Karthik K Srinivasan

Mixed traffic in the cities of many developing countries is characterized by a lack of lane discipline, varying compositions of constituent vehicle types, and significant intraclass variability in static and dynamic characteristics. However, the influence of these factors on traffic flow parameters is not well understood. This study addressed the influence of lane discipline, intraclass variability, and composition on traffic flow characteristics under heterogeneous traffic conditions in Chennai, India. A microscopic traffic simulation model was calibrated and validated with field data from a four-lane divided urban arterial road in Chennai. The preliminary analysis indicated that factors such as composition, intraclass variability, and lane discipline had a statistically significant effect on stream speed. Speed-flow and speed-density relationships were developed on the basis of simulation results. These results showed a clear influence of lack of lane discipline, variability, and composition on stream speed. The influence varied depending on volume level and type of subject vehicle. The effect of composition on capacity was quantified. When two-wheelers had a predominant share, they enjoyed better performance in the absence of lane discipline. However, when cars and heavy vehicles had a significant presence, the impact of the lack of lane discipline was much smaller. The simulation model was applied to evaluate a range of traffic control measures based on vehicle type and lane. The results showed the promise of some measures based on vehicle class, namely, the exclusion of autorickshaws or autorickshaws and heavy vehicles. The findings have interesting implications for efficiency, user experience, and equity in mixed traffic.

Conventional and contemporary models of travel choice make the restrictive assumption of homogeneity in decision rules. Recent literature has shown empirical evidence for potential heterogeneity in decision rules with regard to utility maximization and regret minimization. Notwithstanding these advances in modeling decision rules, behavioral understanding in the differences in these alternative decision rule segments has not been sufficiently understood. Moreover, the factors which influence the choice of these decision rules have not received significant attention. This study proposes a framework which considers decision makers to have both utility maximizing and regret minimizing tendencies. The variation in these tendencies across decision makers renders the framework heterogeneous. A heterogeneous decision rule model is developed assuming the decision rule adopted to be a latent construct. The study characterizes the regret minimizing and utility maximizing segments based on average values of the segmental attributes. The empirical findings show evidence to confirm that utility maximizers tend to be predominantly captive to personal vehicle usage while regret minimizers might be non-captive to any particular mode. The nature and extent of influence of factors affecting the choice of decision rule is also examined.

This paper focuses on systematic behavioral heterogeneity in mode choice decisions of working population in Chennai city. The sources of heterogeneity that are investigated include (a) variation in choice set across decision makers, (b) differences in innate preferences and responsiveness to explanatory factors due to variations in their degree of captivity to non-personal modes and number of co-passengers during work commute. Existing studies that account for such systematic differences among decision-maker in a developing country context capture only the preference heterogeneity, impose unrealistic restrictions on the utility of non-personal modes (say for e.g. for joint (copassengers > 0) trips) and specify choice sets to be fixed and identical across decision-makers. Addressing these issues, the main objectives of the study were to capture the effect of factors influencing perceived availability of alternatives in the choice set; behavioral comparison of alternate choice set representations namely: fixed choice set, choice set with explicit specification of unavailability of alternatives, partial and probabilistic choice set; investigate potential heterogeneity in innate preference to alternatives and responsiveness to explanatory factors due to difference in captivity status and the presence of co-passengers; and examine whether and how these heterogeneity effects differ based on alternative choice set representations. The effect of factors influencing the perceived availability of alternatives is captured using a set of binary logit models. Unlike many studies, where captivity is represented by a binary variable, behavioral differences across three levels of captivity (captive by vehicle ownership, captive by driving knowledge, and semi-captive) are investigated. The results from the empirical analysis show that disregarding preference or response heterogeneity, when present, can lead to poorer goodness-of-fit in models. It also shows that partial and probabilistic choice set representation is more behaviorally consistent than explicit specification of alternative unavailability, which in turn is better than assuming choice set to be identical across decision-makers.

In developing countries such as India, the traffic on roads is highly heterogeneous in nature, with vehicles of widely varying static and dynamic characteristics. This type of traffic is characterized by lack of lane discipline and free movement over the entire width of the roadway (laneless movement) based on availability of space. In such a regime of traffic flow, the phenomenon of merging of vehicles at junctions of two roads is complex and warrants further study. Merging maneuvers at T-junctions under congested traffic conditions were studied microscopically through video recordings. On the basis of these observations, models for normal and forced merging have been developed. Video data of merging maneuvers were employed to calibrate and validate the models. Such models can be used to simulate highly congested traffic flow in a realistic manner under heterogeneous traffic conditions. They can also give insight on devising better traffic control measures at such junctions.

