Batch analysis — IndPenSim-Batch

IndPenSim penicillin fermentation (public benchmark) · prepared by Yahya Nazer, CTRL Designer LLC · 11 September 2026

1. What this is

A batch process cannot be judged the way a continuous plant is. Every batch runs to a different length and nothing about it is steady, so "normal" is not a window of time but a set of good batches. These reports build a golden batch from batches known to be good, stretch every other batch onto the same 0-100% progress axis, and measure how far each one strays from it.

Why there is more than one report. Each is built around a different golden batch. A golden made from every good batch is the best general detector; a golden made from one way of running the plant answers whether that way is genuinely different from the others. Section 9 is the same in all of them and is where the strategies are compared.

2. The reports

Report	What it holds	Size
IndPenSim-100 E-Batch/IndPenSim-100/batch_report.html	100 batches · golden from 90 (every batch with no recorded fault) · 10 of 10 recorded faults caught	51 KB
IndPenSim-Golden-Recipe E-Batch/IndPenSim-Golden-Recipe/batch_report.html	100 batches · golden from 20 (batches 1-20) · 10 of 10 recorded faults caught	74 KB
IndPenSim-Golden-Operator E-Batch/IndPenSim-Golden-Operator/batch_report.html	100 batches · golden from 20 (batches 31-50) · 10 of 10 recorded faults caught	72 KB
IndPenSim-Golden-APC E-Batch/IndPenSim-Golden-APC/batch_report.html	100 batches · golden from 20 (batches 61-80) · 10 of 10 recorded faults caught	74 KB

Charts are live: drag to zoom, double-click to reset, click a name in the legend to hide it.

3. Questions and comments

Every finding is a question until the plant answers it. If something is flagged that you know ran well, that is the feedback that makes the next pass better.

Email info@ctrlDesigner.com

Yahya Nazer · CTRL Designer LLC