This study aims to highlight the significance of weight initialization towards the consistency of Physics-Informed-Neural-Network-based (PINN-based) predictions for spatiotemporal problems in engineering and science. The focus of this study is on a PINN, developed for mass transfer analysis (i.e., PINN-MT) for a single plant cell during drying. While solving Fick's law of diffusion for a cell domain and predicting mass loss and moisture concentration based on convective mass transfer at the cell wall boundary, PINN-MT utilizes moisture concentration at fresh state (i.e., undried) as an initial condition. The governing equations, boundary conditions, and initial conditions are incorporated to the corresponding PINN through the loss function [1]. Residuals of these equations and initial-and-boundary conditions are minimized during the training process of PINN, which predicts moisture concentration variations in time and space scales. However, spatiotemporal problems typically involve a large number of tunable hyperparameters that can make the training process more complicated, leading to inconsistent predictions and loss-convergence issues. This is uncommon in the context of traditional computational approaches [2]. To address this complexity associated with PINNs, pre-trained weight initialization can be adopted, enhancing the ability of PINN-MT to provide consistent solutions via automatic differentiation. In this context, this study assesses effectiveness and efficiency of PINN-MT coupled with weight initialization to address training complexities and provide consistent solutions for spatiotemporal problems in engineering and science.