Sen, DorukGunay, M. ErdemTunc, K. M. Murat2024-07-182024-07-1820190360-54421873-6785https://doi.org/10.1016/j.energy.2019.02.130https://hdl.handle.net/11411/7351Natural gas is a foreign-dependent source of energy in many countries and a rapid increase of its consumption is mainly associated with the increase of living standards and needs. In this work, Turkey was taken as a case study with high degree of foreign dependence of energy, and the future natural gas consumption was predicted by several different multiple regression models using socio-economic indicators as the descriptor variables. Among these, gross domestic product and inflation rate were found to be the only significant ones for this prediction. Next, three different projections for the future values of the significant descriptor variables were tested, and the natural gas consumption was predicted to rise gradually in the range 1.3 +/- 0.2 billion m(3) per year reaching to a consumption of 64.0 +/- 3.5 billion m(3) in the year 2025. It was then discussed that this additional natural gas can be compensated by utilizing local lignite sources or by starting a nuclear energy program although these two methods to reduce the future natural gas consumption have some conflictions with the general European energy matrix and environmental politics. Thus, it was concluded that resuming the wind and solar-based electricity generation programs can be considered as a more reasonable option. (C) 2019 Elsevier Ltd. All rights reserved.eninfo:eu-repo/semantics/closedAccessNatural Gas ConsumptionPredictionTurkish ScenarioGross Domestic ProductMultiple RegressionLigniteNuclear PowerElectricity DemandEnergy-ConsumptionStatistical-ModelNeural-NetworksNuclear-PowerPredictionCombinationSupportSystemArticle2-s2.0-8506221442710.1016/j.energy.2019.02.1301118Q11106173Q1WOS:000464301000088