Suite Of Models Research Articles

Abnormal low-magnitude seismicity preceding large-magnitude earthquakes

Unraveling the precursory signals of potentially destructive earthquakes is crucial to understand the Earth’s crust dynamics and to provide reliable seismic warnings. Earthquake precursors are ambiguous, but recent experimental studies suggest that robust warning signs may precede large seismic events in the short (day-to-months) term. Here, we show that the M6.4-M7.1 2019 Ridgecrest sequence (California) and the M7.1 2018 Anchorage earthquake (Alaska) were preceded by up to ~3 months of tectonic unrest on regional scales, as evidenced by abnormal low-magnitude seismicity spreading over the ~15-25% of Southern California and Southcentral Alaska. This precursory unrest has been discovered with an algorithm that integrates an innovative random forest machine learning approach and statistical features built from earthquake catalogs. Supported by a novel suite of finite element solid mechanics models, we propose that precursory, abnormal, low-magnitude seismicity arises if the pore fluid pressure within large fault segments escalates significantly as they approach failure, which leads to major uneven changes in the regional stress field. Our findings and method may open up new perspectives for surveillance agencies to anticipate when a region approaches an earthquake of great magnitude weeks to months before it occurs.

Digital Economy, Green Dual Innovation and Carbon Emissions

The digital economy serves as a pivotal catalyst for sustainable and eco-friendly development. This study employs a suite of advanced econometric models, including the fixed effects, mediation, threshold and moderation model, to elucidate the intricate dynamics by which the digital economy influences carbon emissions through the lens of green innovation. Building on the existing research on digital economy, green technology innovation and carbon emissions, this paper takes a dual-innovation perspective and divides green technology innovation into disruptive green technology innovation and incremental green technology innovation. And from the government and the public level, it explores how social concerns affect the effect of digital economy on carbon emissions. The analysis is grounded in a comprehensive dataset encompassing a decade of provincial-level data from 2011 to 2021 across China’s 30 provinces. The benchmark regression outcomes indicate the digital economy’s ability to substantially cut down carbon emissions; the threshold effect and mediating effect models demonstrate that a single-threshold effect exists and that disruptive and progressive green technological innovations mediate such ability. Additional research reveals that the digital economy’s impact on carbon emissions could be positively moderated by public and governmental attention. Eastern and western regions in China, as well as those with high levels of foreign investment and low levels of technological transaction activity, are more affected by the digital economy in terms of carbon emission reduction. Our conclusions offer practical recommendations for digital economy’s coordinated advancement and carbon emissions mitigation, and guide local governments to achieve sustainable development goals (SDGs).

BEForeGAN: An image-based deep generative approach for day-ahead forecasting of building HVAC energy consumption

This study presents a pioneering approach in building energy forecasting by introducing a novel reformulation framework that transforms the forecasting task into an image inpainting problem. Based upon the fundamental notion that “forecasting is about generating data of the future”, we propose BEForeGAN, an innovative deep generative approach for day-ahead Building HVAC Energy consumption Forecasting based on multi-channel conditional Generative Adversarial Networks (GANs) with U-Net generators. Our method is evaluated using 96,360 hourly HVAC energy consumption records from 11 buildings, demonstrating significant accuracy improvements of 17%∼76% and a substantial variability reduction of 3%∼96% compared to a suite of conventional and deep learning benchmark models across individual-building and zero-shot cross-building forecasting tasks. Notably, BEForeGAN exhibits robustness to noisy inputs, with an increase below 3% in Coefficient of Variation of Root Mean Square Error (CV-RMSE) for each 10% noise increment. This study addresses critical gaps in existing literature by showcasing the untapped potential of GANs as standalone forecasters, advocating for further exploration of two-dimensional (2D) GAN-based methods in building energy forecasting, and emphasising the need for more studies focusing on cross-building forecasting tasks. In conclusion, our findings underscore the transformative impact of GANs in revolutionising building energy forecasting practices, paving the way for enhanced energy-efficient building management and beyond.

A multi-decadal 1 km gridded database of continental-scale spring onset products

The timing of spring onset is a particularly sensitive climate change indicator that also allows the study of weather interannual variations and extremes. This indicator can be derived from a suite of phenological models called the Extended Spring Indices (SI-x). These models transform daily minimum and maximum temperatures into a set of consistent indices that track the timing of first leaf and first bloom for key indicator plant species. The SI-x also enables the calculation of the so-called frost damage index. Using new computational technologies and high spatial resolution gridded weather data, here we present and evaluate a multi-decadal 1 km spatial resolution version of the SI-x models, which cover North American (1980 to 2022) and European areas (1950 to 2020). Because of the generally good agreement with ground phenological observations, this continental-scale and high spatial resolution product can support both scientists and decision makers in their quest to hind- and forecast weather and climate change impacts.

