Dengue fever, an acute health concern in Bangladesh, has dramatically increased in frequency and severity in recent years. The dengue epidemic experienced in 2023 represented an unprecedented peak in the incidence of the disease within the nation’s recorded history. In this study, our main goal was to create forecasting models for dengue-affirmed cases in the unique context of Bangladesh, primarily driven by the escalating health crisis of the dengue epidemics. We examined historical confirmed cases of dengue infection in Bangladesh from 2008 to 2023 to comprehend the underlying pattern and seasonal fluctuations with the help of univariate time series analysis. We further proposed the SARIMA, Holt-Winters, Prophet, and LSTM models for dengue prediction. The novel aspect of this research was to introduce the utilization of the Holt-Winters method and new parameters for the SARIMA model, within Bangladesh’s particular circumstances of dengue virus. To identify the best model, each one was assessed using simple evaluation metrics like MAE and RMSE. The most effective predictive model indicated a potential surge in future dengue cases that exceeded any previously recorded outbreaks, underscoring the critical need for immediate policy intervention. Finally, the paper concludes with future research scope to enhance dengue surveillance and response. We anticipate that the insights gained from this study will be instrumental in enabling proactive measures to effectively address upcoming dengue outbreaks.