The fundamental goal of a smart environment is reducing carbon emission from all possible aspects of human life. Appropriate flight selection, using shared transportation, tree plantation, switching to clean energy and using energy saving appliances at home are a few pointers towards achieving the same. Air trip contributes 2.5% of global direct CO2 emissions, and 3.5% of non-CO2 emissions globally but cannot be avoided fully in this fast-paced world. Hence, an innovative regression-based model is proposed to predict the total carbon emission for a passenger so that the user can choose the optimal flight accordingly. Distance, Flight No, IATA Code and Airline Name are received as direct input parameters from user; however, few features are derived instead of directly using them, to enhance the model accuracy. climb-cruise-descent (CCD) Phase Time is the first one which indicates the actual flying time, calculated based on Distance. Aircraft Type is computed through Flight No, IATA Code and Airline Name. The last two are Engine Type and No of Engines, derived from ‘Aircraft Type and Engine Details’ reference table which is received through aviation council. All these features are highly correlated with flight emission. Additionally, the ‘Travel Impact Model’ the google API is utilized to generate the training dataset; however, once the model is trained, there will not be any dependency on this API. The model can independently predict passenger-wise emission based on new set of inputs. The USP is that it can be plugged into other applications and independent of travel date, which is a pre-requisite for few existing available APIs in the market.