Cash Flow Forecasting by Using Time Series Methods in Geothermal District Heating Systems: Balcova - Narlidere Case

Abstract

Cash flow forecasting is one of the difficult and important tasks in an economic evaluation of a geothermal investment. Geothermal district heating systems are characterized by a high capital cost. In addition, relatively low operation and maintenance costs occur throughout their life. The aim of this research is to estimate the potential cash flows for Balcova - Narlidere Geothermal District Heating System by using historical data accumulated over a period of time and several forecasting methods: moving average, exponential smoothing, adjusted exponential smoothing and curve fitting functions. Mean absolute percentage deviation (MAPD) which is the most common approach to select the appropriate method to a particular time series is used in the selection of the most suitable model. Alternative methods are compared with each other regarding to their MAPD values. It is found that the models represented by exponential curve fitting functions have smaller MAPD values and give better results in cash flow forecasting of investment investigated.