• AR2Tech has joined Seequent
    Working together to build a better understanding of the earth.
    Learn More
  • The new generation of
    Parallel. Distributed. On Cloud. On Servers.
    Learn More
  • Take geomodeling
    To the next level
    Classical geostatistics. Stochastic optimization. Machine-learning.
    Learn More
  • Design your geomodeling solutions
    Custom workflows
    C++ speed and robustness. Python flexibility
    Learn More
AR2Tech Logo

Advanced Resources and Risk Technology

AR2Tech provides state-of-the-art geostatistical applications. We are experts in geostatistics, geomodeling, and spatial data integration.

Our algorithms can run on desktops or on distributed computing infrastructures, on the cloud or on your own servers.

Geomodeling & Uncertainty Analysis.
Since 2010.

AR2GAS: High Performance library

Geostatistics As a Service

Flexibility & integration

AR2GAS is a series of geostatistical libraries with C++ and Python APIs. All the algorithms can be run from a script or a Jupyter notebook locally or on high-performance servers. AR2GAS is ready to be integrated into machine learning workflows for state-of-the-art modelling workflow.

Designed for speed

AR2GAS is the result of a complete overhaul of our geostatistical libraries, built from the ground up to work efficiently both locally or on distributed computing systems. The same geostatistical workflow can now be executed on the desktop, the cloud or on private clusters.

Highly customizable

AR2GAS algorithms are designed to work across scales and gridding systems with data that varies in density and quality. Geomodeling situations, objectives and workflows vary. The AR2GAS Python API provides the flexibility to fit your needs via a Jupyter Notebook, as part of a complex python script or in conjunction with other tools. The user has full freedom to create custom geomodeling workflows.

Connect to Database

AR2GAS is built to connect to external data sources and extract data on a per need-basis allowing for efficent workflows on very large models (billion of cells). The data stream is optimized for each algorithm providing performance and scaling.

Run as process

Easily build complex modelling algorithms in python as processes with strong-typed parameters. This feature greatly decreases the chance of user errors when initializing algorithms and provide robustness during execution by isolating the execution from the main program.

Create Dashboards

AR2GAS provides several dashboards to efficiently and interactively validate geostatistical models. A dashboard presents relevant interactive charts and metrics in a browser. When hosted on a server, the dashboard can be accessed simultaneously by team members for analysis without the need of external software suite.

Cloud performance

Geostatistics on The Cloud

Creating a diverse portfolio of geological models with millions of cells or optimizing on geomodeling properties of large models requires computing power and scalability. AR2GAS is designed to leverage cloud computing environments and break the feasibility limits of desktop-centric geomodeling tools.

Property estimations and simulations are now accessible in a matter of minutes for grids over one billion points. Multiple models can be generated in parallel, and parallelization can also be achieved optimally within models. Cloud computing makes it possible to perform uncertainty quantification and optimization on large models with no performance bottleneck.

Contact Us

Unstructured grids geostats

Unstructured grids geostatistics and gridless geostatistics

With grid blocks of various geometric shapes and sizes, unstructured grids offer a better representation of geological complexities seen in deposits. The current standard of building block models with rectangular cuboids forces simplifying engineering concepts onto geological modeling processes. Unstructured grids separate the geologist's needs for flexible modeling frameworks from the regularity of engineering grids.

Extending beyond unstructured grids, AR2Tech is actively developing gridless geostatistics. Gridless properties exist outside of any discrete gridding system, only the locations are required not the actual grid. This technology allows for adaptive workflow where the property resolution can be modified in real time in critical area.

  • Tetrahedral meshes
  • PEBI grid
  • Y-faults, pinch-outs, non-neighbor connections
  • Unstructured topology, positioning and volume
Contact Us

Uncertainty quantification, Optimization, and Machine Learning

AR2Tech has developed a toolkit to create a portfolio of geostatistical models by sampling a space of uncertain geostatistical parameters. This approach uses experimental design concepts to create diverse models and evolutionary optimization to generate models that minimize an objective function.

AR2Tech’s Python interface is compatible with industry-standard Machine Learning algorithms for supervised and unsupervised classifications, and regression algorithms. Problems such as facies determination from ancillary data can be solved with machine-learning algorithms using large training data sets.

Contact Us

Geostatistics for embedded systems (IoT)

The AR2GAS library is able to operate on a micro PC (such as a Raspberry Pi) to provide advanced spatial algorithms to embedded systems. This is in active development.

Please contact us if you would like to discuss a project for spatial statistics on embedded systems, be it on medical applications, meteorology, robotics, logging or real-time measurements. We are open to all types of applications.

Contact Us

Geostatistics for World's leading Oil & Gas Companies

While many software packages aim to offer an integrated, all-in-one solution, we specialize in geostatistics.

Advanced geomodeling

Our solutions have been implemented by integrated oil & gas companies that chose not to be limited by standard commercial solutions.

Create your workflows

We understand the role of geostatistics from data to performance forecasts, and we know how to build customized workflows designed for uncertainty quantification, stochastic history-matching, and field development optimization.

Expertise at your fingertips

Our goal is to deliver the most advanced geostatistical toolbox for your projects. We develop new algorithms that meet the industrial demand created by increasing complexity.

Mining and Orebody Modeling

Advanced Resources and Risk Technologies provides software and solutions for conditional simulation and estimation models of complex orebodies. We work with major mining companies to develop the toolkits needed to standardize and optimize their resource estimation workflow. We go beyond the geostatistical algorithms offered in commercial packages, select and further customize the algorithms best-suited to specific deposits.

We have developed a proprietary suite of algorithms and utilities specific for mining applications such as contact analysis, downhole contamination analysis, twin hole analysis, change of support functionalities and grade-tonnage curves, plus several more. We have also implemented direct block simulation algorithm that allows the generation of large models in a more efficient fashion.

  • Multivariate simulations on point or block support,
  • Simulation of geology with indicator or training images,
  • Customization of algorithms and automation,
  • Drill holes placement optimization.

Environmental Sciences

Advanced Resources and Risk Technology understands the challenges inherent to environmental data and applications. We extract values out of data at hand and design schemes to optimizing the information value of current or future data acquisition. From a specific problem we design a customized workflow from data processing to spatial evaluation to meaningful risk assessment and then provide software applications to provide functional autonomy to our clients.

  • Stochastic simulation of subsurface for water modeling,
  • Sampling evaluation and strategy,
  • Integration of remote sensing data in geostatistical model,
  • Spatio-temporal monitoring,
  • Characterization of pollutant.