Application of PSO (particle swarm optimization) and GA (genetic algorithm) techniques on demand estimation of oil in Iran

• Published in: Energy (Oxford) Vol. 35; no. 12; pp. 5223 - 5229
• Main Authors: Assareh, E., Behrang, M.A., Assari, M.R., Ghanbarzadeh, A.
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.12.2010; Elsevier
• Subjects: Applied sciences; Demand; Domestic; Economic data; Energy; Energy economics; Estimates; Exact sciences and technology; Exports; Fossil fuels and derived products; GA (genetic algorithm); General, economic and professional studies; Genetic algorithms; Marketing; Mathematical models; Methodology. Modelling; Oil; Optimization; Projection; PSO (particle swarm optimization); Iran; Asia; Demand; PSO (particle swarm optimization); Projection; Oil; GA (genetic algorithm); Genetic algorithm; Socioeconomic study; Long term; Modeling; Particle swarm optimization; Forecasting; Petroleum
• ISSN: 0360-5442
• DOI: 10.1016/j.energy.2010.07.043

This paper presents application of PSO (Particle Swarm Optimization) and GA (Genetic Algorithm) techniques to estimate oil demand in Iran, based on socio-economic indicators. The models are developed in two forms (exponential and linear) and applied to forecast oil demand in Iran. PSO–DEM and GA–DEM (PSO and GA demand estimation models) are developed to estimate the future oil demand values based on population, GDP (gross domestic product), import and export data. Oil consumption in Iran from 1981 to 2005 is considered as the case of this study. The available data is partly used for finding the optimal, or near optimal values of the weighting parameters (1981–1999) and partly for testing the models (2000–2005). For the best results of GA, the average relative errors on testing data were 2.83% and 1.72% for GA–DEMexponential and GA–DEMlinear, respectively. The corresponding values for PSO were 1.40% and 1.36% for PSO–DEMexponential and PSO–DEMlinear, respectively. Oil demand in Iran is forecasted up to year 2030.