There is a growing interest in modeling travel time uncertainty in transportation networks in addition to optimizing the reliability of travel times at the path and network level. This paper focuses on the analysis and optimization of travel time (including stopped delays) Reliability on the Urban Road Network in Chennai. Specifically, two objectives are investigated. The first objective involves the quantification of travel time reliability at the link and path level. In particular, the distribution of link travel times is quantified for the Chennai Urban road network using empirical data. The results indicate that the shifted log-normal distribution (SLN) reasonably represents link travel time for all facility types and relevant facility wise distribution parameters are estimated. Further, the resulting path travel time distribution is approximated by a SLN distribution, which is computationally less expensive than traditional Monte-Carlo estimation techniques with an acceptable compromise on accuracy. The second objective addresses the optimal reliability path problem on a network with SLN link travel times with general correlation structure. For this problem, it is shown that the sub-path optimality property of shortest path problems does not hold making traditional label-setting/label correcting algorithms inapplicable. Consequently, a sufficient optimality condition based on reliability bounds is established and a new network optimization algorithm is proposed and proof of correctness is presented. The convergence rate of the algorithm was shown to increase at every iteration under some mild conditions. The computational performance of the proposed algorithm is investigated using synthetic and real-world networks and found to be reasonably accurate. © 2013 Elsevier Ltd.

Transportation networks are subject to significant travel time uncertainty as a result of traveler behavior, recurring congestion, capacity variability (construction zones, traffic incidents), variation in demand, and so on. Therefore, interest is growing in modeling and optimizing travel time reliability in such networks. This paper proposes an efficient algorithm to compute the path of maximum travel time reliability on a network with normally distributed and correlated link travel times. For this optimal reliability path (ORP) problem, it is shown that the subpath optimality condition for the deterministic shortest path problem does not hold, and consequently, a new bounds-based optimality criterion is proposed using the K shortest expected time paths and the minimum path variance on the network. An algorithm is developed to solve the ORP problem on the basis of the proposed optimality criterion and an efficient path generation procedure. Computational experiments on various test networks show the proposed algorithm to be efficient, requiring limited path enumeration. With as few as five shortest paths and 50 Monte Carlo draws, the proposed algorithm is able to find the most reliable path for realistic network sizes. Empirical investigations highlight the unreliability of the least expected time path and suboptimality of the independence assumption. The study also underscores the role of risk attitudes (reflected by reliability threshold) on the benefits of the ORP. The algorithm and empirical results have important applications for developing reliability-based routing applications for congestion mitigation and intelligent transportation systems.

This study addresses the problem of determining the most reliable time-adaptive strategy on a stochastic and time-dependent transportation network. The reliability is measured as a conic combination of the mean and standard-deviation of travel time and is termed robust-cost. The stochastic time-dependent network is represented as a directed acyclic hypergraph, where the time-adaptive strategies correspond to the hyperpaths. This representation transforms the problem to that of determining the hyperpath with the least robust-cost on the constructed hypergraph. The minimum robust-cost strategy problem is difficult to solve because of the non-linear objective function. Consequently, the solution procedures commonly adopted in the literature —that are based on substrategy optimality and substrategy non-dominance —are not applicable to this problem. In this light, we propose a novel bounds-based iterative algorithm that determines the minimum robust-cost strategy on the stochastic and time-dependent networks. This algorithm needs to determine the least and K-best strategies in the second moment of travel time, for which an efficient procedure is also proposed. The algorithm is shown to be exact and exhibit parameterically polynomial behavior; computational tests were performed to demonstrate its efficiency. Further, tests showed that the minimum robust-cost strategy compromises little in terms of the mean travel time (0.2%–2.9%) —compared to least expected travel time strategy— with significant reduction in travel time variability (6.2%–29.8%).

