Tom V. Mathew

Pareto Optimal Multiobjective Optimization for Robust Transportation Network Design Problem

A study was done to formulate and solve the multiobjective network design problem with uncertain demand. Various samples of demand are realized for optimal improvements in the network while the objectives of the expected total system travel time and the higher moment for total system travel time are minimized. A formulation is proposed for multi-objective robust network design, and a solution methodology is developed on the basis of a revised fast and elitist nondominated sorting genetic algorithm. The developed methodology has been tested on the Nguyen-Dupuis network, and various Pareto optimal solutions are compared with earlier work on the single-objective robust network design problem. A real medium-size network was solved to prove efficacy of the model. The results show better solutions for the multiobjective robust network design problem with relatively less computational effort.

Multiobjective network design for emission and travel-time trade-off for a sustainable large urban transportation network

Existing optimal road-network capacity-expansion models are based on minimizing travel time and rarely consider environmental factors such as vehicular emissions. In this study we attempt to solve such a transportation network design problem when the planner is environment conscious and thereby tries to minimize health-damage cost due to vehicular emissions along with total system travel time while performing optimal capacity expansion. This problem can be formulated as a multiobjective optimization model which minimizes emissions in addition to travel time, and under budget constraints. A prerequisite for this model is an accurate estimation of vehicle emissions due to changes in link capacities. Since the current practice of estimation of vehicular emissions by aggregate emission factors does not account for the improved speeds resulting from capacity improvements, speed-dependent emission functions for various transport modes and pollutants are used in this study. These functions help in calculating emission factors for use in the proposed model. The model uses a nondominated sorting genetic algorithm as the optimization tool to solve the network design problem. The model is tested on a small hypothetical network and solved for a real large-sized network in India taking into account three pollutants and five transport modes. The Pareto-optimal solutions generated can act as trade-offs between total emissions and total system travel time to account for the planners desired objectives. Also, reduction in travel time as well as in emissions supports the present model compared with the single-objective model.

Intelligent Agent Optimization of Urban Bus Transit System Design

The transit route network design (TRND) problem seeks a set of bus routes and schedules that is optimal in the sense that it maximizes the utility of an urban bus system for passengers while minimizing operator cost. Because of the computational intractability of the problem, finding an optimal solution for most systems is not possible. Instead, a wide variety of heuristic and meta-heuristic approaches have been applied to the problem to attempt to find near-optimal solutions. This paper presents an optimization system that synthesizes aspects of previous approaches into a scalable, flexible, intelligent agent architecture. This architecture has successfully been applied to other transportation and logistics problems in both research studies and commercial applications. This study shows that this intelligent agent system outperforms previous solutions for both a benchmark Swiss bus network system and the very large bus system in Delhi, India. Moreover, the system produces in a single run a set of Pareto equivalent solutions that allow a transit operator to evaluate the trade-offs between operator costs and passenger costs.

Passenger car units and saturation flow models for highly heterogeneous traffic at urban signalised intersections

Signalisation is a traffic control strategy to ease the competition by providing right-of-way in a cyclic manner to conflicting traffic at intersections. Saturation flow is a major component in the design of signals, and is influenced by a variety of factors like vehicle composition, intersection geometry and drivers behaviour. The highway capacity manual (HCM) has recommended a saturation flow model primarily for homogeneous conditions, with limited ability to address heterogeneity. But the traffic in many parts of the world is highly heterogeneous and hence, defining a unified saturation flow concept is a challenging task. The variability in vehicle types necessitates the use of passenger car units (PCUs). This article proposes a methodology to develop a saturation flow model based on dynamic PCUs by a microscopic analysis. Field data from intersections of three Indian cities – Jaipur, Bangalore and Trivandrum – is used for the study. PCU values are derived from the field data, by minimising the difference between the observed and the ideal flow profiles. A new saturation flow model is then developed using the regression method. The model is validated with the saturation flows collected from different locations. In addition, the flow predicted by the saturation flow model is used to estimate delay using the HCM and Websters models, and these delays are compared to the observed delays. The proposed model resulted in lower error compared to the conventional flow estimation techniques.

Vehicle-Type Dependent Car-Following Model for Heterogeneous Traffic Conditions

Car-following behavior forms the kernel of traffic microsimulation models and is extensively studied for similar vehicle types. However, in heterogeneous traffic having a diverse mix of vehicles, following behavior also depends on the type of both the leader and following vehicles. This paper is an attempt to modify the widely used Gipps’s car-following model to incorporate vehicle-type dependent parameters. Performance of the model is studied at microscopic and macroscopic levels using data collected from both homogeneous and heterogeneous traffic conditions. The results indicate that the proposed modifications enhance the prediction of follower behavior and suggest the need of incorporating vehicle-type combination specific parameters into traffic simulation models.

