APPLICATION OF SUBTRACTIVE CLUSTERING AND ANFIS TUNING FOR SUBURBAN COMMUTING

Abstract

In today’s world travel time is important in planning and managing urban roads; on the other hand, it is important in commuters' route and number of trips generated from an area. In recent years, machine learning (ML)-based approaches have gained increasing attention in road traffic load handling research. In this thesis, a hybrid learning algorithm referred to as Adaptive Neuro Fuzzy Inference System (ANFIS) is used to obtain the commuting model. ANFIS has emerged as a highly effective mathematical approach for modeling traffic and transportation processes. This thesis outlines the fundamental principles of ANFIS systems and provides a comprehensive analysis of their application in addressing road traffic load handelling challenges. In this thesis a suburban commuting model has been created using ANFIS. An area has been selected for the experiment. Five demographic factors like population, number of houses, vehicle ownership, median household income, and total employment has been taken as input variable and number of automobile trips generated from that area has taken as output variable. This model may help in traffic controlling. Using this model one can predict about the number of trips in a particular area and likewise arrangements can be done. The model is simulated in MATLAB environment. It is concluded that the model using ANFIS is accurate enough and is useful in traffic controlling actions. Keywords: Fuzzy logic, FIS, ANFIS, Traffic load, epoch.