The product

RASQI Line Simulator

"How much will this line actually make?"

A simulator for serial and assembly lines. It adds the losses a real line has — breakdowns, setups, scrap, a station waiting for parts or held up by a full buffer — so the answer is what the line will deliver, not what the cycle times promise.

How we check the engine

~2% from known queueing-theory results, tested in every run

Automatic tests also check the laws a line must obey: stations in series add up, parallel machines take the maximum, variability never creates capacity.

This checks the engine, not your data. The result is as good as the numbers you enter.

What it models

The line, its losses, the statistics, the money

The line

  • Up to 49 stations, plus a line feeder
  • Parallel identical machines, fed by committed load the way a line PLC does — not blind rotation
  • Assembly with component feeder lines: the assembly waits when a component cannot keep up
  • Finite buffers with real blocking and starvation
  • Conveyors (incl. accumulating), fork and merge, returnable carts
  • Line-level WIP cap (CONWIP)

What gets in the way

  • Seven distributions fitted to your data, per station or per product
  • Breakdowns by MTBF and MTTR, per cycle or per time
  • Scrap and at-station rework
  • Setups and campaigns, with a sequence-dependent changeover matrix
  • Shared operators or robots with travel time between stations
  • Shift calendar and breaks, kept apart from availability losses
  • Optional advanced reliability (Weibull, micro-stops, wear). Off by default

The statistics

  • Check line before every run: what would make the result meaningless, and how to fix it
  • Paired A→B scenarios on the same random numbers, with the payback of the change
  • Replications with 95% confidence intervals in A→B and DOE; warm-up discarded
  • Factorial DOE on up to six factors (half fraction at five or six)
  • Improvement study: your two biggest losses and the constraint's cycle, at three levels, at the line's own horizon. The gains do not add up, and the report says so

The money

  • Loss waterfall in $/year, split into availability, performance and quality
  • Diagnosis ranked by value, each with a TOC, Lean, 6σ or TPM action
  • Payback, ROI, ROIC, NPV, IRR and EVA
  • One-page summary to print; CSV and Excel round-trip
  • Demand mode: sells only what the market buys, pays only the hours needed
  • Prices, margins and demand are never sent to the server

Rewind the run and watch the same jam again

Play from 1× to 300×, pause, drag back — in the 2D diagram, in 3D, or both side by side on one clock. Lay the stations over a photo of your plant if you like. Before recording, the app tells you how many hours fit your connection.

The boundary, before you pay

What it does not do, and why

Job-shop and changing routes

Each product follows the route you declare, and may return to a station on it. A path that changes piece by piece, or on a test result, is out.

Queue priority other than FIFO

It changes the lateness of each order, not throughput, OEE or $/year (Little's Law). That is scheduling, a different job.

Sub-second lines, piece by piece

Cycles down to 1 ms are accepted, but a run's time grows with the pieces it makes. Model the batch cycle instead. Below 5 s the advanced micro-stops switch off, and the app says so.

AGV fleets and tooling

Conveyor and cart transit are modeled. Fleet dispatch, and tools or fixtures as a limited resource, are not.

Adding up several improvements

The improvement study prices one change at a time; the gains do not add. Make one change, then run again.

The full list is open

Every limit, with its reason, on the capabilities page. Not sure about your line? Answer 9 questions or tell us about it.

Where it stops

It gives you the number. Your finance function approves the case.

The app says so on its own screen: the financial page is an engineering estimate for screening — no working capital, financing, sales taxes, inflation, FX, ramp-up or residual value.

Prove it on your own line first

One free month with the real software. If it does not find money, do not buy it.