Strip-Based Approach for the Simulation of Mixed Traffic Conditions

AbstractVehicles in homogeneous traffic follow lane-based movement and can be conveniently modeled using car-following and lane-changing models. The former deals with longitudinal movement behavior, while the latter deals with lateral movement behavior. However, typical heterogeneous traffic is characterized by the presence of multiple vehicle types and non-lane-based movement. Because of the off-centered positions of the vehicles, the following driver is not necessarily influenced by a single leader. Additionally, the following behavior of the subject vehicle depends on the type of the front vehicle. Unlike discrete lane changes in the case of lane-based traffic, heterogeneous traffic streams require modeling of continuous lateral movements. Hence, the existing driver behavioral models may not be able to represent the heterogeneous traffic behavior accurately enough. To address these critical issues of heterogeneous traffic, a space discretization–based simulation framework is proposed. The lane is divid...

Approximation Techniques for Transportation Network Design Problem under Demand Uncertainty

Conventional transportation network design problems treat origin-destination (OD) demand as fixed, which may not be true in reality. Some recent studies model fluctuations in OD demand by considering the first and the second moment of the system travel time, resulting in stochastic and robust network design models, respectively. Both of these models need to solve the traffic equilibrium problem for a large number of demand samples and are therefore computationally intensive. In this paper, three efficient solution-approximation approaches are identified for addressing demand uncertainty by solving for a small sample size, reducing the computational effort without much compromise on the solution quality. The application and the performance of these alternative approaches are reported. The results from this study will help in deciding suitable approximation techniques for network design under demand uncertainty.

Continuous Network Design with Emission Pricing as a Bi-Level Optimization Problem

Traffic network design problem attempts to find the optimal network expansion policies under budget constraints. This can be formulated as a bi-level optimization problem” the upper level determines the optimal capacity expansion vector and the lower level determines the link flows subject to user equilibrium conditions. However, in the context of environmental concerns, driver’s route choice includes travel time as well as emission pricing. This study is an attempt to solve network design problems when the user is environment cautious. The problem is formulated as a bi-level continuous network design problem with the upper level problem determines the optimal link capacity expansions subject to user travel behavior. This behavior is represented in the lower level using the classical Wardropian user equilibrium principles. The upper level problem is an example of system optimum assignment and can be solved using any efficient optimization algorithms. Genetic Algorithm is used because of its modeling simplicity. The upper level will give a trail capacity expansion vector and will be translated into new network capacities. This then invokes the lower level with these new link capacities and the output is a vector of link flows which are passed to upper level. The upper level then computes the objective function and GA operators are applied to get a new capacity expansion vector and the process is repeated till convergence. The model is first applied to an example network and the optimum results are shown. Finally the model is applied to a large case study network and the results are presented.

Hybrid Stochastic Cellular Automata-Driver-Vehicle-Object Simulation Model for Heterogeneous Traffic at Urban Signalized Intersections

The traffic flow at intersections is generally chaotic, and signalization is a control measure to reduce this chaos. Heterogeneous traffic at signalized intersections behave much differently from homogeneous traffic. Also, in many countries, nonlane-based traffic prevails; hence, designing control systems for such situations is a challenging task. Traffic simulation helps the analyst to model the behavior of such complex systems. Cellular automata (CA), a recent entrant in traffic flow modeling, represents the traffic flow by means of simple rules, and thus has proved to be a versatile tool in traffic simulation. The present study aims to develop a computationally efficient traffic flow simulation model integrating the concepts of cellular automata and driver-vehicle-objects, thus making a behavioral model of traffic. The model emphasizes the diversity in human behavior, and represents the traffic using the minimal modeling concept of CA. To represent multiple vehicle types, a multicell representation was adopted. Further, to address the issue of nonlane-based movement, new lateral movement rules were proposed. The model incorporated behavior at amber and lateral movements, thus attempting to achieve a near to reality representation of nonlane-based heterogeneous traffic. The model was calibrated and validated using delay data from selected intersections in India. This model was then used to predict saturation flows at signalized intersections. The model performed reasonably well in predicting the delays, but the saturation flow values showed up to 30% variability.

Semiautomated Tool for Extraction of Microlevel Traffic Data from Videographic Survey

Real-world traffic data collection, extraction, and analysis at a microscopic level is necessary for modeling various traffic and transportation phenomena. In developing countries such as India, where the traffic is characterized by different vehicle types and non-lane-based vehicle movement, an effective tool for vehicle data extraction and analysis is necessary. Development of a fully automated software tool for vehicle trajectory extraction is practically impossible because of the weak lane-based movement of vehicles and occlusion. For addressing these issues, a semiautomated tool, a traffic data extractor, was developed to extract the details of vehicle movements with good accuracy (0.1 s). A videographic survey followed by a camera calibration technique was used to obtain the real-world coordinates from two-dimensional image coordinates. The concept of a vanishing point–based camera calibration technique was used in this tool. The steps involved in handling the developed tool and its applications in microscopic traffic data extraction and analysis are presented.