The novel use of Machine Learning (ML) software along with advanced dosing controllers to optimize odor control chemical dose rates is discussed in this paper. Through the application of ML, the Xylem team was able to achieve better Hydrogen Sulfide (H2S) compliance using industry standard measurement methods. ML modeling is effective in analyzing complex equations and a large amount of historical data within a short period of time. This significantly reduces the time and effort that is required in maintaining H2S compliance and preventing odor complaints. Machine Learning is one of the uprising soft computing and communication technologies widely used in various industries for process monitoring and optimization. ML tools are effective for optimal modelling and data forecasting. Municipal utilities, which are highly driven by environmental and social factors, and subject to immutable infrastructure and budget constraints, will benefit from continuous optimization using ML modeling in their liquid phase odor control programs. In an ideal world, if we know all variables such as sewage flowrate, temperature, chemical composition including BOD, COD, sulfate concentration, water infiltration then a precise minute to minute, hour to hour dosing profile could be calculated. However, those variables are often unavailable due to price, missing technology, or infrastructure limitations. With the help of ML software modeling, we can achieve better optimization with very little effort and time and overcome unavailable variables. By providing accurate upper and lower limits we will be able to operate with higher productivity and confidence. The presented case studies indicate that ML modeling was able to optimize odor control dose rates just as effectively and consistently than optimizing manually in a very short period, proving that this tool can logically predict the required chemical needed.