Highway Research Record

V. TRAFFIC AND TRANSPORTATION
1.PLANNING AND MANAGEMENT
NEW PROJECTS
Modeling Generalized Cost of Travel and Studying the Sensitivity of its Components using Stated Preference (SP) and Revealed Preference (RP) Data

Duration:September 2002 March 2005
Indian Institute of Technology, Kharagpur (R)

Scope and objectives

(i) Modeling direct cost of travel for PV (Private Vehicle), IPT (Intermediate Public Transport) and PT (Bus Transport) users
(ii) Modeling value of in-vehicle travel time for PV, IPT and PT users
(iii) Modeling value of out-of-vehicle time (i.e. waiting time) for PT users
(iv) Modeling value of discomfort for PT users
(v) Modeling value of aesthetics for PT users
(vi) Modeling value of aesthetics for IPT users
(vii) Studying the sensitivities for values of in-vehicle travel time (for PV, IPT and PT), out-of-vehicle time (for PT), discomfort       (for PT) and aesthetics (for IPT and PT) with respect to income level and trip purpose.

Two case studies are considered in the study, One in Rural (or Non-urban) scenario and the other in urban scenario. In rural scenario, a bus route connecting a district head quarter (Midnapur) and a tourist place (Digha) in West Bengal, is considered, whereas, trips made by commuters in Kolkata are considered for urban scenario. In urban scenario three different modes of transport i.e. bus, private vehicle and taxi are considered. For both urban and non-urban scenarios, the scope of works include identification of travel attributes and their levels, design of questionnaire, collection of primary and secondary data, analysis of data and development of utility models, estimation of perceived values associated with different attributes of travel, development of generalized cost models, and studying the sensitivity of generalized cost parameters with respect to trip purpose and socioeconomic attributes.

Methodology
Methodology involves preparation of Questionnaire for collecting RP and SP data after identifying the attributes of travel of respective mode under consideration, and their levels, carrying out personal interviews for collecting trip makers responses, development of utility equations using random utility theory based MNL models, formulating Generalised Cost Model from Utility equations, and studying the sensitivity of Generalised Cost Model parameters with various cross classified groups based on Socioeconomic and Trip characteristics. The methodology used for the work is summarized below:

Design of Experiments:
• Selection of survey method/Type of data: SP and RP data
• Preference elicitation method: Choice Based
• Preparation of Choice sets: Fractional Factorial technique
Source of Data:Primary (Personal Interview) and Secondary information
Type of Model:Multinomial Logit (MNL) Model
Estimation of Values for Travel Attributes:
U = α (In Vehicle Travel Time) + β(Cost of Travel) + γ(Discomfort Level)
Value of travel time (a1) = α/β (Rs. per minute)
Value of discomfort (a2) = γ/β (Rs. per unit change of DL)
Modeling Generalised Cost of Travel: GC =a1 IVTT + a2 DL + Fare + δ

Conclusions

The stated preference technique is found to be useful for developing utility models and estimating perceived values associated with different attributes of travel. Both qualitative and quantitative attributes are considered. Perceived values associated with different attributes are estimated separately for rural bus users, urban bus users, taxi users and private vehicle users. The estimated perceived values associated with different attributes of travel are also found to vary logically with mode of travel, trip purpose, income level etc.

Significance/Utilisation Potential

The knowledge of estimated perceived values associated with different travel attributes, and their variations with trip purpose, mode, income etc. can be used for judicious improvement planning of transportation system and also for formulating suitable policy measures with due regard to users willingness- to-pay for an improved service/system.

Limitations
Estimated values associated with different travel attributes, and variations with trip purpose, mode, income etc., are case specific outputs from the present work. The outputs/results are normally not geographically transferable. However, the approach and methodology indicated in the present work can be applied, with suitable modifications, for other urban and rural areas. Further works may be taken up for formulation of transportation improvement proposals in the study area, based on the estimated perceived values associated with different attributes of travel.

Reports/ Publications
Final report on the project is under preparation and will be submitted soon to DST, New Delhi.

Publications
1. B. Maitra and C. V. Phani Kumar (2004), “Stated Choice Approach for Modeling Generalised Cost of Travel for Rural Bus    Service” Proceedings of International Conference TRANSPO 04, IIT Madras, Chennai, 2004.
2. V. Phani Kumar, D. Basu and B. Maitra (2004), “Modeling Generalized Cost of Travel for Rural Bus Users: A Case Study”,    Journal of Public Transportation, Vol. 7, No. 2, 2004, pp 59-72.
3. C. V. Phani kumar and B. Maitra (2004), “Non-linearity in the Perceived Values of Travel Attributes - An Experience in Rural     India”, Proceedings of International workshop TPMDC 04, IIT Bombay, Mumbai, 2004.
4. C. V. Phani Kumar and B. Maitra (2005), “Modeling Travel Behaviour of Rural Bus Users for Various Trip Purposes”,     Proceedings of International Conference START 2005, IIT Kharagpur, Kharagpur, 2005.
5. C. V. Phani Kumar and B. Maitra (2005), “Conjoint Analysis for Estimating Trip Makers’ Perceived Values of Travel     Attributes”, accepted for publication in Indian Roads Congress.
6. C. V. Phani Kumar and B. Maitra (2005), “Percieved Values of Travel Attributes in Rural Regions”, Proceedings of National     Conference ART 05, NIT Rourkela, Rourkela, 2005.