Extension of the SpK atomic physics code to generate global equation of state data

Global microphysics models are required for the modelling of high-energy-density physics (HEDP) experiments, the improvement of which are critical to the path to inertial fusion energy. This work presents further developments to the atomic and microphysics code, SpK, part of the numerical modelling suite of Imperial College London and First Light Fusion. We extend the capabilities of SpK to allow the calculation of the equation of state (EoS). The detailed configuration accounting calculations are interpolated into finite-temperature Thomas–Fermi calculations at high coupling to form the electronic component of the model. The Cowan model provides the ionic contribution, modified to approximate the physics of diatomic molecular dissociation. By utilising bonding corrections and performing a Maxwell construction, SpK captures the EoS from states ranging from the zero-pressure solid, through the liquid–vapour coexistence region and into plasma states. This global approach offers the benefit of capturing electronic shell structure over large regions of parameter space, building highly-resolved tables in minutes on a simple desktop. We present shock Hugoniot and off-Hugoniot calculations for a number of materials, comparing SpK to other models and experimental data. We also apply EoS and opacity data generated by SpK in integrated simulations of indirectly-driven capsule implosions, highlighting physical sensitivities to the choice of EoS models.

Landscape-Scale Modeling to Forecast Fluvial-Aeolian Sediment Connectivity in River Valleys.

Sedimentary landforms on Earth and other planetary bodies are built through scour, transport, and deposition of sediment. Sediment connectivity refers to the hypothesis that pathways of sediment transport do not occur in isolation, but rather are mechanistically linked. In dryland river systems, one such example of sediment connectivity is the transport of fluvially deposited sediment by wind. However, predictive tools that can forecast fluvial-aeolian sediment connectivity at meaningful scales are rare. Here we develop a suite of models for quantifying the availability of river-sourced sediment for aeolian transport as a function of river flow, wind regime, and land cover across 168 km of the Colorado River in Grand Canyon, USA. We compare and validate these models using topographic changes observed over 10 years in a coupled river sandbar-aeolian dunefield setting. The models provide a path forward for directly linking fluvial hydrology with the management and understanding of aeolian landscapes.

Spectra of GRB 221009A at Low Energies Derived from Ground-based Very Low-frequency Measurements

The gamma-ray burst (GRB) event GRB 221009A was the brightest event that has ever been detected to date. Owing to its unexpected brightness, the temporal and/or spectral information of the prompt emission cannot be accurately measured by many satellites (with the only exception of GECAM-C), since they suffered from significant pulse pileup and data saturation effects. Similarly, the X45 solar flare event occurring on 2003 November 4 saturated space-borne X-ray detectors, and it was through ground-based measurements of very low-frequency (VLF) signals that the magnitude of this event was determined, since VLF signals are particularly sensitive to the disturbance on the D-region ionosphere caused by low-energy photons. Therefore, in this study, we first report measurements of VLF signals from the JJI and VTX transmitter as recorded in Shiyan, China, when GRB 221009A occurred. The amplitude change was ∼1.25 and ∼2.31 dB for the JJI and VTX transmitter, respectively. Using a suite of well-validated models, we have further simulated the influence on the D-region ionosphere induced by low-energy photons (<100 keV) of GRB 221009A. Compared with the pre-GRB condition, the electron density was enhanced by 39.75% and 626.61% at 60 and 70 km altitude for the VTX-SYS path and 39.73% and 621.11% at 60 and 70 km altitude for the JJI-SYS path, respectively, with the altitude of notable electron density change being as low as ∼30 km. Moreover, we have compared modeling results of VLF signal change with our measurements during GRB 221009A. The good agreements obtained in terms of amplitude change and overall trend validate the fluxes and spectra of GRB 221009A at low energies (<20 keV) as measured by GECAM-C.

Results from a synthetic model of the ITER XRCS-Core diagnostic based on high-fidelity x-ray ray tracing.

