基于多维成本的出行方式选择建模及调控

doi:10.13229/j.cnki.jdxbgxb.20250429

摘要/Abstract

摘要：

为揭示城市交通拥堵背景下居民出行方式选择的复杂影响机制，并优化公共交通与私家车出行结构，提出了一种基于多维成本的出行方式选择建模及调控方法。首先，通过整合个体一天内的出行记录，生成8 368条单日出行链，融合个体社会属性特征，构建多项式Logit模型解析出行决策行为；其次，基于13 806条出行记录，结合百度地图API获取多模式交通时空数据，并在多种成本情景下构建5个机器学习模型；最后，进行政策成本调控分析。结果表明，出行费用和时间成本显著影响出行方式选择，5个机器学习模型中XGBoost模型预测精度最高。当公共交通出行成本降低20%时，选择私家车出行的人数比例出现显著下降，降幅达4.65%。本文证实了成本调控工具在出行行为诱导中的有效性，为实施差别化交通定价、推进城市出行结构绿色转型提供了决策依据。

关键词: 交通规划, 出行成本, 出行方式选择, 多项式Logit模型, 机器学习模型

Abstract:

To elucidate the complex mechanisms underlying residents' travel mode choice under urban traffic congestion and to optimize the modal composition of public transport and private-car travel， a multidimensional-cost-based method for travel mode choice modeling and regulation was proposed. By integrating individual daily travel records， 8 368 daily trip chains were generated， and a multinomial Logit model incorporating socioeconomic attributes was developed.Results showed that expenses costs and time costs significantly influenced mode selection. Subsequently， 13 806 filtered travel records were fused with multimodal spatiotemporal data from Baidu Map API to establish cost scenarios. Comparative analysis of machine learning models identified XGBoost as the optimal predictor. Cost regulation simulations demonstrated that reducing public transit costs by 20% led to a significant decline in the proportion of private car usage， with a decrease of 4.65 percentage points. These findings validate the efficacy of cost-driven policy tools in steering travel behavior， providing strategies for differentiated transportation pricing and advancing green urban mobility transitions.

Key words: transportation planning, travel costs, travel mode choice, multinomial Logit model, machine learning modes

中图分类号:

U491.1

孙轶琳,管展易,麻书政,孙浩珊,冯源,黄毅方. 基于多维成本的出行方式选择建模及调控[J]. 吉林大学学报(工学版), 2026, 56(10): 2617-2625.

Yi-lin SUN,Zhan-yi GUAN,Shu-zheng MA,Hao-shan SUN,Yuan FENG,Yi-fang HUANG. Modeling and regulation of travel mode choice based on multi⁃dimensional cost[J]. Journal of Jilin University(Engineering and Technology Edition), 2026, 56(10): 2617-2625.

图/表 8

表1

表2

图1

表3

MNL模型参数估计结果"

变量	系数
变量	公共交通出行	主动出行	出租出行
$λ i$	-4.695 5^***	-9.882 9^***	0.704 7^***
0~17岁	-0.522 9^*	-0.689 5^**	-0.762 3^***
18~59岁	0.279 4	-0.132 5	-0.249 5^*
60岁以上	—	—	—
0~10万元	1.202 1^***	0.945 2^**	0.242 1
10~50万元	0.810 0^***	0.633 6^**	0.012 7
50万元以上	—	—	—
持有驾照	-1.060 7^***	-0.951 8^***	-1.452 1^***
出行费用/元	-40.848 1^***	-52.565 7^***	-1.152 9^***
时间成本/元	12.712 8^***	10.945 1^***	0.254 3
换乘成本/元	-1.101 5^***	-1.790 9^***	-0.315 1^***
体力感知成本/元	-4.978 6^***	-4.849 0^***	0.610 5^***
McFadden's R2	0.692 3
Cox & Snell R2	0.803 6
Nagelkerke R2	0.888 2

