出行即服务促销策略对用户生命周期提升的因果推断研究

doi:10.16097/j.cnki.1009-6744.2025.04.004

交通运输系统工程与信息 ›› 2025, Vol. 25 ›› Issue (4): 34-43.DOI: 10.16097/j.cnki.1009-6744.2025.04.004

• 综合交通运输体系论坛 • 上一篇下一篇

出行即服务促销策略对用户生命周期提升的因果推断研究

俞诚成¹，戴一楚² ，杨超^*1,3 ，徐雷⁴ ，袁泉^1,3

1. 同济大学，城市交通研究院，上海200092；2.南京信息工程大学，数学与统计学院，南京210044；3.同济大学，道路与交通工程教育部重点实验室，上海201804；4.浙江省数智交院科技股份有限公司，物流中心，杭州310030

收稿日期:2025-02-26 修回日期:2025-03-20 接受日期:2025-03-24 出版日期:2025-08-25 发布日期:2025-08-25
作者简介:俞诚成(1999— )，男，浙江杭州人，博士生。
基金资助:
国家自然科学基金 (52172305)。

Casual Inference Research of Mobility-as-a-service Incentive Strategies on User Life-cycle Survival Time Promotion

YU Chengcheng¹, DAI Yichu², YANG Chao^*1,3, XU Lei⁴ , YUAN Quan^1,3

1. Urban Mobility Institute, Tongji University, Shanghai 200092, China; 2. School of Mathematics and Statistics, Nanjing University of Information Engineering, Nanjing 210044, China; 3. The Key Laboratory of Road and Traffic Engineering, Ministry of Education, Tongji University, Shanghai 201804, China; 4. Logistic Center, Zhejiang Institute of Communications Co Ltd, Hangzhou 310030, China

Received:2025-02-26 Revised:2025-03-20 Accepted:2025-03-24 Online:2025-08-25 Published:2025-08-25
Supported by:
National Natural Science Foundation of China (52172305)。

摘要/Abstract

摘要： 针对出行即服务平台短期促销激励策略的有效性评估与精准实施问题，本文提出“准社会实验-因果推断”融合框架，旨在量化短期促销激励策略对出行即服务用户生命周期的平均与异质性处理效应，并实现策略敏感人群的个体级识别。首先，采用准社会实验与倾向得分匹配方法消除在出行频率、消费金额、出行规律性和高峰时段出行占比这4类出行特征上的自选性偏差，在激励策略整体层级估算平均处理效应；然后，引入二元因果森林模型分析用户群组层级行为特征的异质性，计算异质性处理效应与个体处理效应分布；最后，基于个体的增益模型构建自然生存时长与增益值的四象限模型，实现在个体级别的敏感人群动态划分与优先级排序。实证研究结果表明：激励金额与平均处理效应呈正相关，但边际效应递减，尤其是高额激励(＞10元)，尽管平均处理效应为12.18d，但低额激励([1,10]元)在成本效益上更具优势(平均处理效应为11.10d，成本降低82.5%)；用户出行频率与消费金额是驱动异质性处理效应的核心因素，高频率和高消费用户响应较为显著；个体级增益模型有效识别营销敏感人群，其中，60.19%为低出行频率用户，但高消费用户中仍存在12.84%的敏感个体，验证了传统“高价值优先”策略的偏差，并辨识出低出行频率和低消费用户中分别有23.61%与17.37%的反作用人群，需避免干预。

关键词: 城市交通, 用户挽留, 因果推断, 出行即服务, 激励策略, 全生命周期

Abstract: This study addresses the issue of effectiveness evaluation and precise implementation about short-term promotional incentive strategies for Mobility-as-a-Service (MaaS) platforms. This study proposes a "quasi-social experiment-causal inference" integrated framework to quantify the average and heterogeneous treatment effects of short-term promotional incentives on MaaS users' lifecycle, and to achieve individual-level identification of strategy-sensitive user segments. First, a quasi-social experiment combining with propensity score matching is used to eliminate the selection bias across four types of travel characteristics: travel frequency, spending amount, travel regularity, and the proportion of peak-hour travel, and to estimate the average treatment effect at the overall strategy level. Then, a binary causal forest model is introduced to analyze the heterogeneity of user group-level behavioral characteristics, and to calculate the distributions of heterogeneous treatment effects and individual treatment effects. Finally, based on the individual uplift model (Uplift model, UM), a four-quadrant model of natural survival time and uplift values is constructed to achieve dynamic classification and prioritization of sensitive user segments at the individual level. The empirical results show that: (1) The incentive amount is positively correlated with average treatment effects, but with diminishing marginal effects. Although high-value incentives (>10 yuan) yield an average treatment effects of 12.18 days, low-value incentives (1~10 yuan) are more cost-effective (average treatment effects 11.10 days, with a cost reduction of 82.5%). (2) User travel frequency and spending amount are the core e of heterogeneous treatment effects , with higher-frequency and higher-spending users showing more significant responses. (3) The individual-level uplift model effectively identifies marketing-sensitive user segments, with 60.19% of them being low-frequency users. However, among high-spending users, 12.84% are still sensitive, which verifies the bias of the traditional "high-value first" strategy and identifies that 23.61% and 17.37% of low-frequency and low-spending users, respectively, belong to the counterproductive group, requiring avoidance of intervention.

Key words: urban traffic, user retention, casual inference, mobility-as-a-service, incentive strategy, lifecycle survival time

中图分类号:

U121

俞诚成, 戴一楚, 杨超, 徐雷, 袁泉. 出行即服务促销策略对用户生命周期提升的因果推断研究[J]. 交通运输系统工程与信息, 2025, 25(4): 34-43.

YU Chengcheng, DAI Yichu, YANG Chao, XU Lei, YUAN Quan. Casual Inference Research of Mobility-as-a-service Incentive Strategies on User Life-cycle Survival Time Promotion[J]. Journal of Transportation Systems Engineering and Information Technology, 2025, 25(4): 34-43.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/10.16097/j.cnki.1009-6744.2025.04.004

http://www.tseit.org.cn/CN/Y2025/V25/I4/34

参考文献

[1]孙世超,张琳琳.“出行即服务(MaaS)”核心用户群体的使用意愿机理[J].大连海事大学学报,2024,50(4):89 99. [SUN S C, ZHANG L L. The mechanism of usage intentions among core user groups in Mobility-as-a Service[J]. Journal of Dalian Maritime University, 2024, 50(4): 89-99.]

[2]李文翔,程佳楠,刘向龙,等.出行即服务环境下个体出行链碳足迹监测与评估[J].交通运输系统工程与信息, 2023, 23(2): 22-31, 53. [LI W X, CHENG J N, LIU X L, et al. Monitoring and assessing carbon footprint of individual trip chain in environment of mobility as a service[J]. Journal of Transportation Systems Engineering and Information Technology, 2023, 23(2): 22-31, 53.]

[3]LIANG Z, TANG Y, YU J, et al. A collective incentive strategy to manage ridership rebound and consumer surplus in mass transit systems[J]. Transportation Research Part A: Policy and Practice, 2024, 182(4): 104031.

[4]李婉莹,关宏志,韩艳,等.社会偏好视野下旅游出行服务链定价博弈模型[J].交通运输系统工程与信息, 2022, 22(4): 11-22. [LI W Y, GUAN H Z, HAN Y, et al. Pricing game model of travel service chain under social preference view[J]. Journal of Transportation Systems Engineering and Information Technology, 2022, 22(4): 11-22.] [5]刘好德,钱贞国,刘向龙,等.城市MaaS研究热点文献计量分析与发展启示[J]. 交通运输研究,2022,8(3): 130-142. [LIU H D, QIAN Z G, LIU X L, et al. Bibliometric research and development enlightenments of urban MaaS hotspots[J]. Transport Research, 2022, 8 (3): 130-142.] [6]TSOUROS I, TSIRIMPA A, PAGONI I, et al. MaaS users: Who they are and how much they are willing-to-pay[J]. Transportation Research Part A: Policy and Practice, 2021, 148(6): 470-480. [7]SCHIKOFSKY J, DANNEWALD T, KOWALD M. Exploring motivational mechanisms behind the intention to adopt mobility as a service (MaaS): Insights from Germany [J]. Transportation Research Part A: Policy and Practice, 2020, 131(1): 296-312.

[8]LI Y, MAY A, COOK S, et al. Literature review: Why do we need innovative design methods for future Mobility-as-a-Service (MaaS)?[J]. Transportation Research Interdisciplinary Perspectives, 2024, 27(9): 101233.

[9] BRIAND A-S, CôME E, TRéPANIER M, et al. Analyzing year-to-year changes in public transport passenger behaviour using smart card data[J]. Transportation Research Part C: Emerging Technologies, 2017, 79(6): 274-289.

[10] LI M. Using the propensity score method to estimate causal effects: A review and practical guide[J]. Organizational Research Methods, 2013, 16(2): 188-226.

[11] NASRI A, CARRION C, ZHANG L, et al. Using propensity score matching technique to address self selection in transit-oriented development (TOD) areas[J]. Transportation, 2020, 47: 359-371.

[12] MARINESCU I E, LAWLOR P N, KORDING K P. Quasi experimental causality in neuroscience and behavioral research[J]. Nature Human Behaviour, 2018, 2(12): 891 898.

[13] ATHEY S, IMBENS G. Recursive partitioning for heterogeneous causal effects[J]. Proceedings of the National Academy of Science of the United States of America, 2016, 113(27): 7353-7360.

[14] KüNZEL S R, SEKHON J S, BICKEL P J, et al. Metalearners for estimating heterogeneous treatment effects using machine learning[J]. Proceedings of the National Academy of Science of the United States of America, 2019, 116 (10): 4156-4165.

[15] 龚鹤扬.基于信息视角的因果建模及其在互联网个性化激励增益建模中的应用[D].合肥:中国科学技术大学, 2021. [GONG H Y, An informative view of causality and its applications in industrial personalize incentives with Uplift modeling[D]. Hefei: University of Science and Technology of China, 2021.

出行即服务促销策略对用户生命周期提升的因果推断研究

Casual Inference Research of Mobility-as-a-service Incentive Strategies on User Life-cycle Survival Time Promotion

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	谭德明, 陈可沛, 吴大维, 李延欢, 胡四新, 张彩平. 低碳视角下轨道交通与土地利用供需协同发展研究[J]. 交通运输系统工程与信息, 2025, 25(4): 13-23.
[2]	张鹏羽, 李正中, 张翕然, 岳晓辉. 不同时期下城市轨道交通客流的时空影响机制研究[J]. 交通运输系统工程与信息, 2025, 25(4): 24-33.
[3]	王海涌, 张丹, 王孟琳, 田爱爱. 异质交通流下交叉口信号及车辆轨迹融合控制模型[J]. 交通运输系统工程与信息, 2025, 25(4): 96-103.
[4]	宋翠颖, 丁杰, 张春波. 模块化公交车辆调度研究综述[J]. 交通运输系统工程与信息, 2025, 25(4): 175-192.
[5]	雷雪, 范鹏飞, 刘睿, 李松松, 吴亦政, 宋国华. 插电式混合动力汽车效用因子及其减排效益研究[J]. 交通运输系统工程与信息, 2025, 25(4): 337-348.
[6]	王殿海, 徐梦丹, 张萌, 曾佳棋, 蔡正义. 基于客流争夺与补充的轨交对公交非对称竞合关系评价[J]. 交通运输系统工程与信息, 2025, 25(3): 22-31.
[7]	祝静静, 许项东. 路径冗余性对行程时间可靠性的影响分析[J]. 交通运输系统工程与信息, 2025, 25(3): 142-151.
[8]	方腾源, 徐峰祥, 朱其茂, 邹震. 时空约束下的道路交通事故与违法行为关联规则研究[J]. 交通运输系统工程与信息, 2025, 25(3): 246-254.
[9]	王学楷, 鱼凯, 闫茂德. 动态功率下城轨单向供电不足区域行车调整方法[J]. 交通运输系统工程与信息, 2025, 25(3): 255-265.
[10]	李海军, 赵颖, 黄岩, 田卫刚. 城轨虚拟编组列车群多状态运行模型构建与仿真研究[J]. 交通运输系统工程与信息, 2025, 25(3): 266-275.
[11]	郝妍熙, 刘巍, 胡华, 方勇, 刘志钢. 改进的L型通道行人转弯行为仿真模型[J]. 交通运输系统工程与信息, 2025, 25(3): 383-392.
[12]	晋泽倩, 俞诚成, 叶昕. 混合效用和后悔规则下的洪涝交通疏散决策行为研究[J]. 交通运输系统工程与信息, 2025, 25(2): 16-25.
[13]	周雨阳, 赵聪颖, 李京昆, 陈艳艳, 柳堤, 王书灵. 多类城市职住及通勤状态对出行方式选择影响比较[J]. 交通运输系统工程与信息, 2025, 25(2): 26-35.
[14]	彭其渊, 刘思源, 江山, 冯涛, 陈垚, 张永祥. 贯通运营下市域铁路与地铁列车开行方案协同优化[J]. 交通运输系统工程与信息, 2025, 25(2): 36-47.
[15]	蒋阳升, 叶肖甫, 徐尉耀, 刘洪汛, 胡路. 考虑转运与拼车的共享自动驾驶电动汽车动态运营[J]. 交通运输系统工程与信息, 2025, 25(2): 58-68.