The deterministic traffic assignment problem based on Wardrop's first criterion of traffic network utilization has been widely studied in the literature. However, the assumption of deterministic travel times in these models is restrictive, given the large degree of uncertainty prevalent in urban transportation networks. In this context, this paper proposes a robust traffic assignment model that generalizes Wardrop's principle of traffic network equilibrium to networks with stochastic and correlated link travel times and incorporates the aversion of commuters to unreliable routes. The user response to travel time uncertainty is modeled using the robust cost (RC) measure (defined as a weighted combination of the mean and standard deviation of path travel time) and the corresponding robust user equilibrium (UE) conditions are defined. The robust traffic assignment problem (RTAP) is subsequently formulated as a Variational Inequality problem. To solve the RTAP, a Gradient Projection algorithm is proposed, which involves solving a series of minimum RC path sub-problems that are theoretically and practically harder than deterministic shortest path problems. In addition, an origin-based heuristic is proposed to enhance computational performance on large networks. Numerical experiments examine the computational performance and convergence characteristics of the exact algorithm and establish the accuracy and efficiency of the origin-based heuristic on various real-world networks. Finally, the proposed RTA model is applied to the Chennai road network using empirical data, and its benefits as a normative benchmark are quantified through comparisons against the standard UE and System Optimum (SO) models.

Link performance functions commonly used for traffic assignment are often based on Volume Delay Functions (VDF) developed for homogeneous traffic. However, VDFs relating stream speed to the volume of traffic based on homogeneous lane-based traffic are not adequate for traffic assignment in developing countries due to the heterogeneous nature of traffic that is characterized by a mix of a wide range of vehicle classes with significant differences in static and dynamic characteristics and an imperfect lane discipline. Unfortunately, the use of VDFs similar to those for homogeneous traffic flow situations imposes strong restrictions considering two respects: 1) travel times at path and link levels can be obtained for an aggregated stream but not for individual vehicle types; 2) the effect of varying composition and asymmetric interactions is captured only to a limited extent by converting all vehicles into equivalent Passenger Car Unit (PCU). Hence, this paper proposes the development of VDFs specific to different classes of heterogeneous traffic, as it is more realistic in traffic assignment than the use of the same VDF for all classes of vehicles in a link. This study is aimed at developing models to determine the speed of each vehicle class as a function of flow and composition for six lane roads with heterogeneous traffic based on data obtained from Chennai city, India. Heterogeneity in this study mainly refers to differences in vehicle types (two-wheeler, car, bus, etc.) participating in mixed traffic. To develop multiple user class VDFs, the speed and flow of each vehicle class for a wide range of traffic flow conditions need to be recorded. As this is not possible using field measurements, an established micro-simulation model (HETEROSIM) is used for determining speeds for each vehicle type by systematically varying the volume and composition levels over a range of values that represent relevant and practical traffic conditions observed in six lane divided roads in Chennai city. The proposed delay functions are different from standard single user class VDFs in three key respects: first, they enable more realistic behaviour by modelling differences in class wise speeds at a given volume and composition level; second, they allow for capturing asymmetric interactions of different vehicle types on an average speed of a given vehicle class. Finally, speed-flow relationships for each class are also allowed to vary across volume levels which enable the representation of differential interactions at different levels of congestion in mixed traffic. The need for homogenizing the volumes in terms of a single class is obviated. The models significantly outperformed single class VDFs in both calibration and validation datasets. Further, the proposed models are used for analyzing heterogeneous traffic characteristics. Empirical evidence of asymmetric interactions and the impact of composition on classwise performance are also found and quantified. Finally, two applications of the proposed models are demonstrated for the level of service analysis of different classes and impact analysis of excluding some classes. The proposed models may have applications such as determining class wise road user costs and performance measures (e.g. emissions) that depend on class-specific speeds. © 2012 Copyright Vilnius Gediminas Technical University (VGTU) Press Technika.

This study addresses various formulations of the optimal reliability path problem on stochastic networks. Robust-Cost is adopted as the measure of reliability, which is defined as a weighted combination of the mean and standard deviation of travel time. The principal problem solved is the Minimum Robust-Cost Path (MRCP) problem in the presence of Correlated link travel times. It is shown that the subpath optimality and subpath non-dominance principles, which are conventionally adopted in the literature, cannot be used to solve this problem. In this light, this study proposes an algorithm based on the subpath pruning approach, which eliminates nonoptimal subpaths by using a pruning criterion. We construct the novel pruning criterion, which depends on only two independent objectives, by transforming the network and using efficient shortest path algorithms. The correctness of the algorithm is established, and its good practical performance is demonstrated on real-world networks. Furthermore, the pruning procedure is generalized with suitable modifications to solve three related problems: (i) the MRCP problem with independent link travel times, (ii) the K-best Robust-Cost Paths problem, and (iii) the MRCP problem with stochastic nondominance constraints. These extensions demonstrate the potential wider applications of the proposed solution approach.