Nowadays, elastomers, due to excellent work-ability, form-ability, and versatility, play a pivotal role in different industries such as transportation, ship-building, and automotive. During their operation, they are exposed to extreme environmental conditions like temperature, oxygen, water, or chemical substances which change the material structure known as aging. This process changes their properties over time. In this paper, we proposed a physic-informed data-driven model to predict quasi-static mechanical response of elastomers during thermal-induced aging for long-range timescales. We use a swarm of super-simplified neural networks to calculate microstructural strain energy of the material during thermal-induced aging by macroscopic experimental data set. We demonstrate the goodness of the model by using experimental data sets of thermo-induced aging in the literature. We show the behavior of these materials under quasi-static loading and their inelastic behavior such as Mullins effect and permanent set over time. Accuracy and easiness is the most significant achievement of this model.