To achieve the Sustainable Development Goals (SDGs), it is essential to monitor the Infant Mortality Rate (IMR) and Child Mortality Rate (CMR), as these indicators are critical for assessing child health. West Java Province recorded the highest number of IMR cases, with 5,234 infants and 299 children. Compared to 2022, when there were 2,959 IMR-related fatalities and 173 CMR-related deaths, this represents a significant increase. The purpose of this study is to model the IMR and CMR using bi-response nonparametric regression model based on penalized spline estimators which has more flexibility to adapt pattern of data. The flexibility in penalized spline estimator is controlled by smoothing parameters, i.e. Lambda parameter, knot points and the number of knots. For determining the optimal smoothing parameter, we use criteria the minimum of Generalized Cross Validation (GCV) value. Based on the minimum GCV, we find that for an increase in the percentage of health facility deliveries, exclusive breastfeeding, and complete basic immunization tends to decrease IMR and CMR. Conversely, an increase in obstetric complications tends to increase IMR and CMR. The best model with Mean Squared Error (MSE) of 1.932198 and a coefficient of determination (R² ) of 0.8787218. It’s mean that the model satisfy goodness of fits criteria.