digilib@itb.ac.id +62 812 2508 8800

In the recent years, drilling optimization techniques have been used to reduce drilling operation costs. This would be done by reducing the operation time and minimzing the non productive time, since time is always money in drilling operations. Almost all experimental during the drilling progress on site are always looking for the prediction of unexpected events and optimizing the drilling parameters. Analyzing and predicting the Rate of Penetration (ROP) is a prominent factor for drilling engineers due to its effect on the optimization of various parameters that drives to economical and engineering decisions during the well planning. ROP models can be used to estimate the formation drillability by considering the effects of drilling parameters, bits design, bit wear, and fomation properties. Drilling optimization using ROP models us done by changing the drilling parameters and / or bit design to find the optimum drilling scenarios for an entire bit run. Various mathematical equations have been suggested to make a relation between ROP and major controllable and uncontrollable drilling parameters. Over the past several years, the operator required an intellectual and comprehensive method to assess drilling performance prior to spud using the available components of the lithologies, loggings, and bit engagement informations. the drilling created many mathemarical formulas and techncial papers that outline ROP predictions processes. The inpiut parameters are true vertical depth (TVD), weight on bit (WOB), rotation per minute (RPM), mud weight (MW), pump pressure, D.exponent, and equivalent circulating depth (ECD), also rate of penetration (ROP) as a final prediction. All of parameters are classified as per bit size utilized on 4 wells. This study will generate the artificial neural network modelling, evaluate the actual ROP based on each bit utilized, and determine the influence input parameters for ROP prediction. By applying the neural network capabilities for ROP Prediction, the analytical of ROP for the planned well, including quality result for different lithologies with a wide range of rock strength values. This research will focus on the ANN technologies in predicting the drilling performances, also include the multivariate regression analysis to generate the ROP equation with the drilling parameters.