This thesis employs a quantitative and technical approach to evaluate time series functions as a forecasting method for ice cream sales in Sweden, focusing on two distinct ice cream types. Utilizing weekly historical data spanning from 2021 to 2024, the study explores the application of time series forecasting in @Risk, an add-on application in Microsoft Excel, specifically its "Time Series” functionality, to conduct simulations with user-defined parameters and settings. The thesis investigates the impact of varying settings in @Risk on the forecasting simulations for three different historical data periods throughout the year including high and low season for ice cream sales. Evaluation of the simulations is based on the forecast error which is calculated as the difference between forecasted values and actual sales data divided by the forecasted value. Findings reveal that limited historical data leads to significant variations in simulation results, amplifying uncertainties and rendering forecasts unreliable. However, incorporating seasonality into the settings improves forecast accuracy by decreasing the forecast error in many instances. While the project's conclusions are specific to its context, the methodology and results hold valuable insights for similar projects within the Fast-Moving Consumer Goods (FMCG) industry or for time-series functionality.