<p>Despite the crucial public health consequences of incoordination in the pediatric vaccine supply chain, resilience and seasonality have not been addressed. This paper contributes to resilient seasonal pediatric vaccine supply analytics by proposing an analytical framework involving demand forecasting using Machine Learning (ML) models, demand seasonality analytics using data science techniques, resilient seasonal supply of pediatric vaccine by solving a multi-period, multi-season mixed-integer linear mathematical model, and global sensitivity analysis using interpretable ML models. The proposed framework is applied in a case study from Iran, involving 31 demand regions, five pediatric vaccines, including BCG, DTP, OPV, MMR, and HB, and three periods. Five ML models, including ARIMA, SARIMA, linear regression, random forest, and XGBoost, are coded in Python to forecast the monthly demand for each pediatric vaccine in each demand region based on an 80–20% partitioning of the dataset for training and testing. Moreover, a sequence of hypothesis tests is employed to detect the seasonal fluctuation patterns of demand, which are incorporated into the model solved using the CPLEX solver to determine seasonal production, shipping, inventory, and delivery from 2026 to 2028. The results revealed that seasonal demand and seasonal holding cost are the most and least significant input parameters affecting cost variability, respectively. In addition, the designed seasonal supply chain was resilient under a seasonal disruption, losing one of two vaccine factories during a season, due to high seasonal production and holding inventory in factories during previous seasons.</p>