Abstract: Based on the manufacturing data of 11 provinces in the western region from 2015 to 2022, the Moran Index and social network analysis method combined with the modified gravity model were used to construct the high-quality development correlation network of the manufacturing industry, and the spatial structure characteristics and evolution rules of the manufacturing network in the western region were analyzed. The research results show that：1） The high-quality development of manufacturing industry in western China shows a significant positive correlation and is relatively stable according to the Moran Index. 2） Sichuan and Chongqing have a strong driving effect on the surrounding radiation, Xinjiang tends to develop independently, and Ningxia will stay at the edge of the network for a long time before 2022. 3） The network density and network efficiency of the high-quality spatial correlation network in the western region, obtained by using the improved gravity model, are generally on the rise. （4） The degree centrality of the network is gradually increasing, the inter-provincial connections are increasingly close, the intermediary centrality is also increasing, and the network structure is gradually developing toward equilibrium. Therefore, the high-quality development of manufacturing industry in the western region has significant spatial differentiation and unbalanced development characteristics, but the difference is gradually narrowing. Through the analysis, the following conclusions are drawn： the overall strength of the southwest region exceeds that of the northwest region, mainly because its economic base and infrastructure are superior, attracting a large number of production factors and promoting the rapid development of the manufacturing industry. Sichuan and Chongqing, as engines of regional development, have played a significant role in driving the surrounding areas, while Shaanxi is also gradually increasing its influence. Although Xinjiang and Ningxia are relatively backward, they have gradually improved their development level through regional cooperation and industrial transfer. Shaanxi has promoted industrial upgrading through scientific and technological innovation, Ningxia has accelerated the transformation of old and new growth drivers through industrial transformation and upgrading, and Xinjiang has made progress in equipment manufacturing, forming an industrial system with local characteristics. These development strategies and practices have provided strong support for the high-quality development of manufacturing industry in the western region.

Key words: manufacturing industry, gravity model, social network analysis, Moran's index, western region

摘要： 基于2015—2022年西部地区11个省制造业数据，运用莫兰指数和社会网络分析方法，结合修正的引力模型，构建制造业高质量发展关联网络，分析西部地区制造业网络空间结构特征以及演变规律。研究结果表明：1）通过莫兰指数测算出西部地区制造业高质量发展呈现显著的正相关性，且较为稳定。2）四川和重庆对周围辐射带动作用较强，新疆则是倾向于自身独立发展，宁夏则是在2022年之前长期游离在网络边缘。3）利用改进引力模型得到的西部地区高质量发展空间关联网络的网络密度和网络效率总体呈上升趋势。4）网络的度数中心度逐渐增加，省际间联系日益紧密，中介中心度也呈增加趋势，网络结构逐渐向均衡化发展。因此，西部地区制造业高质量发展存在显著的空间分异和非均衡发展特点，但是。其差异逐渐缩小。通过分析得出以下结论：西南地区整体实力超过西北地区，主要因为其经济基础和基础设施更优越，吸引了大量生产要素，促进了制造业的快速发展。四川和重庆作为区域发展的引擎，对周边地区有显著的带动作用，而陕西也在逐渐提升其影响力。新疆和宁夏虽然相对落后，但在区域合作和产业转移中逐步提升发展水平。陕西通过科技创新推动产业升级，宁夏则通过产业转型升级加快新旧动能转换，新疆则在装备制造业方面取得进展，形成了具有地方特色的产业体系。这些发展策略和实践为西部地区制造业的高质量发展提供了有力支撑。

关键词: 制造业, 引力模型, 社会网络分析, 莫兰指数, 西部地区