中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊
重庆交通大学学报（自然科学版）

重庆交通大学学报（自然科学版） ›› 2026, Vol. 45 ›› Issue (9): 89-98.DOI: 10.3969/j.issn.1674-0696.2026.09.11

• 交通运输+人工智能 • 上一篇    

基于云计算与Stackelberg博弈的客流诱导与承载管理协同优化

何若冰1，王小刚1，陈浪2，李璐璐3，杨欣3   

  1. (1. 重庆交通建设管理有限公司，重庆 401121； 2. 中国水利水电第七工程局有限公司，四川 成都 610213； 3. 北京交通大学 系统科学学院，北京 100044)
  • 收稿日期:2026-01-31 修回日期:2026-06-30 发布日期:2026-09-23
  • 作者简介:何若冰（1986—），女，陕西西安人，高级工程师，主要从事城市轨道交通方面的工作。E-mail:32950591@qq.com 通信作者：李璐璐（1994—），女，河北沧州人，博士研究生，主要从事交通系统建模与优化方面的研究。E-mail:23111561@bjtu.edu.cn.
  • 基金资助:
    国家自然科学基金项目（72331001）

Collaborative optimization of passenger flow guidance and carrying capacity management based on cloud computing and Stackelberg game

He Ruobing1, Wang Xiaogang1, Chen Lang2, Li Lulu3, Yang Xin3   

  1. (1. Chongqing Transportation Construction Management Co., Ltd., Chongqing 401121, China; 2. Sinohydro Bureau 7 Co., Ltd., Chengdu 610213, Sichuan, China; 3. School of Systems Science, Beijing Jiaotong University, Beijing 100044, China)
  • Received:2026-01-31 Revised:2026-06-30 Published:2026-09-23

PDF

摘要： 针对城市轨道交通系统中客流诱导与客流承载管理的协同优化问题，从“计算—经济”融合视角出发，构建了一个基于“云—边—端”协同架构的Stackelberg博弈模型。该模型将云计算能力内生化为预测误差方差，系统揭示了其通过影响客流预测精度，进而作用于运营决策与经济绩效的传导机制。采用逆向归纳法求解博弈均衡，并结合数值仿真进行分析。结果表明：最优诱导强度与发车频率决策对云预测精度高度敏感，提升云计算能力可驱动诱导强度提高使发车频率调整更为精细；系统总成本随云计算能力增强呈下降趋势，且边际效益递减，据此可确定云资源投资的经济阈值；云计算通过降低信息不确定性，能显著增强诱导策略与客流承载管理间的战略互补性，提升协同效率。

关键词: 交通运输工程；城市轨道交通；云计算；Stackelberg博弈；客流协同优化；计算—经济均衡

Abstract: Focusing on the collaborative optimization of passenger flow guidance and carrying capacity management in urban rail transit systems, a Stackelberg game model based on a “cloud-edge-terminal” collaborative architecture was constructed from an integrated “computational-economic” perspective. The proposed model endogenized cloud computing capability as the variance of prediction errors, systematically revealing its transmission mechanism that influenced operational decisions and economic performance by affecting passenger flow prediction accuracy. The game equilibrium was solved by backward induction method and analyzed through numerical simulation. The results show that the optimal guidance intensity and train dispatch frequency decisions are highly sensitive to cloud prediction accuracy. Enhanced cloud computing capability can drive an increase in guidance intensity and more precise adjustment of departure frequency. The system total cost decreases with the improvement of cloud computing capability, exhibiting diminishing marginal returns, which provides a clear economic threshold for cloud resource investment. By reducing information uncertainty, cloud computing significantly strengthens the strategic complementarity and synergistic efficiency between guidance strategies and capacity management.

Key words: traffic and transportation engineering; urban rail transit; cloud computing; Stackelberg game; passenger flow collaborative optimization; computational-economic equilibrium

中图分类号: