Time series forecasting is widely used as a basis for economic planning, production planning, production control and optimizing industrial processes. The aim of this study has been to develop a novel two-level Genetic Algorithm (GA) to optimize time series forecasting in order to forecast cost data for fans used in road tunnels by the Swedish Transport Administration (Trafikverket). The first level of the GA is responsible for the process of forecasting time series cost data, while the second level evaluates the forecasting. The first level implements GA based on the Autoregressive integrated moving average (ARIMA) model. The second level utilizes a GA based on forecasting error rate to identify a proper forecasting. The results show that GA based on the ARIMA model produces better forecasting results for the labor cost data objects. It was found that a multi-objective GA based on the ARIMA model showed an improved performance. The forecasted data can be used for Life cycle cost (LCC) analysis.