【报告摘要】Examining the recreational value of urban parks is essential for improving urban park systems and encouraging park activities, which requires a comprehensive consideration of park characteristics, park accessibility, and user preference. Though building discrete choice models (DCMs) has been deemed as a promising way to quantify park recreational values, most of existing studies are based on state preference (SP) experiments instead of revealed preference (RP) datasets. It remains unclear whether park valuation varies among different contexts (e.g., seasons, weekdays/weekends) in real cases. This study explored the feasibility of building DCMs with mobile phone-derived park activities. Using a one-year Tokyo mobile phone dataset provided by ZENRIN DataCom Co., LTD, we built mixed logit models (MXL) to examine residents’ willingness-to-visit (WTV) for urban parks and explored the variation of perceived park recreational values on weekdays/weekends in four seasons. Based on the preliminary experiment, we found that 35.2% of Tokyo residents have visited at least two different parks, and a majority of them didn’t choose the park closest to their home. Moreover, residents tended to travel a longer distance for parks with larger size and higher ratings, and their WTT was obviously larger in spring and on weekends than in other periods. The contribution of our study is two-fold. On the one side, our study proved the feasibility of using mobile phone data to build RP-based models, which is complementary to existing SP park studies. On the other side, our findings provide new perspectives for park planners and policymakers, by quantifying the trade-offs that residents made when choosing different park options. We conclude that park characteristics, park accessibility, and preference heterogeneities should be simultaneously considered to formulate planning and design strategies beyond generic guidelines. Our findings are complementary to existing SP-based park studies, and can provide new perspectives for park planners and policymakers by quantifying park recreational values in a real case.

【报告人简介】关成贺，哈佛大学城市规划设计博士，上海城市设计与城市科学重点实验室联合创始主任，上海纽约大学城市科学与政策研究员，科技部专家项目青年人才，上海市东方学者特聘教授，纽约大学瓦格纳公共政策学院博士生导师。关教授团队创建了上海纽约大学第一个省部级重点实验室，实现了建校10年以来科研方向的重大突破；建立了重点实验室与牛津大学联合成立了博士后培养站点；与上海大数据中心、上海交通大学附属新华医院、人民网人民数据等机构签订了产学研发展协议。

关教授团队的研究领域是计算城市科学，这一跨学科领域，其核心是利用人工智能、大数据分析等科学定量研究方法，结合城市设计、社会科学“厚”数据等定性研究方法，来深入理解、探究和解决城市和大都市区内复杂的时空发展问题。该领域内三个高度关联且与未来城市发展需求密切相关的创新研究方向包括：（1）数字城市转型框架下城市时空分析与设计；（2）全球气候变化背景下低碳城市发展与模拟；（3）城市感知技术和人工智能在环境敏感的城市蓝色、绿色基础设施规划中的应用。

关教授团队发表SCI/SSCI文章60余篇，包括顶级期刊Nature Communications，Cities等。培养了多名遍及本硕博多个阶段的学生，进入哈佛大学、麻省理工学院、牛津大学、伦敦大学学院等继续深造。

重点实验室网站：https://urban.shanghai.nyu.edu/ 

谷歌学者网站：https://scholar.google.com/citations?user=U0NQImYAAAAJ