表3

图2

表4

图3

图4

参考文献 28

[1]	Bai T Z, Li X M, Sun Z Y. Effects of cost adjustment on travel mode choice: analysis and comparison of different logit models[C]∥Proceedings of the 14th World Conference on Transport Research(WCTR), Shanghai, China, 2017: 2653-2663.
[2]	Andani I G A, Puello L L P, Geurs K. Modelling effects of changes in travel time and costs of toll road usage on choices for residential location, route and travel mode across population segments in the Jakarta-Bandung region, Indonesia[J]. Transportation Research Part A: Policy and Practice, 2021, 145: 81-102.
[3]	Huang Y Q, Gao L J, Ni A N, et al. Analysis of travel mode choice and trip chain pattern relationships based on multi-day GPS data: a case study in Shanghai, China[J]. Journal of Transport Geography, 2021, 93: 103070.
[4]	Long Y, Ao Y B, Li H M, et al. Non-linear effects of children's daily travel distance on their travel mode choice considering different destinations[J]. Journal of Transport Geography, 2024, 118: 103921.
[5]	Hamadneh J, Jaber A. Modeling of intra-city transport choice behaviour in Budapest, Hungary[J]. Journal of Urban Mobility, 2023, 3: 100049.
[6]	杨励雅, 赵鹏军. 考虑偏好差异与备选方案相关性的居民出行方式选择模型[J]. 北京大学学报: 自然科学版, 2012, 48(6): 1009-1015.
	Yang Li-ya, Zhao Peng-jun. Travel mode choice model accounting for individual preference heterogeneity and correlation among choice alternatives[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2012, 48(6): 1009-1015.
[7]	Yang Y N, Li N. Research on Residents' travel behavior based on multiple logistic regression model[J]. IEEE Access, 2023, 11: 74759-74767.
[8]	Lu X S, Liu T L, Huang H J. Pricing and mode choice based on nested Logit model with trip-chain costs[J]. Transport Policy, 2015, 44: 76-88.
[9]	孙轶琳, 黄佩, 汤心怡, 等. 基于多项Logit模型的游客组合出行方式选择行为研究[J]. 综合运输, 2024, 46(5): 100-106, 113.
	Sun Yi-lin, Huang Pei, Tang Xin-yi, et al. Tourists' travel mode choices based on multinomial logit model[J]. China Transportation Review, 2024, 46(5): 100-106, 113.
[10]	Cheng L, Chen X W, de Vos J, et al. Applying a random forest method approach to model travel mode choice behavior[J]. Travel Behaviour and Society, 2019, 14: 1-10.
[11]	Hagenauer J, Helbich M. A comparative study of machine learning classifiers for modeling travel mode choice[J]. Expert Systems with Applications, 2017, 78: 273-282.
[12]	Kashifi M T, Jamal A, Kashefi M S, et al. Predicting the travel mode choice with interpretable machine learning techniques: a comparative study[J]. Travel Behaviour and Society, 2022, 29: 279-296.
[13]	Li X W, Shi L X, Shi Y, et al. Exploring interactive and nonlinear effects of key factors on intercity travel mode choice using XGBoost[J]. Applied Geography, 2024, 166: 103264.
[14]	Tang L, Lin X Z, Yu J C, et al. Integrating travel survey and Amap API data into travel mode choice analysis with interpretable machine learning models: a case study in China[J]. IEEE Access, 2025, 13: 27852-27867.
[15]	Vasudevan N, Gore N, Zope R, et al. Determining mode shift elasticity based on household income and travel cost[J]. Research in Transportation Economics, 2021, 85: 100771.
[16]	王殿海, 李逸文, 蔡正义. 收费及奖励策略对通勤出行方式选择的影响分析[J]. 交通运输系统工程与信息, 2024, 24(2): 1-12.
	Wang Dian-hai, Li Yi-wen, Cai Zheng-yi. Impact of charging and incentive strategies on commuting mode choice[J]. Journal of Transportation Systems Engineering and Information Technology, 2024, 24(2): 1-12.
[17]	Xie L Y, Wan H N, Wang Y C. Distributional effects of subway fare surges: evidence from Beijing[J]. Cities, 2025, 158: 105602.
[18]	Zhang Y H, Li H J, Ren G. Data-driven exploration of heterogeneous gasoline price elasticities using generalized random forests[J]. Transportation, 2025, 52(1): 215-237.
[19]	叶盈. 出行方式链成本测算方法研究[D]. 杭州: 浙江大学建筑工程学院, 2015.
	Ye Ying. Study on the method of travel mode chaining cost[D]. Hangzhou: College of Civil Engineering and Architecture, Zhejiang University, 2015.
[20]	孙立军. 沥青路面结构行为理论[M]. 上海:同济大学出版社, 2003.
[21]	王忠仁. 路面管理系统的车辆运营费模型[J]. 北京建筑工程学院学报, 1992(2): 62-69, 34.
	Wang Zhong-ren. Vehicle operating cost model in pavement management system[J]. Journal of Beijing University of Civil Engineering and Architecture, 1992(2): 62-69, 34.
[22]	隗海林, 王劲松, 王云鹏, 等. 基于城市道路工况的汽车燃油消耗模型[J]. 吉林大学学报: 工学版, 2009, 39(5): 1146-1150.
	Kui Hai-lin, Wang Jin-song, Wang Yun-peng, et al. Vehicle fuel consumption model based on urban road operations[J]. Journal of Jilin University(Engineering and Technology Edition), 2009, 39(5): 1146-1150.
[23]	Gwilliam K M. The value of time in economic evaluation of transport projects: lessons from recent research[J/OL]. [2025-04-25]..
[24]	Jara-díaz S R, Gschwender A. Towards a general microeconomic model for the operation of public transport[J]. Transport Reviews, 2003, 23(4): 453-469.
[25]	王殿海, 叶盈, 朱文韬, 等. 基于条件价值法的公交车内拥挤成本测算模型[J]. 吉林大学学报: 工学版, 2016, 46(1): 57-62.
	Wang Dian-hai, Ye Ying, Zhu Wen-tao, et al. Estimation of crowding costs in bus based on contingent valuation method[J]. Journal of Jilin University(Engineering and Technology Edition), 2016, 46(1): 57-62.
[26]	Gschwender A, Jara-diaz S, Bravo C. Feeder-trunk or direct lines? Economies of density, transfer costs and transit structure in an urban context[J]. Transportation Research Part A: Policy and Practice, 2016, 88: 209-222.
[27]	杭州市统计局.2024杭州市统计年鉴[EB/OL]. [2025-04-30]..
[28]	王江波, 连芝锐, 冯涛, 等. 基于机器学习的时空出行选择行为研究综述与展望[J]. 地理科学进展, 2024, 43(8): 1649-1665.
	Wang Jiang-bo, Lian Zhi-rui, Feng Tao, et al. A review and outlook of machine learning-based travel choice behavior research[J]. Progress in Geography, 2024, 43(8): 1649-1665.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

