Section for Travel Mode and Departure Time Tea eam C Li Linghan - PowerPoint PPT Presentation

Nested Logit Model of Combined Section for Travel Mode and Departure Time Tea eam C Li Linghan Zh Zhang , Tuqi uqiang Zh Zhou u , Wei eiyan Zo Zong 1

１．１ Data processing 2 more variables Sun Mon Tue Wen Thu Fri Sat 29 30 31 0:00 7:00 9:30 16:00 19:30 24:00 1 2 3 4 5 6 7 period period period period period 1 2 3 4 5 8 9 10 11 12 13 14 15 16 17 18 19 20 21 All-day time division 22 23 24 25 26 27 20 weekdays + 10 weekends & holidays 2

１． 2 Basic analysis Departure Time Choice Clustering Departure Time Choice Clustering Analysis - 2 Analysis - 1 0.5 0.6 0.5 0.4 0.4 0.3 0.3 0.2 0.2 0.1 0.1 0 0 1 2 3 4 5 1 2 3 4 5 weekdays weekends and holidays male female 3

2 ．１ Model Structure Departure Time and Travel Mode Period 1 Period k μ 1 μ k Walk Bicycle Walk Bicycle Car Bus Rail Car Bus Rail …… Period 1 Period 1 Period 1 Period 1 Period 1 Period k Period k Period k Period k Period k NL model structure with travel mode located in lower layer 4

2 ． 2 Utility Function （ NL model ） , , , x x x x weekend weekday female shopping - Dummy variables x   - OD distance(km) ODdis tan ce V c t * x walking walking traveltime Mode={walking, bicycle, rail, bus, car}   V c t * x bicycle bicycle traveltime  V t * x rail traveltime   V c t * x bus bus traveltime   V c t * x 5 car car traveltime

2 ． 3 Model Estimation Variable Estimation Variable Estimation μ 1 d(km) 0.019*** 0.2064*** μ 2 wn1 -1.2347*** 0.5326 μ 3 1.2657*** 0( base ) wn2 Departure time choice μ 4 g1 -1.1341*** 0.5134 μ 5 g2 0.7908*** 0.3564* sp 1.5602*** Sample size 1522 t(h) -0.8748*** Rho-squared 0.221 Ajusted Rho- c walking 0.2774*** 0.223 Travel mode squared c bicycle -0.0832 choice Significance level *** - 99.9% ** - 99% c bus -1.8644*** * - 95% . - 90% 0.1538*** c car 6

3 ．１ Conclusions Departure time and travel mode • The substitutability between different travel modes exists when decision makers decide to depart in period 1 and 5 as the estimation results of μ 1 and μ 5 are significant . • The sign of OD distance in the utility functions of period 2 and 4 are positive. • Women have more obvious preferences when making departure time choices based on the parameter signs of female dummies. 7

3. 2 Policy Implications Shopping behavior in evening peak hours (based on the findings of the relationship between departure time choice and travel mode choice, and between departure time choice and travel purpose) • Discount in restaurant and supermarket could start from 8 pm. This may make some commuters to change their schedules, such as delaying their leaving time from work. Women-only passenger car (based on the finding of departure time choice and gender) • It is possible to adjust the number of WOPC according to women preferences. 8

Thank you ！ 9