Geology–Engineering Double Sweet Spot in Unconventional Reservoirs: A Comprehensive Review

Summary Amid the global energy transition, the precise identification of geology–engineering “sweet spots” has become critical to the efficient development of unconventional oil and gas resources worldwide. Focusing on the low-porosity/permeability and high brittleness characteristics of unconventional reservoirs, in this study, we first systematically review the controlling factors for double sweet spot. Key findings reveal that geology sweet spot is primarily governed by hydrocarbon generation potential, reservoir quality, and preservation conditions; engineering sweet spot depends on intrinsic mechanical properties (brittleness, fracture toughness) and external environmental factors (natural fracture, in-situ stress, sedimentation, and diagenesis). Next, we summarize the applicability and limitations of key prediction methods, including core testing, well logging/seismic interpretation, and machine learning (ML). ML demonstrates significant advantages in critical tasks such as total organic carbon (TOC) prediction, fracture identification, and 3D in-situ stress field reconstruction. Based on accurate parameter acquisition and correction, we establish a double sweet-spot model by integrating four core mechanisms, which are (1) correction factor optimization, (2) weighted coupling of multiscale data, (3) ML-driven parameter fusion, and (4) integrated classification of key geological and engineering factors. Future research should advance lithofacies-controlled integration, rock mechanics constitutive model, and life cycle intelligent prediction via multiscale data-ML synergy, enabling precise characterization of double sweet spot for efficient unconventional reservoir development.