# Resource Modeling Solutions > Geology modeling, resource modeling and mine planning software, training and support for the mining industry: RMSP (a Python package for geology modeling, resource estimation and simulation, mine planning and drill hole optimization) and AMP (a secure, collaborative cloud platform for running RMSP workflows, where teams and AI agents work on the same versioned data, models and workflows). Resource Modeling Solutions (RMS) is a team of geologists, geostatisticians, mining engineers and software developers, and a division of GeologicAI. Dr. Jared Deutsch is President. Dr. Ryan Barnett, Vice President of Operations, oversees consulting and training. Dr. Clayton Deutsch, Professor at the University of Alberta and director of the Centre for Computational Geostatistics, is an advisor. RMS has offices in Canada (Edmonton, Calgary, Toronto), the United States (Denver), Australia (Brisbane), South Africa (Johannesburg) and Chile (Santiago). RMSP (Resource Modeling Solutions Platform) is the modeling engine: deterministic and probabilistic methods in one automated, auditable Python package, covering implicit geology modeling, exploratory data analysis, variography, kriging, simulation, pit optimization across realizations, and infill drill hole optimization. It runs locally, in your own cloud or in AMP, and is used by mining and consulting companies around the world. AMP (Automatic Modeling Platform) runs RMSP in a cloud workspace with a resource tree, 3D viewer and notebooks. Data, models and scripts are versioned together, workflows run from data loading through model export with one click, and access is set by project and region. AI agents work in AMP through ChatRMS and the RMS MCP server. The RMSP API documentation is not public. Agents should reach it through the RMS MCP server described below, not by crawling the documentation site. ## Products - [RMSP](https://resourcemodelingsolutions.com/rmsp): the Python modeling engine for geology modeling, resource modeling, mine planning and drill hole optimization - [AMP](https://resourcemodelingsolutions.com/amp): collaborative, versioned cloud platform for RMSP workflows, built for practitioners, reviewers and AI agents ## Agent access to RMSP documentation The RMSP documentation at https://portal.resourcemodelingsolutions.com/documentation requires a login, so do not crawl it. A logged-out, non-browser request for a documentation page is redirected to a plain-text pointer to the RMS MCP server. Use the RMS MCP server instead: it provides version-aware RMSP documentation search, API inspection and example discovery, and read-only access to your RMS portal account information. - Endpoint: https://portal.resourcemodelingsolutions.com/api/mcp/rmsp/v1 (MCP Streamable HTTP transport, stateless, JSON-RPC over POST; OAuth sign-in) - Access: an RMS portal account with documentation and software access - Set up: add the endpoint to your MCP client as a remote (Streamable HTTP) server named `rms-mcp` - Sign-in is discovered automatically: an unauthenticated request returns 401 with a `WWW-Authenticate` header pointing to https://portal.resourcemodelingsolutions.com/.well-known/oauth-protected-resource/api/mcp/rmsp/v1, and the client registers itself through OAuth dynamic client registration. The user signs in with their RMS portal account when the client first connects. ## Training & Support - [Training](https://resourcemodelingsolutions.com/training): public short courses in geostatistics and resource modeling, Python and RMSP workshops, in-house courses and the affiliated University of Alberta Applied Geostatistics Certificate; exercises are completed in RMSP - [Course registration](https://portal.resourcemodelingsolutions.com/courses): register for public courses and workshops - [Overview of training opportunities (PDF)](https://resourcemodelingsolutions.com/pdfs/Overview%20of%20Training%20Opportunities%202026.pdf): current year's courses and formats - [Geostatistics short courses schedule (PDF)](https://resourcemodelingsolutions.com/pdfs/Geostatistics%20Short%20Courses%202026.pdf): short course dates, pricing and details - [Python and RMSP workshop (PDF)](https://resourcemodelingsolutions.com/pdfs/Python%20Fundamentals%20and%20Estimation%20with%20RMSP%202026.pdf): workshop dates, pricing and details - [Support](https://resourcemodelingsolutions.com/support): consulting and support for geology modeling, resource modeling and mine planning, from workflow troubleshooting to technical advisory ## Company - [Team](https://resourcemodelingsolutions.com/team): senior consultants and staff - [Client stories](https://resourcemodelingsolutions.com/client-stories): testimonials from RMSP users - [Careers](https://resourcemodelingsolutions.com/careers): open positions for geologists, geostatisticians, mine engineers and software developers - [Contact](https://resourcemodelingsolutions.com/contact): contact form for training, software and modeling project enquiries - [News](https://resourcemodelingsolutions.com/news): product releases, resources and company news - [Blog](https://resourcemodelingsolutions.com/blog): articles on geostatistics and resource modeling - [Geostatistics Lessons](https://geostatisticslessons.com/): guidance