A high-fidelity synthetic diagnostic has been developed for the ITER core x-ray crystal spectrometer diagnostic based on x-ray ray tracing. This synthetic diagnostic has been used to model expected performance of the diagnostic, to aid in diagnostic design, and to develop engineering tolerances. The synthetic model is based on x-ray ray tracing using the recently developed xicsrt ray tracing code and includes a fully three-dimensional representation of the diagnostic based on the computer aided design. The modeled components are: plasma geometry and emission profiles, highly oriented pyrolytic graphite pre-reflectors, spherically bent crystals, and pixelated x-ray detectors. Plasma emission profiles have been calculated for Xe44+, Xe47+, and Xe51+, based on an ITER operational scenario available through the Integrated Modelling & Analysis Suite database, and modeled within the ray tracing code as a volumetric x-ray source; the shape of the plasma source is determined by equilibrium geometry and an appropriate wavelength distribution to match the expected ion temperature profile. All individual components of the x-ray optical system have been modeled with high-fidelity producing a synthetic detector image that is expected to closely match what will be seen in the final as-built system. Particular care is taken to maintain preservation of photon statistics throughout the ray tracing allowing for quantitative estimates of diagnostic performance.

Dynamical formation of Gaia BH3 in the progenitor globular cluster of the ED-2 stream

Context. The star–black hole (S–BH) binary known as Gaia BH3, discovered by the Gaia Collaboration is chemically and kinematically associated with the metal-poor ED-2 stream in the Milky Way halo. Aims. We explore the possibility that Gaia BH3 was assembled dynamically in the progenitor globular cluster (GC) of the ED-2 stream. Methods. We used a public suite of star-by-star dynamical Monte Carlo models to identify S–BH binaries in GCs with different initial masses and (half-mass) radii. Results. We show that a likely progenitor of the ED-2 stream was a relatively low-mass (≲105 M⊙) GC with an initial half-mass radius of ∼4 pc. Such a GC can dynamically retain a large fraction of its BH population and dissolve on the orbit of ED-2. From the suite of models we find that GCs produce ∼3 − 30 S–BH binaries, approximately independently of initial GC mass and inversely correlated with initial cluster radius. Scaling the results to the Milky Way GC population, we find that ∼75% of the S–BH binaries formed in GCs are ejected from their host GC, all in the early phases of evolution (≲1 Gyr); these are expected to no longer be close to streams. The ∼25% of S–BH binaries retained until dissolution are expected to form part of streams, such that for an initial mass of the progenitor of ED-2 of a few 104 M⊙, we expect ∼2 − 3 S–BH to end up in the stream. GC models with metallicities similar to Gaia BH3 (≲1% solar) include S–BH binaries with similar BH masses (≳30 M⊙), orbital periods, and eccentricities. Conclusion. We predict that the Galactic halo contains of order 105 S–BH binaries that formed dynamically in GCs, a fraction of which may readily be detected in Gaia DR4. The detection of these sources provides valuable tests of BH dynamics in clusters and their contribution to gravitational wave sources.

Predicting transit ridership using an agent-based modeling approach

Accurate ridership estimation is pivotal in the advancement of sustainable transit systems, be it for proposed or existing transit networks. A multitude of methods, including travel demand models, direct ridership models, and regression models, have been employed by practitioners and researchers to estimate ridership at both station and network levels. However, travel demand models, frequently utilized for new transit lines, exhibit intrinsic limitations due to their aggregate nature and complexity based on their types. Researchers have also identified deficiencies, such as the incapacity to capture small spatial resolutions and specific station characteristics, as these models are predominantly designed for large-scale analyses.This study aims to overcome these limitations by introducing a novel approach that utilizes three microscopic agent-based models to develop a travel demand modeling suite, providing a policy-sensitive forecasting tool. The suite comprises three agent-based models: SILO-MITO-MATSim. Validation of the model against previous year data is conducted, and projections are made for future years. The model is applied to estimate network-level ridership for the proposed ‘Purple Line,’ a light rail transit line planned by MDOT, MTA, Maryland, which will integrate with the Washington D.C. Metro, the fourth largest transit system in the USA, boasting an average daily ridership of half a million. The study’s findings indicate an anticipated ridership of approximately 31,230 passengers in the inaugural year of 2027. The proposed model offers a robust and policy-sensitive solution empowering decision-makers to make informed choices to support a sustainable transportation system.

