<![CDATA[Infant vaccination is important in supporting growth and strengthening the immune system. One of the challenges faced is the imbalance between vaccine supply and demand in various regions, which can lead to distribution shortages. This study aims to predict the supply of infant vaccines to reduce distribution gaps using the Holt-Winters Exponential Smoothing method. This method is applied using two approaches: an additive and a multiplicative model based on monthly data from 2021 to 2024. The results show that the multiplicative model is more accurate for the bivalent oral polio vaccine (BOPV), hepatitis B (HBO), and measles-rubella (MR) vaccines because demand exhibits significant fluctuations. The additive model is more accurate for Bacillus Calmette-Guérin (BCG), diphtheria-pertussis-tetanus (DPT), and inactivated poliovirus vaccine (IPV) because demand tends to be stable around a constant average value. The BOPV vaccine yields perfect accuracy (MAPE< 10%) and reasonably good accuracy for the HBO vaccine (MAPE< 20%). The BCG and MR vaccines have low accuracy levels (MAPE< 50%). The DPT and IPV vaccines have bad accuracy levels (MAPE> 50%). Accuracy levels can be influenced by demand fluctuations, uneven distribution, and adjustments to the α, β, and ꝩ parameters. The results of this study indicate that the Holt-Winters Exponential Smoothing method can help predict vaccine supply fluctuations more accurately, thereby supporting more even distribution across all regions.]]>
