Rate of Penetration is defined as the speed at which the drill bit can break the rock under it and thus deepen the well bore. This speed is usually reported in units of feet per hour or meters per hour. In general, the rate of penetration (ROP) optimization means that drilling parameters such as Weight on Bit (WOB), rotary speed, hydraulic properties, mud weight and other variables are adjusted to drill the current formation most efficiently. To bring down the operational costs and to achieve business objectives it is important to identify the factors that influence ROP and optimize them to get the best efficiency. Optimization of various parameters is also the key to reducing operational costs and handling unpredictable or unforeseen problems in drilling operations. To forecast and plan various real-time transactions, it is important to get performance metrics and key performance indicators (KPIs) from historical data as well as monitor current data to detect deviations at the early as possible which is necessary to minimize damage. This thesis proposes a model to predict ROP by identifying key parameters that affect ROP and other individual parameters that play a role in determining penetration rate. The proposed model uses predictive learning algorithms to predict Rate of penetration and Identify key factors that can be optimized to achieve desired ROP.