In this paper, a new model for estimating the remaining electrical life of the AC contactor is proposed. The model involves an equation that use the rated and operational parameters, material properties and geometries to estimate electrical health of contactor. The variation in the contact resistance as a function of time is found varying in accordance with the remaining electrical life of the contactor. This characteristic parameter along with the data regarding the quantity of material lost at the contact surface has been fed as the sample data for the machine learning model. The sample data is carefully chosen from the contactors operating at different environmental conditions and utilization categories to include wide range of data. Using the collected data from contactors operating in varied environments and utilization categories a Stochastic Gradient Descent Classifier model is generated and used to estimate the remaining electrical life.