In this paper we propose a new version of the Ant Colony Optimization (ACO) to predict the natural gas demand problem. This heuristic approach will estimate the Turkey's natural gas demand based on economic indicators. The well-known ACO is multi-agent systems in which the behavior of each ant is inspired by the foraging behavior of real ants to solve optimization problem. In this paper the proposed approach presents that the many ant colonies begin the searching from different random points. Ant Colony Optimization Natural Gas Demand Estimation (ACONGDE) model is developed using population, gross domestic product (GDP), import and export. All equations proposed here are linear and quadratic. Quadratic_ACONGDE provided better fit solution due to the fluctuations of the economic indicators. The ACONGDE model plans the natural gas demand of Turkey until 2025 for three scenarios. We also present an estimation of the natural gas demand for Central Anatolia Region of Turkey using population rates.