|
|
| |
|
|
| 本文已被：浏览 57次 下载 0次 |
|
|
|
| 基于居民活动数据的产城融合单元识别与评价—以武汉市为例 |
|
焦洪赞1, 赵灿1, 潘启胜2, 刘学军1, 李延新3, 王彤3
|
|
1.武汉大学;2.同济大学;3.武汉市自然资源保护利用中心
|
|
| 摘要: |
| 合理识别居民活动空间并基于此评价产城融合单元对于实现职住平衡和优化城市功能布局具有重要意义。然而，传统方法多依赖行政区划或土地利用规划，未能充分考虑居民实际活动行为的动态特征。提出一种基于大数据的居民活动空间识别与产城融合单元评价方法：利用手机信令数据提取居民的居住、工作及生活活动范围，通过社区发现算法精确识别居民活动空间的特征与边界，进而构建产城融合研究单元。为验证有效性，以武汉市为案例进行实证分析，结果显示，识别出的居民活动空间与研究单元呈现显著的空间分异特征，中心城区显示出高密度的职住活动模式，而外围区域则体现功能分散的特征，验证了其在产城融合研究中的实用性与创新性。 |
| 关键词: 产城融合 社区发现算法 出行大数据 居民活动空间 单元识别 |
| DOI： |
| 分类号:TU 984.11 |
| 基金项目: |
|
| Identification and Evaluation of Industry-City Integration Units Based on Resident Activity Data: A Case Study of Wuhan |
|
Jiao HongZan1, ZhaoCan1, Pan QiSheng2, Liu XueJun1, Li YanXin3, Wang Tong3
|
|
1.Wuhan University;2.TongJi University;3.WUHAN NATURAL RESOURCES CONSERVATION AND UTILIZATION CENTER
|
| Abstract: |
| The accurate identification of resident activity spaces and their subsequent evaluation for industry-city integration units are critical for achieving job-housing balance and optimizing urban functional layouts, which are foundational goals for sustainable and livable urban development. However, traditional urban analysis and planning methods, which predominantly rely on static, politically defined administrative divisions or predefined, often idealized land-use planning schemes, often fail to capture the dynamic, fluid, and complex characteristics of residents' actual daily movements and spatial behaviors. This fundamental disconnect between planning frameworks and lived spatial reality can lead to significant inefficiencies, including prolonged commutes, traffic congestion, segregated urban functions, and ultimately, a diminished quality of life. To bridge this critical gap, this study proposes an innovative, bottom-up, and data-driven computational framework for identifying genuine, organic resident activity spaces and, based on them, constructing and evaluating Production-City Integration Units (PCIUs). The methodology is grounded in the analysis of large-scale, passive mobile phone signaling data, which provides continuous, high-resolution spatiotemporal footprints of a large and representative sample of the urban population. Through a series of advanced data mining and cleaning processes, we extract and geographically delineate the core activity anchors for individuals—primarily distinguishing between residential locations (identified through prolonged nighttime presence), employment centers (identified through consistent daytime presence on weekdays), and nodes of other lifestyle and recreational activities. The core analytical innovation of this research lies in the application of community detection algorithms from network science and complex systems theory. We construct a city-wide spatial interaction network, where nodes represent fine-grained geographic areas (e.g., traffic analysis zones or grid cells), and the links between nodes are weighted by the intensity of human movement flows derived from the mobile data. Algorithms such as the Louvain method are then employed to detect densely connected groups of areas—communities—within this mobility network. These algorithmically derived clusters represent naturally emergent Functional Urban Areas or resident activity spaces. Their boundaries are not drawn on a map but are revealed by the data, precisely delineating areas of intense internal spatial interaction. These units form the fundamental, behavior-based PCIUs for all subsequent evaluation, replacing arbitrary administrative units with functional ones. To rigorously validate the effectiveness, robustness, and practical utility of this method, a comprehensive empirical case study was conducted in Wuhan, China, a major metropolitan region undergoing rapid transformation. The application yielded highly insightful and spatially explicit results. The analysis revealed significant and systematic spatial heterogeneity in the morphology and function of the identified PCIUs. Within the central urban core and inner-city districts of Wuhan, the PCIUs are typically compact, densely interconnected, and exhibit a highly integrated, high-density job-housing activity pattern. This indicates a strong degree of functional mix and potential for shorter, more sustainable internal commutes. Conversely, the PCIUs identified in peripheral, suburban, and emerging urban expansion areas present a stark contrast. They are generally larger in spatial extent and demonstrate a more dispersed, segregated, and often monofunctional layout. Many are predominantly residential "bedroom communities" with weak internal employment bases, while others appear as isolated employment sub-centers. This pattern highlights a spatial structure reliant on long-distance cross-city commuting, underscoring classic challenges of urban sprawl and functional separation. These findings robustly confirm the practicality, innovativeness, and diagnostic power of the proposed methodology. It provides urban planners, policymakers, and researchers with a powerful, evidence-based, and finely-grained analytical tool. This tool moves beyond descriptive land-use maps to offer a dynamic, human-centric understanding of the city's actual functional anatomy. The PCIU framework enables targeted, context-sensitive interventions for improving public transit networks, optimizing the allocation of public services, promoting local job-housing balance, and guiding strategic land-use policy, thereby contributing directly to the creation of more efficient, equitable, and people-oriented urban environments. |
| Key words: Urban-industrial integration, Community detection algorithm, Mobility big data, Resident Activity Space,Unit recognition |
|
|
|
|