An Empirically Constrained Forecasting Strategy for Induced Earthquake Magnitudes Using Extreme Value Theory

Abstract Induced seismicity magnitude models seek to forecast upcoming magnitudes of induced earthquakes during the operation of subsurface industries such as hydraulic fracturing, geothermal stimulation, wastewater disposal (WWD), and carbon capture and storage. Accurate forecasting models could guide operational decision making in real time; for example, operations could be reduced or paused if forecast models indicate that magnitudes may exceed acceptable levels. Robust and transparent testing of forecasting models is required if they are to be adopted by operators and regulators of such industries. We develop and test a suite of models based on extreme value estimators to forecast the magnitudes of upcoming induced seismic events based on observed seismicity. We apply these models to multiple induced seismicity cases from WWD in Oklahoma and in western Texas, as well as other cases of seismicity caused by subsurface fluid injection in North America, Europe, and China. In total, our testing dataset consists of &amp;gt;80 individual sequences of induced seismicity. We find that all the models produce strong correlation between observed and modeled magnitudes, indicating that the forecasting provides useful information about upcoming magnitudes. However, some models are found to systematically overpredict the observed magnitudes, whereas others tend to underpredict. As such, the combined suite of models can be used to define upper and lower estimators for the expected magnitudes of upcoming events, as well as empirically constrained statistical expectations for how these magnitudes will be distributed between the upper and lower values. We conclude by demonstrating how our empirically constrained distribution can be used to produce probabilistic forecasts of upcoming induced earthquake magnitudes, applying this approach to two recent cases of induced seismicity.

Experimental demonstration of magnetic tunnel junction-based computational random-access memory

The conventional computing paradigm struggles to fulfill the rapidly growing demands from emerging applications, especially those for machine intelligence because much of the power and energy is consumed by constant data transfers between logic and memory modules. A new paradigm, called “computational random-access memory (CRAM),” has emerged to address this fundamental limitation. CRAM performs logic operations directly using the memory cells themselves, without having the data ever leave the memory. The energy and performance benefits of CRAM for both conventional and emerging applications have been well established by prior numerical studies. However, there is a lack of experimental demonstration and study of CRAM to evaluate its computational accuracy, which is a realistic and application-critical metric for its technological feasibility and competitiveness. In this work, a CRAM array based on magnetic tunnel junctions (MTJs) is experimentally demonstrated. First, basic memory operations, as well as 2-, 3-, and 5-input logic operations, are studied. Then, a 1-bit full adder with two different designs is demonstrated. Based on the experimental results, a suite of models has been developed to characterize the accuracy of CRAM computation. Scalar addition, multiplication, and matrix multiplication, which are essential building blocks for many conventional and machine intelligence applications, are evaluated and show promising accuracy performance. With the confirmation of MTJ-based CRAM’s accuracy, there is a strong case that this technology will have a significant impact on power- and energy-demanding applications of machine intelligence.

Closing the gap: secular evolution of bar-induced dark gaps in the presence of thick discs

ABSTRACT The presence of dark gaps, a preferential light deficit along the bar minor axis, is observationally well known. The properties of dark gaps are thought to be associated with the properties of bars, and their spatial locations are often associated with bar resonances. However, a systematic study, testing the robustness and universality of these assumptions, is still largely missing. Here, we investigate the formation and evolution of bar-induced dark gaps using a suite of N-body models of (kinematically cold) thin and (kinematically hot) thick discs with varying thick disc mass fractions and different thin-to-thick disc geometries. We find that dark gaps are a natural consequence of the trapping of disc stars by the bar. The properties of dark gaps (such as strength and extent) are well correlated with the properties of bars. For stronger dark gaps, the fractional mass-loss along the bar minor axis can reach up to ${\sim} 60\!-\!80$ per cent of the initial mass contained, which is redistributed within the bar. These trends hold true irrespective of the mass fraction in the thick disc and the assumed disc geometry. In all our models harbouring slow bars, none of the resonances (corotation, inner Lindblad resonance, and 4:1 ultraharmonic resonance) associated with the bar correspond to the location of dark gaps, thereby suggesting that the location of dark gaps is not a universal proxy for these bar resonances, in contrast with earlier studies.

