Saudi Aramco has recently embarked on an exploration program targeting unconventional gas reservoirs to meet the ever increasing local gas demand. One of the targets is a tight sandstone play located to the east of a giant hydrocarbon reservoir with limited reservoir data through the few exploratory wells drilled in that play. The target formation is considered a deep reservoir with very low permeability and high fracture gradient. In order to evaluate the potential of the subject formation, wells have to be completed with a hydraulic fracture. Given the challenging nature of the reservoir in terms of stress, pressure and temperature, a sophisticated design of both well completion and fracturing treatment must be performed to successfully evaluate the potential of those wells.

Developing tight sandstone across vast area requires proper data collection and analysis. Due to the tight nature and heterogeneity of these reservoirs, several vertical and horizontal wells need to be drilled and completed with multistage hydraulic fractures to assess their potential. Initial post-frac flowback tests, in addition to long-term pressure build-ups, have already been conducted on several of the wells. Data Analysis have assisted in characterization of the tight hydrocarbon reservoirs and evaluating of hydraulic fracture geometry. The results have aided to investigate the drainage radius and well interference, to determine the optimal frac and well spacing design. These information are highly needed to build and calibrate single and full field dynamic models to estimate and address the uncertainty on the ultimate recovery and to come up with an optimized development strategy of the field. The paper presents findings and key lessons learned to efficiently design pressure build-up tests in tight sandstone reservoirs.

In heterogeneous tight sand formations, horizontal wells encounter intervals deposited under varying depositional environments along the lateral portion of the wellbore between landing point and total depth. Horizontal wells in this study were drilled in tight sands deposited in a marine environment where lateral depositional facies changes are common, and hydraulic fracture stimulation is necessary to achieve economic hydrocarbon extraction due to the relatively low permeability of the formation. Without geomechanical logs currently derived from wireline logging, it is not possible to optimize cluster spacing and placement. This step provides necesary information used to optimize completion design, which is crucial to the ultimate productivity of a well. Due to formation heterogeneity, expensive wireline logs must be collected in order to optimize fracture stimulation or else new methods to estimate these logs must be employed. This paper presents a technique to optimize cluster selection for hydraulic fracturing in unconventional tight gas development horizontal wells without wireline logging by leveraging Measure While Drilling (MWD) Gamma Ray logs and surface drilling parameters together with Artificial Intelegence (AI) algorythms to predict density, compressional and shear slowness logs for use in geomechanical evaluation.

In unconventional reservoirs, not only do we have to concern ourselves with having a sufficient hydrocarbon volume to justify development (Resource Fairway), but also whether producibility characteristics, especially the ability to effectively fracture stimulate, are favorable (commercially). While volumetric characteristics tend to be regionally variable resulting in changes in hydrocarbon fluid types, reservoir pressures, thickness, porosity, saturation, etc., at the basin scale, producibility characteristics tend to be more locally variable resulting in changes in natural fracturing, subsurface stresses, permeability, effectiveness of fracture stimulation barriers, etc., at the field scale. A stochastic approach encompasses both these trend variations (regional and local) while assessing the commerciality of the play. This paper will explain the workflow involved in creating a probabilistic model using a commercial tool for engineers to understand the mechanics of the model, quality-check the input distributions, compare the model results to those from other sources (such as empirical, analytical, or numerical models). This workflow also will help in seamlessly modifying the input parameters to generate "what-if" scenarios and stress test the base case of development projects. Due to the capital-intensive nature of Unconventional projects, it is important to economically model a series of staged-investments. The purpose of the staging is to responsibly expose incremental capital and identify course corrections. Several factors and combination of inputs affect the commerciality of unconventional developments. These inputs will then be sampled using Monte Carlo simulation to generate a multitude of trials and their summary statistics. The recoverable resources, costs, and economics can also be exported for any given trial, and the ranges of input parameters sampled to generate a given trial can also be extracted Five critical parameters and their ranges were identified for developing a stochastic Before Tax (BTAX) economic output reflecting the full range of outcomes. Well type curves and select components of Capex and Opex were identified as the primary drivers. Associated products, such as condensate, NGLs, ethane, sulfur and water, were modeled as a function of the primary product using dependent ratios. Variability in cycle time was added to account for delays in implementation and acceleration as learnings are applied. This paper will present how technological integration between legacy in-house developed tools and stochastic analysis engines, as the centerpiece, can provide management the required level of visibility on project economics and evaluation ranges. With the considerable time spared in carrying out economic assessments due to this workflow, engineers are able to spend more time in developing case scenarios, run sensitivities, and analyze outcomes and results to make informed portfolio management decisions.