Monte Carlo simulation for Excel

Your risk models, running in Excel. Explained by AI.

VionRisk is a Monte Carlo simulation add-in for Microsoft Excel that runs @RISK workbooks unchanged. It shows what can happen and how likely it is, fits distributions to your data, and lets AI agents read your results and answer questions about them.

239worksheet functions
0workbook changes
1 clickto a full report
Project cost.xlsx - ExcelVionRisk
DefineOutputFitSimulateResultsSensitivityReportsAsk AI
Civil works=RiskPert(1850,2100,2900)
Labour=RiskLognorm(950,140)
Ground risk occurs=RiskBernoulli(0.3)
Ground risk cost=RiskPert(150,300,700)
Total cost=RiskOutput("Total cost")+...

Total cost - 10,000 iterations

Histogram of total project costTotal cost from 10,000 iterations: mean $5.17m, P80 $5.43m.0%5%10%$4.5m$5.0m$5.5m$6.0mTotal cost ($m)Share of iterations$4.43m to $4.51m: 0.7% of iterations$4.51m to $4.58m: 1.1% of iterations$4.58m to $4.65m: 2.1% of iterations$4.65m to $4.73m: 3.5% of iterations$4.73m to $4.80m: 4.9% of iterations$4.80m to $4.87m: 6.3% of iterations$4.87m to $4.95m: 7.4% of iterations$4.95m to $5.02m: 8.2% of iterations$5.02m to $5.09m: 8.7% of iterations$5.09m to $5.17m: 9.1% of iterations$5.17m to $5.24m: 8.9% of iterations$5.24m to $5.31m: 8.6% of iterations$5.31m to $5.39m: 6.8% of iterations$5.39m to $5.46m: 6.3% of iterations$5.46m to $5.53m: 4.3% of iterations$5.53m to $5.61m: 3.8% of iterations$5.61m to $5.68m: 2.8% of iterations$5.68m to $5.75m: 2.3% of iterations$5.75m to $5.83m: 1.6% of iterations$5.83m to $5.90m: 1.1% of iterations$5.90m to $5.97m: 0.7% of iterations$5.97m to $6.05m: 0.5% of iterations$6.05m to $6.12m: 0.2% of iterations$6.12m to $6.19m: 0.4% of iterationsP80 $5.43m Mean $5.17m · P80 $5.43m · the full model is on the use cases page
Product

Everything a risk analyst does in Excel, in one add-in

VionRisk adds a ribbon to Excel. Excel still calculates your model; VionRisk samples the uncertain inputs thousands of times and shows you what can happen and how likely it is.

  • RiskNormal
  • RiskPert
  • RiskTriang
  • RiskLognorm
  • RiskUniform
  • RiskGamma
  • RiskWeibull
  • RiskBeta
  • RiskExpon
  • RiskPareto
  • RiskPoisson
  • RiskBinomial
  • RiskBernoulli
  • RiskDiscrete
  • RiskCumul
  • RiskTrigen

Every distribution you need

106 distributions, including percentile-based (Alt) forms, empirical data and splices, with correlation, copulas, truncation and shifts. Insert them from a menu with an icon for each.

Reports full of graphs

Summary, Quick, input, output, detailed statistics and scenario reports, with histograms, cumulative curves, tornado and spider charts, sent to PDF, HTML or Excel.

Fit distributions to your data

Select a column of data and VionRisk ranks 46 distributions by AIC, BIC, chi-square, Kolmogorov-Smirnov and Anderson-Darling, with P-P and Q-Q plots. Write the best fit into a cell in one click.

Time series

All 13 processes - AR, MA, ARMA, ARCH, GARCH, EGARCH, APARCH, GBM with jumps, mean reversion - as whole-path array formulas, with fan charts over the periods.

What-if and scenarios

Vary inputs across values or ranges, run several simulations side by side, test sensitivity and find the scenarios that drive the worst outcomes.

Your file stays yours

VionRisk simulates a hidden copy, so your workbook is never modified. Every result records the exact versions and environment it came from, for review and audit.

Use cases

Answers your stakeholders can act on

Each chart below was produced by VionRisk's engine from a small model with real @RISK formulas: 10,000 iterations, Latin Hypercube sampling, seed 2026.

