH K, REZA V, JAKOB P, et al. Exploring the
impact of user preferences on shared autonomous
vehicle modal split: a multi-agent simulation
approach[J]. Transportation Research Procedia,
2019(37): 115-122.
FIRDAUSIYAH N, TANIGUCHI E, QURESHI A
G. Modeling city logistics using adaptive dynamic
programming based multi-agent simulation[J].
Transportation Research Part E: Logistics and
Transportation Review, 2019(125): 74-96.
SHULGA R N, PUTILOVA I V. Multi-agent direct
current systems using renewable energy sources
and hydrogen fuel cells[J]. International Journal of
Hydrogen Energy, 2019(4): 288-290.
TENG F, ZHANG H G, LUO C M, et al. Delay
tolerant containment control for second-order multiagent systems based on communication topology
design[J]. Neurocomputing, 2020(380): 11-19.
BONABEAU E. Agent-based modeling: methods
and techniques for simulating human systems[C]//
Proceedings of National Academics and Sciences,
2002(99): 7280-7287.
DIJKSTRA J, TIMMERMANS H. Towards a
multi-agent model for visualizing simulated user
behavior to support the assessment of design
performance[J]. Automation in Construction,
2002(2): 135-145.
綦伟琦. 城市设计与自组织的契合[D]. 上海：同
济大学，2006.
QI Weiqi. The alignment of urban design and selforganization[D]. Shanghai: Tongji University, 2006.
NEIL L. The limits of urban simulation: an
interview with Manuel Delanda[J]. Architectural
Design, 2010(4): 50-55.
季民河，MICHAEL M，MIGUEL A. 基于多代
理模型的城市土地利用博弈模拟[J]. 地理研究，
2009（1）：85-96.
JI Minhe, MICHAEL M, MIGUEL A. Multiagent based modeling of land-use decision making
process in a democratic setting[J]. Geographical
Research, 2009(1): 85-96.
孔慧子，刘天明，廖家婕，等. 多智能体模型空间
分析导向下的城市公园空间结构生成[J]. 风景
园林，2025，32（3）：73-81.
KONG Huizi, LIU Tianming, LIAO Jiajie, et al.
Generation of spatial structure of urban parks
based on spatial analysis of agent-based models[J].
Landscape Architecture, 2025, 32(3): 73-81.
WU H, LIU L, YU Y, et al. An agent-based model
simulation of human mobility based on mobile
phone data: how commuting relates to congestion[J].
ISPRS International Journal of Geo-Information,
2019, 8(7): 313.
YANG C X, ZHAN M. Construction of recreation
behavior simulation model of public space in urban
waterfront—taking Huangpu River in Shanghai as
an example[C]//The International Conference on
Computational Design and Robotic Fabrication.
Singapore: Springer Nature Singapore, 2023: 201-
210.
CHIOU Y S, BAYER A Y. Microscopic modeling
of pedestrian movement in a Shida night market
street segment: using vision and destination
attractiveness[J]. Sustainability, 2021, 13(14): 8015.
杨春侠，徐思璐，耿慧志，等. 基于多代理行为模
拟的上海市北外滩滨水公共空间诊断和优化[J].
风景园林，2022，29（12）：78-84.
YANG Chunxia, XU Silu, GENG Huizhi, et
al. Diagnosis and optimization of waterfront
public space in North Bund, Shanghai based on
multi-agent behavior simulation[J]. Landscape
Architecture, 2022, 29(12): 78-84.
KLABUNDE A, WILLEKENS F. Decisionmaking in agent-based models of migration: state
of the art and challenges[J]. European Journal of
Population, 2016, 32(1): 73-97.
LI F, LI Z, CHEN H, et al. An agent-based
learning-embedded model (ABM-learning) for
urban land use planning: a case study of residential
land growth simulation in Shenzhen, China[J].
Land Use Policy, 2020, 95: 104620.
LIANG X, LU T, YISHAKE G. How to promote
residents’ use of green space: an empirically
grounded agent-based modeling approach[J].
Urban Forestry & Urban Greening, 2022, 67:
127435.
孙琳. 基于主体建模方法的住区户外空间休憩
活动模拟应用研究——以深圳市典型住区为例
[D]. 哈尔滨：哈尔滨工业大学，2020.
SUN Lin. Research on the application of residential
outdoor space recreation activity simulation based
on agent-based modeling method[D]. Harbin:
Harbin Institute of Technology, 2020.
马婕，成玉宁. 基于集群智能行为模拟与空间句
法分析的城市公园优化设计研究[J]. 中国园林，
2021，37（4）：69-74.
MA Jie, CHENG Yuning. Research on optimization
of urban park design based on swarm intelligent
simulation and spatial syntax analysis[J]. Chinese
Landscape Architecture, 2021, 37(4): 69-74.
