356 Sekhar C.R. et al. Multimodal Choice Modeling Using Random Forest Decision Trees UDC: 656.073:005.311.6 DOI: http://dx.doi.org/10.7708/ijtte.2016.6(3).10 MULTIMODAL CHOICE MODELING USING RANDOM FOREST DECISION TREES Ch.Ravi Sekhar 1 , Minal 2 , Errampalli Madhu 3 1, 2, 3 Transportation Planning Division, Central Road Research Institute, New Delhi 10025, India Received 11 May 2016; accepted 23 August 2016 Abstract: Mode choice analysis forms an integral part of transportation planning process as it gives a complete insight to the mode choice preferences of the commuters and is also used as an instrument for evaluation of introduction of new transport systems. Mode choice analysis involves the procedure to study the factors in decision making process of the commuter while choosing the mode that renders highest utility to them. This study aims at modeling the mode choice behaviour of commuters in Delhi by considering Random Forest (RF) Decision Tree (DT) method. The RF model is one of the most efficient DT methods for solving classification problems. For the purpose of model development, about 5000 stratified household samples were collected in Delhi through household interview survey. A comparative evaluation has been carried out between traditional Multinomial Logit (MNL) model and DT model to demonstrate the suitableness of RF models in mode choice modeling. From the result, it was observed that model developed by Random Forest based DT model is the superior one with higher prediction accuracy (98.96%) than the Logit model prediction accuracy (77.31%). Keywords: mode choice analysis, household survey, MNL model, random forest decision tree. 1 Corresponding author: ravisekhar.crri@nic.in 1. Introduction Dealing with the present bottlenecks as well as creating long lasting and sustainable transport systems has been the greatest challenge of urban transportation planning. Calibrating the present need and forecasting the future demand is the underlying agenda of travel demand forecasting. Mode choice forms an integral part of this process as it gives a complete insight to the mode choice preferences of the commuters validating the introduction of new transport systems to existing ones. Mode choice analysis is the procedure to study the factors and decision making process of the trip maker and to be able to model it. Trip makers seem to choose the mode that renders highest utility to them. Multinomial Logit (MNL) is one of the classic models used in the development of mode choice models (Ben-Akiva and Lerman, 1985). It is a method of logistic regression of classification. Recently, methods of “ensemble learning” are being used. Since the inception of machine learning and use of related algorithm in transportation problems, it finds a prominent place in contemporary modeling. In these methods different classifiers are generated and a final output is obtained by aggregating their results. Two such well-known techniques are that of Boosting and Bagging of classification trees.