Implementação e análise espaco-temporal do valor presente líquido (dinâmico) em reservatórios de petróleo

Abstract

The Net Present Value (NPV) is recognized as one of the most fundamental and widely adopted metrics in the financial analysis of oil and gas exploration projects. This metric assumes critical importance, considering that oil extraction from a well is not carried out until the resource is exhausted, but rather as long as it is economically viable. Given the intrinsic volatility of oil prices, the decision to shut down a well, even when it still contains oil, is a reality when extraction and maintenance costs exceed potential profits. Although NPV has been extensively studied in the literature, a common simplification in calculations is observed, where a fixed Minimum Attractiveness Rate (MAR) and the costs to construct the NPV are also fixed. This approach does not adequately reflect the volatility of the financial market. Therefore, this study proposes the implementation of a new NPV, a dynamic NPV (DNPV), where a variable MAR will be used, aligned with Brazil's risk-free interest rate (Selic Rate), which plays a crucial role in controlling inflation in the country. Unlike previous studies that consider fixed costs in the NPV calculation, this work also adopts an approach where all costs are variable and indexed to inflation, estimated by the IGP-M index. The use of this index is essential to update oil and gas prices and accurately reflect the economic conditions that varied throughout the analyzed time series. Furthermore, the research advances beyond traditional temporal analysis, proposing a spatio-temporal study of DNPV in oil well reservoirs using real data. To prove the efficiency of the proposed DNPV, its calculation was applied to real data, proving a rate of over 97% accuracy in the decision to maintain or shut down an oil well according to its profit, measured by the DNPV. By developing the temporal analysis of DNPV, it was possible to estimate ARIMA and SARIMA models for the analyzed oil wells. From these models, forecasts were made, generating confidence intervals with 92.50% accuracy in the predictions. By developing the spatial analysis, it was possible to detect the regions and wells with the highest profit, as well as recognize the regions with the lowest DNPV. After conducting spatial predictions, a degree of 95% accuracy was confirmed. The DNPV is expected to provide a more in-depth financial analysis of the profitability of oil and gas production, considering both time and space. This two-dimensional approach may serve as a valuable decision-support tool for the continuity of oil exploration in specific wells and adjacent regions.