The three things process engineers hate most about their simulation software
Over the last few months I’ve had 300+ customer discovery conversations with process engineers. Gas-processing consultancies, oil sands contact engineers, a metallurgy shop, EPC veterans, a nuclear-services team. Different industries, different simulators, same question from me: what actually eats your week?
Somewhere around call fifty the answers stopped surprising me. Everyone hates the same three things. This isn’t a survey with checkboxes. It’s what came up over and over in my call notes, anonymized.
Out now
Try Reflux Student free on Aspen Plus
Reflux Student drives your own Aspen Plus V14 from plain English. You type what you want, it opens the case, makes the change in Aspen, runs it, reads the result back and tells you what it verified. Aspen does the math. You keep the judgment. Windows, your own licence, and free to start is a trial with a limit on it, not a free product.
Enter your email on the next page and the download link lands in your inbox, so you can open it on the Windows machine Aspen lives on. Windows will say it does not recognise the app the first time: click More info, then Run anyway. The certificate is new and Windows trusts it by reputation, which takes downloads to build. Nothing is wrong with the file.
Not on Aspen Plus? Tell me which simulator you use → and you hear the day your build exists.
The setup grind
One engineer told me about the week a heat-transfer fluid got flagged as illegal in Canada, and he had to swap it across 17 streams. One stream at a time. Another types petroleum assay cut points off a PDF into the simulator, cell by cell. A third imports lab analyses with 38 components and renames every one by hand so the software will recognize them.
Nobody asked me to automate the thinking. They want the thinking. It’s the data entry wrapped around it they hate. A consulting engineer put a number on it: 30 to 50 percent of simulation hours are recoverable manual labor, not engineering judgment.
It won’t converge, and it won’t say why
A nuclear-services engineering team told me their biggest day-to-day pain isn’t building models at all. It’s convergence. Add a recycle loop and the solver goes unstable, and the error that comes back tells you almost nothing about the actual cause.
A 35-year refinery veteran said the same thing from the other end of a career: reading stream data is basic, converging a distillation column is the fight. A greenfield crude tower still takes one to two weeks, and a lot of that is guess, tweak, rerun, wait.
Licensing that belongs in 1998
Aspen runs $20K to $50K per seat per year. I heard that exact range on multiple calls. Below the big packages it gets stranger. Simulators locked to USB dongles. One tool licensed to a single physical laptop. Plants rationing token-based seats until access itself becomes the bottleneck.
The detail that got me: consultants building models their own clients can’t open, because the client doesn’t hold a license. And after all that money, one firm described vendor support as a two-day wait for a callback that doesn’t help.
Reflux exists because of these calls
Nobody asked me to automate the thinking. They want the thinking. It is the data entry wrapped around it that eats the week, and that is the part a machine should be doing. Reflux drives the simulator you already have from plain English: it takes the setup grind, helps you chase down convergence, and every change it makes is read back and verified before it is reported.
We’re opening a small trial to people who ask for it, and we are not precious about which software you run. Whatever you use, we’ll build for it.
Nathan Ruberto · Co‑Founder, CEO
Back to reflux.sh ->