In recent years, extensive research has focused on applying machine learning (ML) techniques to predict the properties of engineered cementitious composites (ECCs). ECCs exhibit crucial characteristics such as compressive strength (CS), tensile strength (TS), and tensile strain (TSt). Accurate forecasting of these critical properties can reduce material waste, lower construction expenses, and expedite project timelines for engineers and designers. This study investigates mixture design components and corresponding strengths of ECCs based on only polyethylene fiber drawing from existing literatures. Artificial neural network (ANN) models are developed to predict CS, TS, and TSt using a dataset of 339 experimental results with twelve input variables. The ANN models, implemented in MATLAB, consider various hidden layers and neurons to optimize accuracy and validation metrics demonstrate the model's high accuracy. Sensitivity analysis explores individual parameter impacts. Drawing inspiration from this study, it would be advantageous to enhance the predictive modeling toolkit by leveraging the progress made in existing technologies, thereby driving the green and low-carbon development of civil engineering. This approach not only improves the efficiency and sustainability of construction practices but also aligns with global environmental goals by reducing the carbon footprint associated with civil engineering projects.