数据驱动的国土空间规划:理论、范式及趋势

Data-driven Territorial Space Planning:Theory,Paradigm and Trends

研究目的:从“新数据”与国土空间规划之间的关系出发,分析数据驱动规划的理论基础、基本概念、逻辑体系、主要范式、本质特征,以期为构建面向未来人与自然和谐共生的中国式现代化国土空间治理新格局提供理论参考与决策依据。研究方法:文献研究法和系统分析法。研究结果:(1)数据驱动规划是一种通过模型、算法实现多源异构数据的获取筛选、分析挖掘、模拟预测,实现国土空间规划全流程、全链条、全方位定量化的重要模式方法,具有系统性、科学性、实质性等“3S”特性;(2)数据驱动规划遵循基于“总量—增量—存量—流量—质量”过程优化的自然空间、基于“流态—业态—形态”多维复合的社会空间、基于“数字化—智能化—智慧化”技术支撑的数字空间三大逻辑体系;(3)数据驱动规划可划分为全要素协同驱动、多时空尺度驱动、全生命周期驱动和多元化场景驱动4种范式,其本质是以数据要素为核心与纽带,有效链接和无缝衔接自发生长型规划和人为引导型规划,是一种追求人与自然和谐共生的规划,并呈现出全面感知、适应学习、反馈治理等主要特征。研究结论:面向未来数字技术发展趋势,应加强“大数据”和“小数据”以及“新数据”和“老数据”之间的结合,“真数据”和“伪数据”之间的鉴别,以提高数据驱动规划的全面性和有效性;此外,应加强“生成”与“构成”以及“传统”与“新兴”之间的结合,以进一步体现数据驱动规划“以人为本、和谐共生”的核心价值和终极目标。

The purposes of this study are to explore the relationship between new data and territorial space planning under the influence of technological changes,and to deeply analyze the theoretical basis,basic concepts,logical system,main paradigms and essential characteristics of data-driven planning.It provides theoretical reference and decision-making basis for building a new pattern of territorial space governance for the Chinese path to modernization facing the harmonious coexistence of human and nature in the future.The research methods include literature review and systematic analysis.The result shows that 1)data-driven planning is an important mode that achieves the acquisition,screening,analysis,mining,simulation and prediction of multi-source heterogeneous data through models and algorithms,and achieves the full process,full chain,and all-round quantification of territorial space planning.It has the 3S characteristics,which are systematic,scientific,and substantive.2)Data-driven planning follows three logical systems:a natural space based on a process optimization of“total quantity-incremental quantity-stock quantity-flow quantity-quality”,a social space based on a multidimensional composite of“flow pattern-business format-form”,and a digital space based on a technical support of“digitization-intelligence-wisdom”.3)Data-driven planning can be divided into four paradigms:full-factor collaborative driving,multi temporal-spatial scale driving,full lifecycle driving,and diversified scene driving.Its essence is to effectively link and seamlessly connect self-occurring planning and human-guided planning by using data elements,which is a planning that pursues harmonious coexistence between humans and nature.And it presents the main characteristics of comprehensive perception,adaptive learning,and feedback governance.In conclusion,facing the future development trend of digital technology,we should strengthen the combination between big data and small data,new data and old data,and the identification between real data and fake data to improve the comprehensiveness and effectiveness of data-driven planning.In addition,we should strengthen the combination between generation and composition and tradition and emerging,to further reflect the core values and ultimate goals of data-driven planning that are“people-oriented,harmonious coexistence”.