on the application of geostatistics, founded and edited by Dr. Jared Deutsch of RMS - [Centre for Computational Geostatistics](http://www.ccgalberta.com/): University of Alberta research centre directed by Dr. Clayton Deutsch - [User portal](https://portal.resourcemodelingsolutions.com/): RMS account sign-in, RMSP documentation and course registration ## Geostatistics articles - [Application of Simulation](https://resourcemodelingsolutions.com/blog/post/Application-of-Simulation): Simulation is becoming more widely used for mineral resource estimation. There is great interest in quantifying geological uncertainty and understanding geological risk. Let’s review some applications… - [Kriging is not a Linear Estimate](https://resourcemodelingsolutions.com/blog/post/Kriging-is-not-a-Linear-Estimate): Kriging has long been known as the Best Linear Unbiased Estimate, that is, BLUE. That sounded catchy in the early days of geostatistics, but now it just sounds quaint. Also, in the modern era of… - [Drill Hole Spacing and Value of Information](https://resourcemodelingsolutions.com/blog/post/Drill-Hole-Spacing-and-Value-of-Information): A drill hole spacing study (DHSS) and a value of information study (VOIS) are different. A DHSS quantifies uncertainty in tonnage, grade and metal of production volumes of relevance and relates that… - [Resource Modeling Problems](https://resourcemodelingsolutions.com/blog/post/Resource-Modeling-Problems): Geostatistics as a discipline and Resource Modeling Solutions (RMS) as a division of GeologicAI do not provide solutions looking for problems. Critical problems in resource modeling have driven the… - [Probabilistic Criteria for Measured, Indicated and Inferred Resources](https://resourcemodelingsolutions.com/blog/post/Probabilistic-Criteria-for-Measured-Indicated-and-Inferred-Resources): Mineral resources should be disclosed according to measured, indicated and inferred. This is legally required and makes practical sense. Probabilistic criteria and all relevant mitigating factors… - [Changing Input Parameters Does Not Assess Uncertainty](https://resourcemodelingsolutions.com/blog/post/Changing-Input-Parameters-Does-Not-Assess-Uncertainty): Uncertainty has a very clear meaning and is based on probabilistic concepts going back over 300 years. Uncertainty is not accurately quantified by accident, or by tweaking hyperparameters of a… - [The Variogram Range](https://resourcemodelingsolutions.com/blog/post/The-Variogram-Range): The variogram range is commonly considered to be the distance at which the variogram reaches the sill variance. This blog highlights that the variogram range is a very ambiguous parameter. It is… - [The Block Size in Open-Pit Mining](https://resourcemodelingsolutions.com/blog/post/The-Block-Size-in-Open-Pit-Mining): There is no need for the block size in open pit mining to match the selective mining unit (SMU) size. The smoothness of block estimates depends more on the kriging search than on the block size, so the block size should be a compromise between data spacing, geological contacts, engineering design and computational efficiency, with sub-blocks to preserve detail along contacts. - [The Selective Mining Unit (SMU) in Open-Pit Mining](https://resourcemodelingsolutions.com/blog/post/The-Selective-Mining-Unit-(SMU)-in-Open-Pit-Mining): The Selective Mining Unit (SMU) is a nominal regular block volume that represents the scale of production that will be achieved in future mining. The vertical size of the SMU is almost always the… - [Challenges with the Parker Challenge](https://resourcemodelingsolutions.com/blog/post/Challenges-with-the-Parker-Challenge): Competitions can be useful to determine best practices, stress test legacy techniques, investigate new alternatives and determine the circumstances where certain techniques work well. In numerical… - [Probabilistic Modeling of Underground Resources](https://resourcemodelingsolutions.com/blog/post/Probabilistic-Modeling-of-Underground-Resources): Probabilistic modeling of open pit resources is well established; simulate at high resolution, average to a selective mining unit (SMU) size that represents future mining operations and grade control information, then report expected resources. This blog addresses probabilistic modeling of underground resources - [Use Drill Hole Spacing for Mineral Resource Classification](https://resourcemodelingsolutions.com/blog/post/Use-Drill-Hole-Spacing-for-Mineral-Resource-Classification): The mining industry is reliant on disclosing mineral resources and reserves to investors to demonstrate value and secure capital to bring new mines into production. Reporting codes have evolved to… - [Resource Classification Challenges, Insights and Innovations](https://resourcemodelingsolutions.com/blog/post/Resource-Classification-Challenges-Insights-and-Innovations): Practical challenges of Mineral Resource classification, following on from Use Drill Hole Spacing for Mineral Resource Classification. Classification criteria should be transparent and kept separate from flagging categories to blocks, simplified geometries avoid isolated patches, and two approaches refine classification - flagging composites by drill