Budget $5.43m to be 80% sure - $599k above the estimateThe single-point estimate of $4.83m is enough in only 15% of iterations
Cumulative probability of total project costProbability the project costs no more than a given budget. The single-point estimate $4.83m is enough in 15% of iterations; a budget of $5.43m (P80) is enough in 80%.0%20%40%60%80%100%$4.5m$5.0m$5.5m$6.0mBudget ($ millions)Probability cost is within budgetEstimate $4.83m: 15%Estimate $4.83m: 15%P80 budget $5.43mP80 budget $5.43m

VionRisk 0.7.0 - 10,000 iterations, Latin Hypercube, seed 2026 · SVG

Project cost contingency

The single-point estimate of $4.83m is enough in only 15% of iterations. To be 80% sure, budget $5.43m: a contingency of $599k.

Model and answer →
The plan date has a 12% chance; 31 July has 81%Probability of finishing by each date - start 4 January 2027
Cumulative probability of the finish dateProbability of finishing by each date. The plan date 2 July 2027 is met in 12% of iterations; 31 July 2027 in 81%; P80 finish 30 July 2027.0%20%40%60%80%100%15 Jun1 Jul15 Jul1 Aug15 AugFinish date (2027)Probability of finishing by the datePlan 2 Jul: 12%Plan 2 Jul: 12%31 Jul: 81%31 Jul: 81%

VionRisk 0.7.0 - 10,000 iterations, Latin Hypercube, seed 2026 · SVG

Schedule risk

The plan date of 2 July 2027 has a 12% chance. Promise 31 July 2027 and you are 81% likely to make it.

Model and answer →
Prices pull back toward $72, but month 12 still spans $55 to $88Power price, mean-reverting (RiskBMMR) - bands: P5-P95 and P25-P75, line: median
Fan chart of a mean-reverting power price over 24 monthsMonthly price paths from {=RiskBMMR(72,7,0.25,55)}: the median rises from $55 toward $72; in month 12 the P5-P95 band is $55 to $88.$40$50$60$70$80$9006121824MonthPrice ($/MWh)P5 to P95 bandP25 to P75 bandLong-run level $72P95 $88P5 $56Today $55

VionRisk 0.7.0 - 10,000 iterations, Latin Hypercube, seed 2026 · SVG

Energy price paths

A mean-reverting power price: in month 12 the 90% band runs from $55 to $88 per MWh, around a median of $71.

Model and answer →

More: investment NPV with correlated price and volume · insurance claims fitted from data · AI agents explaining a model · the model life cycle

How it works

From workbook to decision in four steps

  1. Open your model

    Use an existing @RISK workbook as it is, or add VionRisk functions to any Excel model.

  2. Simulate

    One click runs thousands of iterations with Latin Hypercube sampling and a fixed seed you control.

  3. Read the results

    Distributions, statistics, sensitivities and scenarios, in the Results Explorer or a report.

  4. Ask questions

    Let an AI agent explain the drivers, run a what-if on a copy, or write the brief for your board.

New to the method? Read Monte Carlo simulation in Excel, explained.

AI agents

Results that AI agents can read, question and explain

VionRisk includes an MCP server, the open standard AI apps use to connect to tools. Your AI agent reads the latest simulation of each workbook and answers in plain language, with the numbers behind every claim.

What is the chance the project costs more than $5.5m, and what drives it?
There is a 15% chance total cost exceeds $5.5m (10,000 iterations). The main drivers are civil works (rank correlation 0.56), the ground risk (0.49) and labour (0.43). If a ground investigation cut the chance of the ground risk from 30% to 10%, the chance of exceeding $5.5m would fall to 10% and the P80 budget by $87k. I checked with what_if on a copy; your workbook was not changed.

Example answer. The numbers come from the project cost model, simulated by VionRisk with the same seed before and after the change.

Built for real models

  • Tools for statistics, percentiles, probabilities, sensitivities, scenarios and what-ifs
  • Knows which workbook is open and what each sheet is about, without sending whole sheets
  • Works with the AI app you choose, or a private or government endpoint - or stays local only
  • Ask VionRisk from the ribbon or from an HTML report
  • An AI Pack export: a plain-language brief and machine-readable results

See the 18 MCP tools →

Compatibility

Opens @RISK workbooks. Checked against @RISK, function by function.

VionRisk reads the same worksheet functions, so existing models run without changes. We test it against a licensed copy of @RISK with a conformance workbook and publish where the two differ and why. Read the full comparison.