姚梓莹. 基于多代理行为模拟的城市滨水区公
共空间品质综合测度指标体系研究——以上海
黄浦江为例[D]. 上海：同济大学，2022.
YAO Ziying. Multi-agent simulation-based urban
waterfront public space quality comprehensive
measurement indexes—taking the Huangpu
waterfront area of Shanghai as an example[D].
Shanghai: Tongji University, 2022.
CHEN Y, YANG J. Historic neighborhood design
based on facility heatmap and pedestrian simulation:
case study in China[J]. Journal of Urban Planning
and Development, 2020(2): 1-12.
马婕. 基于数字技术的城市公园行为预测与空
间设计研究[D]. 南京：东南大学，2021.
MA Jie. Research on behavior prediction and
spatial design of urban parks based on digital
technology[D]. Nanjing: Southeast University,
2021.
连海涛. 基于视觉注意力理论的智能体购物行
为模型[J]. 装饰，2018（11）：128-129.
LIAN Haitao. Intelligent agent shopping behaviour
model based on visual attention theory[J]. Art &
Design, 2018(11): 128-129.
顾卓行，杨春侠，朱玮. 基于智能体行为模拟的
城市公共空间使用诊断研究——以上海市北外
滩滨水公共空间为例[J]. 新建筑，2025（3）：
154-158.
GU Zhuoxing, YANG Chunxia, ZHU Wei.
Urban public space use diagnosis based on agent
behavior simulation: a case study of the North
Bund waterfront public space in Shanghai[J]. New
Architecture, 2025(3): 154-158.
陶叶康. 多智能体技术在住区居民户外行为模
拟和健康空间设计中的应用研究[D]. 南京：东
南大学，2023.
TAO Yekang. Research on the application of multiagent technology in residential outdoor behavior
simulation and healthy space design[D]. Nanjing:
Southeast University, 2023.
王硕. 基于GIS的景区游人游憩模型及可视化系
统[D]. 北京：北京建筑大学，2018.
WANG Shuo. Tourist recreation model and visualization system of scenic spots based on GIS[D].
Beijing: Beijing University of Civil Engineering and
Architecture, 2018.
杨春侠，闵若遥. 行为模拟下滨水健身空间品质
测度与优化预判[J]. 风景园林，2025，32（4）：
106-115.
YANG Chunxia, MIN Ruoyao. Quality measurement
and optimization prediction of urban waterfront
fitness space based on behavior simulation[J].
Landscape Architecture, 2025, 32(4): 106-115.
YANG C X, LIU M X, FAN Z X. Modelling and
visualization behavior simulation in public open
spaces based on MLP and SFM[J]. Simulation
Modelling Practice and Theory, 2025, 140: 103076.
LIU M X, YANG C X, FAN Z X. Prediction,
simulation and evaluation of stay behavior in
public space integrated with microclimate[C]//
Proceedings of the 29th International Conference of
the Association for Computer-Aided Architectural
Design Research in Asia (CAADRIA). Hong Kong:
CAADRIA, 2024, 2: 129-138.
YANG C X, LYU C Z, YAO Z Z, et al. Study on
the differences of day and night behavior in urban
waterfront public space based on multi-agent
behavior simulation[C]//Proceedings of the 26th
International Conference of the Association for
Computer-Aided Architectural Design Research in
Asia (CAADRIA). Hong Kong: CAADRIA, 2021,
2: 559-568.
鞠梦恬. 基于多代理行为仿真的城市设计教学
实践[D]. 上海：同济大学，2024.
JU Mengtian. Teaching practice of urban design
based on multi-agent behavior simulation[D].
Shanghai: Tongji University, 2024.
YANG C X, ZHAN M, YAO Z Y. Multi-agent
simulation-based urban waterfront public space
quality comprehensive measurement indexes[C]//
The International Conference on Computational
Design and Robotic Fabrication. Singapore:
Springer Nature Singapore, 2024: 536-548.
邬伦，宋刚，吴强华，等. 从数字城管到智慧城
管：平台实现与关键技术[J]. 城市发展研究，
2017（6）：99-107.
WU Lun, SONG Gang, WU Qianghua, et al. From
digital city management to smart city management:
the basic platform and its key technologies[J].
Urban Development Studies, 2017(6): 99-107.
张兵，门晓莹，叶斌，等. “落实中央城市工作会
议：新规划实践探索”学术笔谈[J]. 城市规划学
刊，2025（5）：1-13.
ZHANG Bing, MEN Xiaoying, YE Bin, et al.
Symposium on “implementing the Central Urban
Work Conference: exploration of new planning
practices”[J]. Urban Planning Forum, 2025(5): 1-13.