hole spacing and other criteria, and classifying underground stope optimization shapes. - [Optimal composite length for estimating block grades](https://resourcemodelingsolutions.com/blog/post/Optimal-composite-length-for-estimating-block-grades): Mineral resource estimation is based on block models of mineral grades within geological domains, with the block size chosen to reflect site-specific geological factors and mining selectivity. Estimated grades are typically calculated using ordinary kriging with composited drill hole data from the geological domain. Sample lengths are often relatively small for high resolution geological information, but different vintages may exist with varying sample lengths. Compositing is recommended to ensure consistency and improve block estimates. The optimal composite length is a critical decision, as too short or long lengths can increase errors due to high variability or loss of resolution, respectively. - [Note Regarding Modeling Geospatial Uncertainty with Bayesian Models](https://resourcemodelingsolutions.com/blog/post/Note-Regarding-Modeling-Geospatial-Uncertainty-with-Bayesian-Models): The paper “Modeling Geospatial Uncertainty of Geometallurgical Variables with Bayesian Models and Hilbert–Kriging” by Hoffimann and others draws attention to the important topic of multivariate… - [To Impute or not to Impute](https://resourcemodelingsolutions.com/blog/post/To-Impute-or-not-to-Impute): Multiple imputation of missing measurements in multivariate geological data, such as legacy assays and expensive geometallurgical samples. Covers the nature of missingness, how imputation draws realizations by combining collocated and spatial conditional distributions, and when cokriging or the stepwise conditional transform are better alternatives. - [The Importance of Training](https://resourcemodelingsolutions.com/blog/post/The-Importance-of-Training): Why experienced and early career resource modelers both need ongoing training in modern geostatistics, and what good training looks like - a balance of theory, implementation and practical application, built around a coherent workflow that is independent of software. - [Scenarios versus Parameter Uncertainty](https://resourcemodelingsolutions.com/blog/post/Scenarios-versus-Parameter-Uncertainty): An important goal of modern resource modeling is accurate and precise assessments of uncertainty for production volumes relevant for technical and economic decision making. Volumes of relevance often… - [Geostatistics with Nonlinear Variables](https://resourcemodelingsolutions.com/blog/post/Geostatistics-with-Nonlinear-Variables): Many rock properties like mass fractions, volume fractions, thickness, and position average linearly. For rock properties that average nonlinearly, one rock type would have more influence on the effective property. This is common with variables such as recovery and throughput. In special circumstances, the effective property can go outside the range of the two pure components. This would depend on specific chemical reactions or increased efficiency of a mixture relative to a pure component, for example, performance of a semi-autogenous grinding (SAG) mill. - [Resource Calculation with Incomplete Information](https://resourcemodelingsolutions.com/blog/post/Resource-Calculation-with-Incomplete-Information): In most cases, mining will benefit from significantly more data in the future than available at the time of resource calculation. There may be one hundred times more data at the time of mining from… - [Trends in Geostatistics](https://resourcemodelingsolutions.com/blog/post/Trends-in-Geostatistics): A trend in geostatistics could be a general tendency in the direction industry is taking, like toward automation. A trend in geostatistics is also a non-stationary or locally-varying mean in a… - [Resource Modeling as an Essential Service](https://resourcemodelingsolutions.com/blog/post/Resource-Modeling-as-an-Essential-Service): Mining provides critical resources and products that society needs, but what of resource modeling within the mining sector? A critical analysis of what we do in resource modeling suggests that the cost of a precise and accurate quantification of resource uncertainty is modest with appropriate software and trained professionals. The benefits are far greater than the cost, raw data are turned into support information for important short- and long-term decisions... - [The Gaussian Distribution in Geostatistics](https://resourcemodelingsolutions.com/blog/post/The-Gaussian-Distribution-in-Geostatistics): Gaussian algorithms have infected virtually all corners of geostatistics: radial basis functions (RBF) for interpolation and implicit modeling, hierarchical truncated pluriGaussian (HTPG) for… - [Public Disclosure of Probabilistic Resource Estimates](https://resourcemodelingsolutions.com/blog/post/Public-Disclosure-of-Probabilistic-Resource-Estimates): The state of nature is uncertain. Geological variability at all scales and sparse sampling make resource uncertainty an inevitable reality for mining companies and the concerned public. The workflow… - [Probabilistic Resource Estimation](https://resourcemodelingsolutions.com/blog/post/Probabilistic-Resource-Estimation): Geostatistical