International Conference on Emerging Frontiers in Technology for Rural Area (EFITRA) 2012 Proceedings published in International Journal of Computer Applications® (IJCA) 13 Need for a Comprehensive Traffic Simulation Model in Indian Context M. Sreekumar Post Graduate Student Department of Civil Engineering Indian Institute of Technology Guwahati, Assam Akhilesh Kumar Maurya Assistant Professor Department of Civil Engineering Indian Institute of Technology Guwahati, Assam ABSTRACT The large body of literature in the area of traffic flow theory and most of the existing traffic simulation models cannot be used as such in countries like India. Since the traffic condition in the Indian subcontinent are coarser- with innumerable ill- defined classes of vehicles interacting in an undisciplined manner, any study on this scenario will be meaningful only if it can incorporate both vehicular heterogeneity and no lane discipline conditions. This paper briefly discusses various past simulation simulation models and some recent models which are found more relevant to Indian context. Through the above discussions, the paper highlights the need of a comprehensive model which can simulate Indian traffic conditions in a realistic way. The description of CUTSiM model, which is found closer to the above mentioned approach is also presented. The results obtained, ensure the applicability of CUTSiM in Indian conditions in an appreciable way. General Terms Traffic simulation model, Driver behaviour, Heterogeneous traffic. Keywords Simulation model, Car-following, No-lane discipline, Heterogeneity, Microscopic models, CUTSiM 1. INTRODUCTION In developing countries like India, the rapid growth of cities has resulted an extensive increase in the road network. The numbers of motorized vehicles have increased manifolds, especially during the peak hours. Congestion problems in cities simultaneously alarm both the users as well as the decision makers. Controlling, suggesting remedial measures, making changes to the existing traffic system in urban India is becoming more and more complex day after day. Traffic Engineering is hence coming to the limelight in India to understand and analyze the traffic stream behaviour for an efficient design of traffic facilities. Unfortunately, the large body of literature that exists in the area of traffic flow theory cannot be used as is in countries like India due to the following reasons. Firstly, traffic here includes innumerable classes of vehicles having different operating characteristics. Secondly, drivers in India do not follow lanes i.e., the traffic streams do not have lane discipline. Therefore any study on Indian traffic scenario will be meaningful only if it can incorporate both vehicular heterogeneity and “no lane discipline” conditions. To analyze the traffic stream behaviour which is very complex and stochastic in nature, one need to understand how vehicles interact each other not only along the longitudinal but also along the lateral direction of the lane and also with the existing traffic facilities? So while modeling the heterogeneous traffic conditions prevailing in most of the developing countries, the system have to be viewed microscopically. The complexity of stream behaviour and the difficulty that exists with performing experiments make computer simulation an important analysis tool in traffic engineering. Secondly, through simulation, it is easier to model individual vehicle actions in different situations than to understand the system dynamics directly. By making use of different traffic simulation models, one can simulate large scale real-world situations in great detail. This paper is being written, identifying the necessity of a comprehensive microscopic traffic simulation model which can simulate Indian traffic scenario in a realistic manner. A brief review of various categories of models consisting of longitudinal control models, Lateral control models and combined lateral and longitudinal control models is included in this paper. It is felt that the quality of any model lies in the level of realism present in the traffic stream simulated using that simulation model. A few recent models are concisely explained to address the fact that the assumptions made in them are far from Indian traffic conditions. 2. A PRECIS OF MICROSCOPIC TRAFFIC SIMULATION MODELS A realistic model of driving or driver behaviour must be a comprehensive model which models both lateral control (steering control) and longitudinal control (speed control) under the impact of both roadway and traffic features [1]. According to this aspect of control, microscopic models can be categorized into (i) longitudinal control models, (ii) lateral control models, and (iii) comprehensive models and discussed in present section. 2.1 Longitudinal Control Models Longitudinal control models assume that the driver behaviour is influenced only by leading vehicles traveling in the same path or lane. This assumption is largely true where the road characteristics are reasonably same for long distances, vehicles have well demarcated paths and vehicles do not generally cross these demarcations. The process of achieving the longitudinal control is referred to as car-following in transportation literature. The car-following behaviour is basically, a human perception-reaction process, where driver of the following vehicle (FV) attempts to reach a stable condition (by accelerating or decelerating) and then maintain it while following a leading vehicle (LV). In this process of reaching a stable situation and maintaining it, the driver of following vehicle continuously takes corrective actions by accelerating or decelerating [1]. Research on car-following theory started dates back to the mid 20 th century. From then, a lot of studies have been done in