| 摘要: |
| 人类正大规模生成海量人居环境数
据，传统规划分析方法已不足以应对，亟需机
器学习等人工智能（Artificial Intelligence, AI）
工具。这既对规划师技能提出新挑战，也凸
显高校规划专业AI教育的迫切性。我国高校
城乡规划专业AI教育虽蓄势待发，却存在着
内容零散、规模较小、体系缺失等问题，同
时还受限于学制的缩短调整。美国高校规划
专业AI教育近年来迅速铺开，通过选取规划
教育排名前列的32 所高校，对其规划专业的
目标定位、专业设置、课程体系和课程简
介，以及AI相关课程的教学大纲、教职人员
和项目团队等文献开展全面系统的内容分
析。结果表明：美国高校规划专业AI教育具
有明确的技术工具属性定位，以数据科学与
城市分析教学为依托，以编程语言、数据科
学与AI、AI 原理与算法、AI 规划应用为内
容主线，形成了多模式、多层次的培养方案
和以规划AI专业教师为基础的多渠道师资融
合态势。这些成功经验对我国具有重要启示
意义和参考价值。 |
| 关键词: 人工智能 人居环境数据 城乡规
划 规划专业AI教育 美国高等院校 |
| DOI：10.13791/j.cnki.hsfwest.20240620001 |
| 分类号: |
| 基金项目:西南民族大学科研启动金资助项目（RQD2023024）；四川省高等教育人才培养和教学改革重大项目（JG2023-37） |
|
| The current situation and enlightenment of AI education in planning programs of highereducation in the United States |
|
LIU Xin,ZHAO Bing,WEN Xiaofei,NIE Kangcai
|
| Abstract: |
| Large quantities of data about human settlements are being generated on an unprecedented
scale, while traditional data analysis methods in urban and rural planning are insufficient to
understand and utilize these new data. Thus, advanced artificial intelligence (AI) analysis tools such
as machine learning are required for big data. This presents new challenges for planners to improve
their skills and creates an urgent need for universities to incorporate AI education into planning
programs. Such education in China is poised to expand, yet it remains weak in content, small in scale,
unsystematic in framework, and constrained by the reduced duration of planning programs. Hence,
fundamental questions that urgently need to be addressed include: What is the role of AI education in
planning programs? What models can be adopted to deliver such education? How should AI education
be differentiated across degree levels? How can curriculum content be effectively designed and
faculty resources strategically allocated? As U.S. higher education institutions have rapidly launched
AI-related planning programs in recent years, this research samples thirty-two top-ranked universities
in graduate planning programs in the United States to examine their experiences and answer the above
questions. It conducts a systematic and in-depth content analysis of 457 documents, including
strategic plans, handbooks, curricula, course profiles, syllabi related to AI in planning, instructor
information, and lab descriptions. NVivo 14 was used to code and integrate these data. Through
continuous iterative comparison and refinement, five dimensions and their corresponding sub-nodes
were extracted, summarizing the content analysis results. Firstly, the positioning dimension shows that
AI education in U.S. planning programs began in 2017 and expanded comprehensively from 2021,
with increasingly diverse courses and degree programs. As of 2024, 65.6% of the sampled universities
have incorporated AI into their curricula, with higher adoption rates among top-tier institutions.
Although only 15.6% explicitly list AI as a strategic goal, over half emphasize the importance of data
science and urban analytics, positioning AI as a core technology for data-driven decision-making and
modern planning practice. Secondly, the model dimension reveals three modes of AI education in
planning programs: establishing joint/dual degree or short-term degree programs to systematically
incorporate AI courses; offering data science specializations or certificate programs within planning
degrees to focus on AI instruction; and providing independent elective courses and AI labs to help
students master fundamental theories and applications. Thirdly, the hierarchy dimension shows that AI
education in planning is highly concentrated at the master’s level, with only 8.1% at the
undergraduate level and no independent AI specializations offered. It presents a progressive structure:
undergraduate for basic understanding, master’s for proficient application, and doctoral for advanced
research and development. Fourthly, the content dimension encompasses four aspects: programming
languages, data science, AI algorithms, and AI applications. Python and R are the main programming
languages and are typically integrated into comprehensive courses. Introductory courses on data
science and AI are the most widespread, often combining theory, application, and programming. AI
algorithm courses focus on machine learning and require a stronger mathematical foundation. AIplanning application courses are divided into basic introductions and advanced topics, covering diverse fields such as transportation, environment, and smart
cities. Fifthly, the faculty dimension shows three approaches to faculty allocation: over half of the universities rely solely on in-house faculty for AI instruction,
mainly offering introductory courses; about one-third depend primarily on external or cross-departmental faculty to support more specialized AI courses; and a
small number achieve a balance between internal and external faculty. AI faculty resources remain generally limited, yet nearly half of the institutions have
actively recruited specialized AI teachers in recent years. Additionally, this research examines three representative types of courses specifically. 1) Introductory
data science and AI courses, exemplified by Tufts University, focusing on urban data analysis and the application of Python tools in AI; 2) AI algorithms
courses, as represented by Columbia University, covering various machine learning algorithms and their practical applications in planning; 3) AI in planning
application courses, such as those at the University of Florida, concentrating on the practical use of AI technologies in the built environment. All three types of
courses emphasize the integration of theory and practice, highlighting the cutting-edge and practical nature of AI in planning. In light of the above findings, this
research summarizes three key insights. 1) Clarifying the dual roles of planners in the AI era—both as users of AI tools and participants in their development; 2)
emphasizing the balance between technical and interpersonal skills while building a data-driven skill framework; 3) establishing a multi-level, integrated AI
education system that combines degree programs, course modules, and practical support. The successful experience of the United States can serve as a valuable
reference for China to systematically advance AI education in planning programs at the university level. |
| Key words: artificial intelligence data in human settlements urban and rural planning AI education in planning programs higher education in the United States |
