Simulation of oil drilling time series using Monte Carlo and Bayesian Networks

During oil drilling the main goal is to minimize the total cost of the process. There are two major costs: the cost of the drills and the operation cost. It is necessary to find the best combination of bits in order to maximize the Rate of Penetration (ROP). Many environmental and operational variables influence ROP, but the relationship between them is not always clear. In addition, the lack of historical data makes this problem an even bigger challenge. This paper proposes an approach using Bayesian Networks with the Monte Carlo simulation for generating data for an oil drilling process and compares it with the historical data.