The current research work is based on the optimization approach that is used in this research can increase the energy conservation potential of a petrol engine that is fuelled with a mixture of petrol and ethanol without raising the cost. In addition to that, this technique may be used to calibrate the engine's universal characteristics, which can then give direction throughout the real process of engine development. This has practical technical relevance as well as demonstrative impact for better promoting the deployment of biofuels. The concentration of ethanol with gasoline at different working parameters such as blend (E0, E8, E16, and E24), speed (1500, 2000, 2500, and 3000 rpm), and load (10, 20, 30, and 40kg), to obtain an optimal dataset on the input factors, which is then fed into two predicted methods, namely multiple regression analysis, and multi-objective genetic algorithm for investigate engine performance and emission characteristics. The Taguchi approach is employed for experimental design. Regression approaches are used to build mathematical models for brake thermal efficiency, brake-specific fuel consumption, and emissions of CO, HC, CO2, and NOX. The regression equations are used to simultaneously optimize combustion by considering characteristics such as brake-specific fuel consumption, as well as emissions of CO, HC, CO2, and NOX. A multi-objective optimization problem is designed to maximize Brake Thermal Efficiency (BTE) and minimize Brake Specific Fuel Consumption (BSFC), Carbon Monoxide (CO), Hydrocarbon (HC), Carbon Dioxide (CO2), and Nitrogen Oxide (NOX) emissions. The genetic algorithm is utilized for forecasting the Pareto optimum sets of solutions. The optimal operating settings yielded the following results: BTHE = 24.89%, BSFC = 0.36 kg/kWh, CO = 0.045%, HC = 3.5%, NOx = 24 ppm, and HC = 267 ppm. The experimental results are in excellent agreement with the optimization results, as shown by the verification research, which found an error ratio of less than 5%. We compare the values of the response variables that were gathered experimentally with the values that were predicted by the Multi-Objective Genetic Algorithm. It turns out that the anticipated values are rather similar to the experimental ones. All things considered, the results show that the multi-objective optimization method used in this study has very efficient predictive tool. The novelty in this chapter is no work has been reported in the literature to predict output values for reducing emissions by using an optimal set of ethanol blends with gasoline for the selection in a spark ignition engine through a combination of multi-regression and multi-objective genetic algorithm search tool which will help to Figureure out how to control the different emission factors and make the engine run better. This research can be used as a teaching tool to increase public knowledge of the possible advantages of ethanol/gasoline blends, including their ability to improve engine efficiency and reduce hydrocarbon emissions.