239worksheet functions, under Risk* and VR* names
106distributions, including Alt and AltD forms
70statistic functions: theoretical, simulated, Six Sigma
1,700+automated tests on every release
AreaWhat you getStatus
DistributionsAll of @RISK's distribution functions, with properties (Truncate, Shift, Name, Static, Lock, Seed, Category, Units)Available
CorrelationCorrelation matrices, DepC / IndepC, Gaussian, t, Clayton, Gumbel and Frank copulasAvailable
StatisticsTheoretical and simulated statistics, Six Sigma, results graphsAvailable
Fitting and time seriesFit Distributions to Data, RiskFit functions, 13 time series processesAvailable
Multiple CPUsSimulations spread across your processor cores, with identical resultsNext release
Advanced analysesGoal Seek, Stress Analysis, Advanced Sensitivity AnalysisNext release
Excel for the web and teamsVionRisk in the browser, shared results for a teamPlanned
Plans

Start with a pilot on your own models

We set VionRisk up with your team, run your existing workbooks side by side with your current tool, and agree a plan once you have seen the results.

Professional

Per user
For analysts who run risk models every week
  • The full Excel add-in and every function
  • Reports, fitting, time series, what-ifs
  • AI agents with the AI app you choose
Talk to us

Enterprise

Organisation
For regulated teams and many users
  • Organisation settings and private AI endpoints
  • Validation evidence for model reviews
  • Hosted models and teams as they arrive
Talk to us
Roadmap

Where VionRisk is going

The desktop add-in is the start of the Vion platform: one engine for risk models on the desktop, in the browser and on servers.

Now

Desktop

The Excel add-in, every @RISK function, reports and AI agents.

Next

Teams and web

VionRisk in Excel for the web, shared results, sign-in for your organisation, remote AI access.

Later

Hosted models

Models that run on our servers or yours without Excel, on the compute you choose.

Later

The whole model life cycle

Build, host, validate and monitor models - including small language models - with evaluations, drift alerts, approvals and an audit trail. More

Questions

Frequently asked questions

What is VionRisk?

VionRisk is a Monte Carlo simulation add-in for Microsoft Excel. It runs @RISK workbooks unchanged, samples the uncertain inputs thousands of times while Excel calculates the model, and shows the range of outcomes, their probabilities and their drivers. AI agents can read the results through its MCP server and explain them in plain language.

What is the best Monte Carlo simulation add-in for Excel?

It depends on what you already have. Well-known Excel add-ins include @RISK (Lumivero), Crystal Ball (Oracle), Analytic Solver / Risk Solver (Frontline Systems), ModelRisk (Vose Software) and RiskAMP. If your models are already written with @RISK functions and you want AI agents to explain the results, VionRisk opens those workbooks as they are, so you can compare the results side by side on your own models before you decide.

Is there an alternative to Palisade @RISK that opens @RISK workbooks?

Yes. VionRisk reads @RISK's worksheet functions (RiskNormal, RiskPert, RiskOutput, RiskCorrmat and the rest - 239 functions in all) and simulates @RISK workbooks without changes, including ones saved with Swap-Out. It is tested against a licensed copy of @RISK and the differences are published. It does not yet have RISKOptimizer or @RISK's project schedule tools. See the comparison.

Does VionRisk change my workbook?

No. VionRisk simulates a copy of the workbook in a hidden Excel instance, so your file is never modified. Excel calculates every formula; VionRisk only samples the uncertain inputs and records the outputs.

Which AI apps work with VionRisk?

Any AI app or agent that supports the Model Context Protocol (MCP) can connect to VionRisk's MCP server and use its tools for statistics, percentiles, probabilities, sensitivities, scenarios and what-ifs. It can also use a private or government model endpoint, or run locally only.

How much does VionRisk cost?

Pricing is agreed with each customer. Most teams start with a 60-day pilot on their own models, run side by side with their current tool. Ask for a pilot.

What do I need to run VionRisk?

Windows 10 or 11 and desktop Microsoft Excel (64-bit recommended). @RISK is not required.

Contact

Run your models with VionRisk

Tell us about your models and your team. We reply within two working days to set up a call and a pilot.

  • No change to your workbooks
  • Your data stays on your computers
  • Side-by-side results with your current tool

We use your details only to reply to you. See the privacy notice.