Compare the numbers
Compare costs, capacity, throughput, and other measures across scenarios. Time series show how those results develop over the course of a simulation.
BUSINESS SIMULATION / HUMAN + AI
nMODL2
Explore a business decision before making it for real.
I am building nMODL2 so a person and an AI can work with the same model of a business. Either can write a scenario, run the simulation, and examine what happened. The simulation is deterministic: the same model and inputs produce the same results.
It has its own modeling language, a command-line interface, an API, an MCP interface, and skills for AI users. Those are different ways into the same system. A person can set the question, an AI can explore alternatives, and both can inspect the assumptions and results.
01Run the models
Business modelScenario A
Business modelScenario B
02Compare A and B
NumbersNumeric resultsWhat changedWhen
nMODL2Semantic contextHowWhy
03Explore possible strategies
NumbersCompare results
nMODL2Reason about alternatives
The numbers tell part of the story.
A simulation can tell you what happened to costs, capacity, or throughput. I also want to know what happened to the people, roles, and work inside it. nMODL2 produces numerical results and semantic accounts of those changes, giving an AI more to work with than a table of outcomes.
Examine what happened
The semantic output preserves the modeled causes, constraints, dependencies, and tradeoffs behind the results. An AI can use that context to reason about why the simulated futures differ.
What could you learn from a hundred scenarios?
Imagine setting an objective at the end of the day and leaving an AI to explore it. It could write and run a scenario, compare the results with the starting model, and use what it found to design the next one. By morning, it might have worked through a hundred.
I want you to come back to alternatives you would not have had time to prepare by hand, with enough information to examine the promising ones. This is the kind of workflow I am building toward; a result inside a simulation still needs judgment before it becomes a real business decision.
A CHANGE-MANAGEMENT EXAMPLE
8,000 → 7,000
A thousand fewer people. What happens to their work?
Suppose an organization is considering a reduction from 8,000 people to 7,000. A simple model can subtract the salary costs. That tells us very little about how the remaining people are supposed to do the work.
With nMODL2, an AI could explore reductions of 50 people at a time, reroute the work, and test different arrangements at each step. It could compare the modeled performance, workload shifts, and risks, then use those findings to try another arrangement.
The goal would be several possible paths, with the reasoning behind each. Leadership could examine what the model predicts before deciding whether, or how, to make the change.
Give the decision-maker something to work with.
I want the output to be a small set of strategies a person can examine: what each would cost, what might go wrong, what would change in the work, and which results support it. An unexplained recommendation is not enough.
The larger ambition is to make part of what people buy from expensive change-management consulting available as repeatable software. I am developing nMODL2 as an open-source platform for that purpose. People set the objective and make the decision; AI takes on much more of the scenario-building and comparison needed to get there.
Work with me
Tell me what you have in mind.
I build websites and apps, provide AI training, and consult with businesses and organizations getting AI into their work. Tell me what you want to do and what has been getting in the way.