出行方式	拥有成本	使用成本
电动自行车	折旧成本	维修成本、耗电成本
共享自行车	—	使用费用
公共汽车	—	票费
轨道交通	—	票费
燃油汽车	折旧成本、保险费	燃油消耗、停车费、维修成本
电动汽车	折旧成本、保险费	电费消耗、停车费、维修成本
网约车/出租车	—	打车费

变量类型	属性	变量名称	说明
特征变量	个人属性	0~17岁	年龄在0~17岁之间（包含0和17）取1；否则取0
		18~59岁	年龄在18~59岁之间（包含18和59）取1；否则取0
		60岁以上	年龄在60岁以上（包含60）取1；否则取0
		0~10万元	家庭年收入低于10万元取1；否则取0
		10~50万元	家庭年收入高于10万元且低于50万元取1；否则取0
		50万元以上	家庭年收入高于50万元取1；否则取0
		持有驾照	持有驾照取1；否则取0
	出行属性 MNL模型	出行费用C	单日出行费用总和，元
		时间成本TC	单日出行时间成本总和，元
		换乘成本TRC	单日出行换乘成本总和，元
		体力感知成本PC	单日出行体力感知成本总和，元
	出行属性机器学习模型	距离	出发地与目的地的距离，km
		私家车成本	采取私家车出行情况下的4项成本总和与距离之比，元/km
		公共交通成本	采取公共交通情况下的4项成本总和与距离之比，元/km
		骑行成本	采取骑行情况下的4项成本总和与距离之比，元/km
		步行成本	仅采取步行情况下的4项成本总和与距离之比，元/km
		网约车/出租车成本	采取网约车或出租车出行情况下的4项成本总和与距离之比，元/km
目标变量	出行方式	私家车出行	包含以私家车为主的出行方式
		公共交通出行	包含轨道交通、公交的出行方式
		主动出行	包含步行、骑行的出行方式
		出租出行	包含网约车、出租车的出行方式

私家车成本调整幅度/%	私家车比例变化/%	公共交通成本调整幅度/%	私家车比例变化/%
-25	+10.97	-25	-4.86
-20	+9.53	-20	-4.65
-15	+7.08	-15	-2.83
-10	+5.57	-10	-1.53
-5	+3.75	-5	-0.21
+5	-2.05	+5	+2.15
+10	-3.08	+10	+2.41
+15	-4.81	+15	+2.04
+20	-6.86	+20	+2.25
+25	-7.96	+25	+2.29