摘要
城镇化、机动化的持续快速发展催生了城市停车问题,为缓解日益严峻的停车窘境,共享停车应运而生并受到广泛关注与研究。为弥补经典共享泊位分配模型对用户停车成本刻画及用户效益评价的缺失,基于共享停车既有研究及调研数据,构建共享停车选择意愿结构方程模型以分析用户偏好,标定结果显示:模型识别出停车费用、步行距离、违停罚款、寻停时间四类停车偏好,且路径系数依次降低。将结构方程模型识别结果及路径系数作为用户停车成本建模的依据,以经营平台收益与用户停车成本的差值表示系统效益,将系统效益最大化作为优化目标,综合泊位时空间约束,提出考虑用户偏好的停车场共享泊位分配模型。设计随机生成分配方案的蒙特卡洛算法,定义平台让利程度以评价用户效益。数值算例表明:设计的蒙特卡洛算法能够有效求解模型,在模拟次数约10000次后得到系统效益最大值;考虑用户偏好的停车场共享泊位分配模型能够将用户效益提升约7.5%,为共享经营平台权衡自身收益及用户效益提供一定的理论参考。
Continuous and rapid urbanization and motorization has led to the problem of urban parking problems.To alleviate this increasingly severe parking dilemma,shared parking has emerged and received extensive attention and research.To address the shortcomings of the classic shared parking space allocation model in characterizing user parking costs and evaluating user benefits and preferences,this study constructs a structural equation model(SEM)of shared parking choice willingness based on shared parking theory and survey data.The calibrated results reveal that the model identifies four types of parking preferences with decreasing path coefficients:parking fees,walking distance,illegal parking fines,and time spent searching for parking.The identification results and path coefficients of the SEM serve as the basis for user parking cost modeling,with the difference between platform revenue and user parking costs representing the system benefits.We set maximizing system such benefits as the optimization objective,considering both user preferences and spatial-temporal constraints of parking spaces.The study proposes a shared parking space allocation model for parking facilities,taking user preferences into account.A monte carlo algorithm is designed to generate random allocation schemes,and the extent of profit sacrifice by the platform is defined to evaluate user benefits.Numerical examples demonstrate that the algorithm can effectively solve the model and obtain the maximum system benefit after approximately 10,000 simulations.The parking facility shared parking space allocation model considering user preferences can increase user benefits by approximately 7.5%,thereby providing theoretical references for shared platform operators to balance their own profits and user benefits.
作者
蒋军
张义
马志明
JIANG Jun;ZHANG Yi;MA Zhiming(College of Traffic and Transportation,Chongqing Jiaotong University,Chongqing 400074,China)
出处
《交通运输工程与信息学报》
2024年第3期196-206,共11页
Journal of Transportation Engineering and Information
基金
国家自然科学基金项目(52072264)。
关键词
城市交通
共享停车
结构方程模型
蒙特卡洛算法
泊位分配模型
urban transportation
shared parking
structural equation model
monte carlo algorithm
parking lots allocation model