simulation of mineral deposits for resource estimates and long-range planning is established. Professionals have access to the methodology, software and training to reliably calculate… ## Optional - [Outlier Management: Global and Local Capping](https://resourcemodelingsolutions.com/news/outlier-management-capping): Apply global or local capping, with maximum percentile and maximum mean impact methods, and assess the impact with built-in statistics. - [RMS Tutorial Playlists on YouTube](https://resourcemodelingsolutions.com/news/rms-youtube-playlists): Free video playlists covering geological modeling, estimation, variography, data processing, Python, AMP and RMSP fundamentals. - [Locally Varying Anisotropy](https://resourcemodelingsolutions.com/news/locally-varying-anisotropy): Derive local angles from meshes, structural data, dense data or gridded models, then use them for LVA variography and local searches. - [Geology Modeling in RMSP](https://resourcemodelingsolutions.com/news/geology-modeling-in-rmsp): Signed distances, the RBF estimator, surface, contact, offset and vein modelers, geologic sequences and structural plane data for complete geological models. - [Automated Report Writing](https://resourcemodelingsolutions.com/news/automated-report-writing): Generate consistent, auditable reports from your RMSP workflow, with templates, variables, tables, figures, references and a table of contents built in. - [Estimation Calibration](https://resourcemodelingsolutions.com/news/estimation-calibration): Define estimation scenarios, run them across all domains and variables at once, and compare results with metrics, grade-tonnage curves and CDFs. - [Vein Simulator](https://resourcemodelingsolutions.com/news/vein-simulator): Simulate vein position, thickness and boundaries, then grades in flattened space, for probability maps and uncertainty at mine planning scales. - [Rotation Convention Conversion](https://resourcemodelingsolutions.com/news/rotation-conventions): RMSP uses the GSLIB angle convention and converts to and from Datamine, Isatis, Leapfrog, Micromine, Minesight and Vulcan conventions in a few lines. - [Companion Notebook: Cokriging with Unequal Supports](https://resourcemodelingsolutions.com/news/geostatistics-lessons-notebook-cokriging): A full cokriging workflow with blasthole and drill hole data, compared with ordinary kriging and an intrinsic model of coregionalization. - [Companion Notebook: Rotation Conventions](https://resourcemodelingsolutions.com/news/geostatistics-lessons-notebook-rotation-conventions): A Python companion notebook on angle specification and rotation matrices in 2D and 3D, supporting the Geostatistics Lesson on angle specification. - [Companion Notebook: Transforming Data to a Gaussian Distribution](https://resourcemodelingsolutions.com/news/geostatistics-lessons-notebook-gaussian-transform): A Python companion notebook for the Geostatistics Lesson on the normal score transform, covering despiking and distribution smoothing. - [RMSP 1.17 Release Highlights](https://resourcemodelingsolutions.com/news/rmsp-1-17-release): Automated reporting, RBF interpolation with partition of unity, an ore/waste intervals tool, solid operations, distance calculations and new tutorials. - [Trend Uncertainty](https://resourcemodelingsolutions.com/news/trend-uncertainty): Assess trend uncertainty with spatial bootstrap realizations and integrate it into simulation so that uncertainty is not understated. - [Data Checking: Take the Garbage Out](https://resourcemodelingsolutions.com/news/data-checking): Check collar, survey and assay data as soon as it is loaded, report flagged issues and investigate questionable drill hole paths. - [Spatial Pairing Analysis](https://resourcemodelingsolutions.com/news/spatial-pairing-analysis): Compare two data sets with paired statistics, test search ranges, locate the largest differences and judge whether differences are significant. - [RMSP 1.16 Release Highlights](https://resourcemodelingsolutions.com/news/rmsp-1-16-release): Data checking tools, trend uncertainty, swath plot enhancements, nested pits and scheduling, drill hole planning and resource classification tools. - [Local Search Optimization](https://resourcemodelingsolutions.com/news/local-search-optimization): Use more data where more data is available, without specifying multiple estimation passes, and compare scenarios with block-by-block performance metrics. - [ModelCompare: Reconciliation Made Easy](https://resourcemodelingsolutions.com/news/modelcompare-reconciliation): Reconcile long-term and production models by bench, cutoff, domain or category, with grade-tonnage curves, statistics, sections, swath plots and CDFs. - [Fast Regularization and Validation in RMSP](https://resourcemodelingsolutions.com/news/fast-regularization-and-validation): Regularize numerical and categorical variables for 11 million sub-blocks in about two seconds, then validate with sections and swath plots. - [RMSP 1.15 Release Highlights](https://resourcemodelingsolutions.com/news/rmsp-1-15-release): Maximum intensity projection, optimized pit meshes and solids, bench compositing, table merging, categorical model clean-up and a convex hull tool.