Food World Summit 2026 · Grapevine, TX · Oct 5–7

Every quality decision
has a throughput price.

A longer hold-and-test. A tighter allergen changeover. An extra metal-detect reject loop. Each one costs you cases — and almost nobody can put a number on it before they commit. ChiAha builds a validated working copy of your line so you can.

Book a meeting at the booth Try the free sandbox →
Validated within 1% of measured OEE · 300+ organizations since 1995 · 35 years in production simulation
Dates
October 5–7, 2026
Venue
Gaylord Texan, Grapevine TX
Find us at
Booth TBA
Bring
Your line’s stop/start log
Why We’re at This Show

Safety and quality get decided
without a throughput number.

Food and beverage manufacturing runs on a standing argument. Quality and food safety own the protocol; manufacturing owns the case count. When the two collide — a longer hold, a wider sanitation window, a stricter reject rule — the trade-off is settled by whoever argues best, because nobody in the room can say what the protocol actually costs in cases per shift.

01 / The claim

“That protocol will kill the line.” Usually asserted, rarely measured. Sometimes right, sometimes wildly overstated — and there is no way to tell which without running it.

Who says it: Manufacturing · Plant leadership
02 / The counter

“We can absorb it.” Backed by a spreadsheet of averages. But a real line does not run on averages — machines wait on each other, and one stop starves the next.

Who says it: FSQ · Quality · Compliance
03 / The answer

Run the change on a validated working copy of the line first. Same machines, same speeds, same breakdowns — a full year of production in seconds, before you commit to anything.

What it takes: Data you already record
The Technical Reason This Is Hard

A food line isn’t a parts line.
Most simulation treats it like one.

Your process is a hybrid. Upstream is continuous — mixing, extruding, proofing, frying, a spiral freezer holding hours of product in transit. Downstream is discrete — fill, seal, case-pack, palletize. Conventional discrete-event simulation was built for the discrete half and approximates the continuous half badly, which is exactly where food lines hide their losses.

Conventional Discrete-Event Simulation
Models every unit as a discrete entity
Bulk and continuous stages approximated
Work-in-process inside a freezer or proofer is invisible
Days to set up, hours per scenario
Discrete Rate Simulation
Continuous flow and discrete packaging in one model
Bulk stages modeled as flow, not approximated
Pipeline fill and drain represented explicitly
Minutes to set up, seconds per scenario

ChiAha’s founder developed bulk-flow discrete rate simulation in 1990 for precisely this class of process. It is the reason a food line is our home ground rather than an awkward fit.

From a current engagement · National frozen-foods manufacturer

The model kept coming in under standard. On a frozen-foods line we were modeling, simulated output sat consistently below the plant’s standard rate — not drifting, not noisy, but short by the same amount every single run.

A constant gap is a signature. A rate error would scale with run length; this did not. The missing product was sitting in the spiral freezer — work-in-process that has entered the line but not yet come out the other end. The plant’s standard, computed as a rate, has no way to represent product in transit. The simulation does, because it models the freezer as a real stage with real residence time.

Neither number was wrong. They were answering different questions, and only one of them accounts for the line actually filling up. That distinction is worth real money at changeover and startup — and it is completely invisible to a spreadsheet standard.

Come to the booth and we will show you the run.

At the Booth

Four things you can do
in fifteen minutes.

No slideware. Every one of these is a working demonstration you can run yourself, and the last one you can keep.

2 minutes
New to simulation? Start here.
A short primer, in plain English, on what a simulation of a production line actually is and why averages cannot answer the questions you are asking. No modeling background assumed — most of this room does not have one, and does not need one.
5 minutes
Watch a line get built from a stop/start log.
Your line already records when each machine stopped and started. That is the raw material — nothing new to install, no IT project. We build a model in front of you and validate it against measured OEE.
5 minutes
See why gain never equals loss.
Your loss tree ranks failure modes by downtime. Eliminate the top one and you rarely recover what it said you would — sometimes far less, sometimes far more. We eliminate each failure mode in turn, re-simulate the whole system, and measure the throughput actually recovered. The ranking changes.
3 minutes
Price a quality or food-safety change.
Extend a hold. Tighten a reject rule. Widen the sanitation window. Add a changeover step for allergen control. Watch what each one does to cases per shift on a validated model — and take the answer back to the argument you are currently losing.
Reserve a booth slot

Slots run 20 minutes and fill first. Walk-ups welcome at Booth TBA

Who Should Stop By

If you own a number
on a production line.

This is a capital and operating decision tool, not a monitoring dashboard. It is most useful to the people who have to defend a forecast.

Food Safety & Quality leadership

You are asked to justify protocols in throughput terms and have no model to do it with. Bring the change you are fighting for — leave with its cost in cases per shift.

Manufacturing & Operations

You are ranking capital by downtime, which is the wrong measure. See what your line actually recovers when each failure mode is removed — before the capital request goes in.

Operational Excellence & CI

Your loss tree tells you where time went, not where throughput comes back. Simulation closes that gap, and it runs on data your plants already keep.

Supply Chain & Planning

Standard rates ignore pipeline fill, changeover interaction, and buffer dynamics. Size buffers and validate a line’s real capacity before committing to a plan.

Engineering & Capital Projects

Validate a greenfield or expansion line before it is built. A thousand year-long scenarios in under a minute, on your own hardware, behind your own firewall.

Anyone new to simulation

Most of this show is. Start with the two-minute primer and a free browser sandbox — eight ready-made lines, nothing to install, no license required.

~1%
OEE Accuracy
15 min
To a Working Model
300+
Organizations Since 1995
35+
Years Expertise
Tom Lange
"Those 120 one-minute interruptions are creating cascading problems throughout your system that don't show up under the original problem's name. When you eliminate them, you often recover 180–220 minutes of uptime — significantly more than the original downtime suggests."
Tom Lange
36 years, Procter & Gamble · Retired Director, Modeling & Simulation, Corporate R&D
Co-author, "High Accuracy Discrete Rate and Reliability Modeling" (Winter Simulation Conference 2020)
Before October 5

The booth slots
go first.

Twenty minutes, on your line, with your numbers. Send us a stop/start export ahead of the show and we will have something built by the time you walk up — or just book the slot and bring the question.

Book a booth meeting Open the free sandbox

Can’t make the show? Book a live walkthrough instead →