Vinod Kumar Katiyar

A new multi-class continuum model for traffic flow

In this paper, we propose a new continuum model and study some qualitative properties. The new model contains an additional speed gradient term (anisotropic term) in comparison to Bergs model (Berg et al., 2000, Physical Review E, 61, 1056–1066). This anisotropic term guarantees the property that the characteristic speeds can be made less than or equal to the macroscopic flow speed. We extend this model for multi-class traffic flow with heterogeneous drivers. Each user class is characterized by their choice of speeds in a traffic stream. The choice of the speed of a particular user class at any location on the highway is affected by the presence of all user class on that location. Numerical simulations show that the model is able to explain some of the observed traffic phenomena such as platoon dispersion that challenge old homogeneous models presented in the literature.

Short term traffic flow prediction in heterogeneous condition using artificial neural network

Traffic congestion is one of the main problems related to transportation in developed as well as developing countries. Traffic control systems are based on the idea to avoid traffic instabilities and to homogenize traffic flow in such a way that risk of accidents is minimized and traffic flow is maximized. There is a need to predict traffic flow data for advanced traffic management and traffic information systems, which aim to influence traveller behaviour, reducing traffic congestion and improving mobility. This study applies Artificial Neural Network for short term prediction of traffic volume using past traffic data. Besides traffic volume, speed and density, the model incorporates both time and the day of the week as input variables. Model has been validated using actual rural highway traffic flow data collected through field studies. Artificial Neural Network has produced good results in this study even though speeds of each category of vehicles were considered separately as input variables.

A New Real Coded Genetic Algorithm Operator: Log Logistic Mutation

In this study, a new mutation operator for real coded genetic algorithms called the Log Logistic Mutation (LLM) is proposed. The performance of LLM is compared with existing real coded mutation operator namely Power Mutation (PM). LLM is used in conjunction with a well known crossover operator; Laplace Crossover (LX), to obtain a new generational real coded genetic algorithm called LX-LLM. LX-LLM is compared with the existing LX-PM. The performance of both the genetic algorithms is compared on the basis of success rate, average function evaluation, average error and computational time, and the supremacy of the proposed mutation operator is established.

Fuzzy Logic Model for the Prediction of Traffic Volume in Week Days

This paper presents a model for traffic volume prediction which can be effectively used for transportation planning, management and security assessment at any time. Fuzzy logic is applied in order to realize effective and efficient traffic prediction. In this paper, ‘day’ of a week and ‘time’ of a day are taken as inputs for proposed model and the output will be the predicted the traffic volume. The ‘time’ is divided into nine triangular membership functions. The second input ‘day’ is divided into five triangular membership functions and the output forecasted traffic volume has been divided into eight triangular membership functions. The predicted traffic volume when compared with actual traffic volume has MAPE within acceptable level of error. Prediction results show that the proposed fuzzy logic system produces more accurate and stable traffic volume predictions.

Two-Lane Traffic Flow Simulation Model via Cellular Automaton

Road traffic microsimulations based on the individual motion of all the involved vehicles are now recognized as an important tool to describe, understand, and manage road traffic. Cellular automata (CA) are very efficient way to implement vehicle motion. CA is a methodology that uses a discrete space to represent the state of each element of a domain, and this state can be changed according to a transition rule. The well-known cellular automaton Nasch model with modified cell size and variable acceleration rate is extended to two-lane cellular automaton model for traffic flow. A set of state rules is applied to provide lane-changing maneuvers. S-t-s rule given in the BJH model which describes the behavior of jammed vehicle is implemented in the present model and effect of variability in traffic flow on lane-changing behavior is studied. Flow rate between the single-lane road and two-lane road where vehicles change the lane in order to avoid the collision is also compared under the influence of s-t-s rule and braking rule. Using results of numerical simulations, we analyzed the fundamental diagram of traffic flow and show that s-t-s probability has more effect than braking probability on lane-changing maneuver.

Global Optimization of Lennard-Jones Potential Using Newly Developed Real Coded Genetic Algorithms

In this paper recently developed real coded genetic algorithms are applied to the challenging problem of finding the minimum energy configuration of a cluster of identical atoms interacting through the Lennard-Jones potential. Finding the global optimum of this function is very difficult because it has a large number of local minima, which grows exponentially with molecule size. Computational results for atomic cluster containing up to 15 atoms are obtained and presented. The obtained results show the remarkable performance of newly developed real coded genetic algorithms as compared to to the earlier published results.

Multi Objective Extraction Optimization of Bioactive Compounds from Gardenia Using Real Coded Genetic Algorithm

In this paper the problem of extraction optimization of bioactive compounds from Gardenia (Gardenia jasminoides Ellis) fruits is modeled into a non-linear multi-objective optimization problem, in which there are three objectives viz. yield of crocin, yield of geniposide and yield of total phenolic compound and three decision variables namely ethanol concentration, temperature and time.

Traffic Accident Prediction Model Using Support Vector Machines with Gaussian Kernel

Road traffic accident prediction models play a critical role to the improvement of traffic safety planning. The focus of this study is to extract key factors from the collected data sets which are responsible for majority of accidents. In this paper urban traffic accident analysis has been done using support vector machines (SVM) with Gaussian kernel. Multilayer perceptron (MLP) and SVM models were trained, tested, and compared using collected data. The results of the study reveal that proposed model has significantly higher predication accuracy as compared with traditional MLP approach. There is a good relationship between the simulated and the experimental values. Simulations were carried out using LIBSVM (library for support vector machines) integrated with octave.

Mathematical modeling to study influence of porosity on apple and potato during dehydration

Several structural and physical changes in foodstuffs are the consequence of water removal during the drying process. Porosity (volume fraction of pores) is one of the key parameter that affects the quality and other properties of foods (such as apple and potato). To understand the effect of dehydration in apple and potato, in the present study an arbitrary small cubic volume element is considered which contains pores (intracellular spaces) distributed in it. Further, it is assumed that each pore in the cubic volume element is spherical. A mathematical relation is developed between porosity (volume fraction of pores) and pressure generated (due to contraction of cells during water removal) in outward direction on the surface of spherical elements containing pore. The developed relation is satisfactory in respect of experimental observations given in the literature. For the given pressure range, acquired porosity range is 0.1 to 0.92 for apple and 0.03 to 0.89 for potato which is matched with the existing experimental values. The results showed that the porosity is increasing with the increasing values of pressure, as expected, during moisture removal. Further, it is observed that the current porosity is depended on the initial porosity for both apple and potato.

Artificial neural network modeling for road traffic noise prediction

Several attempts have been made by the researchers to predict and model urban road traffic noise mathematically and statistically. There has been a lot of interest in the new techniques for analyzing data. Neural networks offer a new strategy with enormous potential for many tasks in the domain of geospatial planning. ANN technique for modeling provides smaller errors in comparison to other classical methods. Neural networks have been applied to many interesting problems in various areas including road traffic noise prediction. In the present study an attempt has been made to explore the application of neural networks to road traffic noise prediction in Lucknow city, capital of Uttar Pradesh, India. Traffic volume, speed and noise level data were collected at ten selected locations. For development of model, classified traffic volume (Car/Jeep/Van, Scooter/ Motorcycle, LCV/ Minibus, Bus, Truck and 3-Wheeler), traffic speed on both sides of the road were taken as input data. Output was estimated as Leq. Performance of the model was tested by root mean square error (RMSE), mean absolute error (MAE) and coefficient of correlation (R). It was observed that there is no significant difference between observed and predicted noise levels in the present case, indicating the accuracy of model.