West Virginia University
Forecasting Automobile Demand Via Artificial Neural Networks & Neuro-Fuzzy Systems
Abstract
dc:description.abstractThe objective of this research is to obtain an accurate forecasting model for the demand for automobiles in Iran's domestic market. The model is constructed using production data for vehicles manufactured from 2006 to 2016, by Iranian car makers. The increasing demand for transportation and automobiles in Iran necessitated an accurate forecasting model for car manufacturing companies in Iran so that future demand is met. Demand is deduced as a function of the historical data. The monthly gold, rubber, and iron ore prices along with the monthly commodity metals price index and the Stock index of Iran are Artificial neural network (ANN) and artificial neuro-fuzzy system (ANFIS) have been utilized in many fields such as energy consumption and load forecasting fields. The performances of the methodologies are investigated towards obtaining the most accurate forecasting model in terms of the forecast Mean Absolute Percentage Error (MAPE). It was concluded that the feedforward multi-layer perceptron network with back-propagation and the Levenberg-Marquardt learning algorithm provides forecasts with the lowest MAPE (5.85%) among the other models. Further development of the ANN network based on more data is recommended to enhance the model and obtain more accurate networks and subsequently improved forecasts.
Degree
thesis:*- Name thesis:degree_name
- MS
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- Industrial and Managements Systems Engineering
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Niaki, Armin
- Contributors dc:contributor
-
- Majid Jaridi
- Kenneth Currie
- Feng Yang
Identifiers
dc:identifier.*- Identifier
- https://researchrepository.wvu.edu/etd/7336