Smart reception: An artificial intelligence driven bangla language based receptionist system employing speech, speaker, and face recognition for automating reception services

In recent times, service robots (SR) have become widely accepted in a variety of fields as an alternative to traditional reception methods. Artificial Intelligence (AI) driven systems are seen as efficient labor alternatives, resulting in several SR models already being developed, primarily designed for English-speaking environments. Despite a large Bangla-speaking population, the exploration and implementation of Bangla language support within AI reception systems remain unexplored due to limited resources, posing significant challenges for deployment. This study presents a novel AI-enabled receptionist framework tailored to Bangla-speaking environments, addressing the limitation of Bangla language resources for developing automated reception systems. Leveraging advanced AI technologies including Face Recognition, Speaker Recognition, Automatic Speech Recognition (ASR), Text-to-Speech Synthesis, and Question Answering System, our integrated system demonstrates promise to automate Bangla reception systems. Our suite of models yields positive performance, with a face recognition accuracy of 99.38%, a speaker recognition system with 5.83 Equal Error Rate (ERR), ASR model achieving a Word Error Rate (WER) of 9.005%, TTS model scoring a Mean Opinion Score (MOS) of 4.10, and a question-answering system with a validation loss of 0.03%. Real-world evaluation among 1664 users in a university setting achieved over 75% user satisfaction, with 88% expressing interest in real-life implementation, showcasing usability across different domains. Limitations such as limited training data, scalability, and environmental sensitivity persist, underscoring the need for further development. Nevertheless, our framework demonstrates potential for real-life implementation, fostering human-robot interaction in Bangla-speaking contexts and paving the way for future innovations in AI-driven Bangla receptionist systems.

The Effect of Age and Stellar Model Choice on Globular Cluster Color-to-metallicity Conversions

The photometric colors of globular clusters (GCs) act as effective proxies for metallicity, since all normally used optical/IR color indices exhibit a nonlinear but monotonic relation between their integrated color and their metallicity. One color index, (g − z) or (F475W–F850LP), has been spectroscopically calibrated in several studies, providing leverage to define color-to-metallicity conversions for other indices. In this paper, building on the work of Hartman et al., we study the GC color-metallicity relation in more detail by testing the dependence of the relations on different suites of stellar models and different assumed GC ages. Though noticeable differences between models exist, we find that the net effect on the derived GCS metallicity distributions is small.

Regression analysis and its application to oil and gas exploration: A case study of hydrocarbon loss recovery and porosity prediction, China

In oil and gas exploration, elucidating the complex interdependencies among geological variables is paramount. Our study introduces the application of sophisticated regression analysis method at the forefront, aiming not just at predicting geophysical logging curve values but also innovatively mitigate hydrocarbon depletion observed in geochemical logging. Through a rigorous assessment, we explore the efficacy of eight regression models, bifurcated into linear and nonlinear groups, to accommodate the multifaceted nature of geological datasets. Our linear model suite encompasses the Standard Equation, Ridge Regression, Least Absolute Shrinkage and Selection Operator, and Elastic Net, each presenting distinct advantages. The Standard Equation serves as a foundational benchmark, whereas Ridge Regression implements penalty terms to counteract overfitting, thus bolstering model robustness in the presence of multicollinearity. The Least Absolute Shrinkage and Selection Operator for variable selection functions to streamline models, enhancing their interpretability, while Elastic Net amalgamates the merits of Ridge Regression and Least Absolute Shrinkage and Selection Operator, offering a harmonized solution to model complexity and comprehensibility. On the nonlinear front, Gradient Descent, Kernel Ridge Regression, Support Vector Regression, and Piecewise Function-Fitting methods introduce innovative approaches. Gradient Descent assures computational efficiency in optimizing solutions, Kernel Ridge Regression leverages the kernel trick to navigate nonlinear patterns, and Support Vector Regression is proficient in forecasting extremities, pivotal for exploration risk assessment. The Piecewise Function-Fitting approach, tailored for geological data, facilitates adaptable modeling of variable interrelations, accommodating abrupt data trend shifts. Our analysis identifies Ridge Regression, particularly when augmented by Piecewise Function-Fitting, as superior in recouping hydrocarbon losses, and underscoring its utility in resource quantification refinement. Meanwhile, Kernel Ridge Regression emerges as a noteworthy strategy in ameliorating porosity-logging curve prediction for well A, evidencing its aptness for intricate geological structures. This research attests to the scientific ascendancy and broad-spectrum relevance of these regression techniques over conventional methods while heralding new horizons for their deployment in the oil and gas sector. The insights garnered from these advanced modeling strategies are set to transform geological and engineering practices in hydrocarbon prediction, evaluation, and recovery.

