We build simulation software and AI agents for production lines and supply chains, for manufacturers that include many Fortune 500 operators. The work pairs artificial intelligence with operations research to explore the efficient frontier for your problem.
ReliaSim models production lines, plants and supply chains from your own stop data and checks the model against what the plant actually did. A customer’s published paper reports the model within about 1% OEE of the plant. Then you test the change on the model before you try it on the plant.
Fit failure and repair patterns from your stop history with ReliaStats®.
Find the real bottleneck and rank which stops are worth fixing, including where a bigger buffer pays.
Discrete rate simulation for high-speed and bulk flow, the method Andrew Siprelle created in 1990.
Network design and safety stock on the same engine, including tariff scenarios.
“We leveraged the model to recommend solutions that saved $21.7M in inventory carrying costs and reduced transit time by 19%.”
ReliaSim is the main one; each of the others handles one kind of problem. Most have a free sandbox, and six also run inside AI assistants such as Claude and ChatGPT.
In “How data-based simulations find the fastest wins,” Tom Lange and Andrew Siprelle talk about the paradox of averages, and why your Pareto can point you at the wrong failures.
A line that won’t hit its number, say, or a network decision coming up. Tell us what you’re up against and a person will get back to you.