Flex fuel vehicles (FFV) can operate effectively from E5 (Gasoline 95%, ethanol 5%) fuel to E100 (Gasoline 0%, ethanol 100%) fuel. It is necessary to meet the performance, drivability, emission targets and regulatory requirements irrespective of fuel mixture combination. This research work focuses on optimizing the combustion efficiency and conversion efficiency of catalytic converter of a spark-ignited less than 200 cc engine for FFV using Taguchi methods robust optimization technique. The study employs an eight-step robust optimization approach to simultaneously minimize engine out emissions and maximize catalytic converter efficiency. Six control factors including type of fuel, catalyst heating rpm, lambda (excess-air ratio), injection end angle, lambda controller delay, and ignition timing are optimized. Four noise factors like compression ratio, clearance volume, catalyst noble metal loading, and catalyst aging are also considered. Through approximately 100 physical experiments on a chassis dynamometer, the impact of control factors on engine out emissions and conversion efficiency has been evaluated. Signal-to-noise (S/N) ratios has been calculated for THC and NOx engine-out emissions (combustion efficiency) and their conversion efficiency. Based on the S/N ratio, optimal levels has been chosen for different control factors. The study predicts significant gains in THC conversion and combustion efficiency while modest improvements for NOx conversion. Confirmation runs has been conducted to validate the prediction. The confirmation run yielded slightly lower gains when compared to prediction. The study acknowledges limitations due to control factor interactions but highlights the substantial reduction in experimentation time and effort compared to traditional EFI system calibration methods. Ultimately, the research demonstrates the effective application of robust optimization techniques to meet emission regulations for flex fuel vehicles without significantly increasing production costs. This approach streamlines the optimization process, reducing the number of experiments required and addressing production variations efficiently.