A multi-modal, asymmetric, weighted, and signed description of anatomical connectivity

The macroscale connectome is the network of physical, white-matter tracts between brain areas. The connections are generally weighted and their values interpreted as measures of communication efficacy. In most applications, weights are either assigned based on imaging features–e.g. diffusion parameters–or inferred using statistical models. In reality, the ground-truth weights are unknown, motivating the exploration of alternative edge weighting schemes. Here, we explore a multi-modal, regression-based model that endows reconstructed fiber tracts with directed and signed weights. We find that the model fits observed data well, outperforming a suite of null models. The estimated weights are subject-specific and highly reliable, even when fit using relatively few training samples, and the networks maintain a number of desirable features. In summary, we offer a simple framework for weighting connectome data, demonstrating both its ease of implementation while benchmarking its utility for typical connectome analyses, including graph theoretic modeling and brain-behavior associations.

Construction of a Real-Time Forecast Model with Deep Learning Techniques for Coastal Engineering and Processes: Nested in a Basin Scale Suite of Models

This study aims to develop a robust and adaptable real-time forecasting system for coastal and estuarine regions, considering the challenges posed by the limited or unavailable forecast data. A real-time forecast system has been designed to handle three distinct scenarios: forecast data from global models are available, unavailable, or intermittently unavailable. To address the challenge of data unavailability, this research proposes the application of a deep learning model (DLM). This study involves the construction of a high-resolution numerical model nested within global models. The performance of the DLM is evaluated by comparing the simulation results of the nested model generated using DLM-predicted data against those obtained using data from global models. The results demonstrate a high correlation and around 90% accuracy for up to the initial 5 days of the forecast. Even in the absence of hindcast data up to 4 days prior to the forecast beginning time, the real-time forecast with DLM can achieve an accuracy exceeding 70% of the forecast data from global models. However, direct application of the developed DLM to an unknown domain results in significantly lower performance, highlighting the importance of retraining the DLM with local data. Overall, this study presents a comprehensive approach to constructing a real-time forecasting system for coastal and estuarine regions that adeptly handles various data availability scenarios through the integration of machine learning techniques.

Intercomparison of ocean temperature and circulation near the Galápagos islands in high-resolution models and observations

The mean structure and variability of the Equatorial Undercurrent (EUC) have important implications for upwelling, sea surface temperature (SST), and productivity in the ecologically vital Galápagos Cold Pool. Historically, global coupled climate model simulations have exhibited considerable biases in their simulation of the EUC due to the requirement of relatively high spatial resolution to represent its dynamics. Particularly in the eastern equatorial Pacific, models must also adequately resolve important topographic features to accurately simulate the regional circulation. Here, we examine the extent to which a high-resolution configuration of the NCAR Community Earth System Model (CESM) and a suite of models from the High-Resolution Model Intercomparison project (HighResMIP) adequately represent the regional ocean circulation and other important climatological features of the eastern equatorial Pacific such as the EUC and the associated temperature patterns defining the cold tongue/Galápagos Cold Pool complex. Comparisons with satellite SST and in situ velocity observations, and a high-resolution ocean reanalysis product, illustrate that the high-resolution configuration of the CESM captures many key aspects of the SST field and EUC uniquely well, including its seasonal-to-interannual variability in the eastern equatorial Pacific. Specific strengths and biases of this model with direct comparison to the HighResMIP ensemble are discussed in detail, along with the potential for application of these models to interdisciplinary research topics such as projecting climate change impacts on marine ecosystems.

Enhancing refinery heavy oil fractions analytical performance through real-time predicative modeling.

This study introduces a suite of robust models aimed to advance the determination of physiochemical properties in heavy oil refinery fractions. By integrating real-time analytical technique inside the refinery analysis, we have developed a single analyzer capable of employing six partial least square regression equations. These designed models enable to provide real-time prediction of critical petroleum properties, such as sulfur content, micro carbon residues (MCR), asphaltene content, heating value, and the concentrations of nickel and vanadium metals. Specifically tailored for heavy oil in refinery feeds with an American petroleum institute (API) gravity range of 3° to 32° and sulfur content of 2.8 to 5.5 wt%, the models streamline the analysis process within refinery operations, bridging the gap between catalytic and non-catalytic processes across refinery units. The accuracy of our physiochemical prediction models has been validated against American Society for Testing and Materials (ASTM) standards, demonstrating their capability to deliver precise real-time property values. This approach not only enhances the efficiency of refinery analysis but also sets a new standard for the monitoring and optimization of heavy oil processing in real-time approach.