Inspection and traceability that survive washdown and an audit
Willowark builds inspection and traceability systems for food and beverage production: fill-level and headspace checks, cap and seal verification, label placement, date and lot code reading, and software that ties a finished pallet back to the raw material lots inside it. The problem it removes is the one that keeps plant managers up — a recall you cannot bound, where the safe answer is pulling far more product than was ever affected.
We design for the sanitation cycle first, because that is what destroys equipment here. Cameras and sensors go into IP69K stainless enclosures with sloped tops and no ledges, cabling is rated for caustic and high-pressure hot water, and optics get air purge so a lens does not fog on the first wet clean. Then comes the inspection: real production samples, pass and fail criteria written down, and a false-reject rate low enough that operators do not learn to bypass it.
Most engagements here start with one line and one problem the plant already knows it has: a code reader that is not being trusted, a checkweigher whose data goes nowhere, or a mock recall that took too long. We survey the line during a sanitation cycle, not just during production, because the enclosure and cable decisions get made there. The first build is deliberately narrow, sized so it can be commissioned during a scheduled changeover rather than a lost shift, and it produces records that the QA team can hand to a customer auditor unchanged.
Reviewed
Sound familiar?
If you've said any of these, we should talk.
“If we had a recall tomorrow, I could not tell you which pallets to pull.”
We build lot genealogy that scans at each handoff: raw material lot at receiving, batch record at mixing, fill time window, case code, and pallet. A recall query returns the exact pallet list and which customers received them, with supporting records attached.
“We give away product on every fill just to stay above the label weight.”
We capture checkweigher data continuously and trend it by filler head rather than as a pass or fail. Where the filler supports it we close the loop against measured drift; where it does not, we alarm the heavy head before a full shift ships overfilled.
“Date codes get missed and we hear about it from the customer.”
A camera reads the printed code with OCR and verifies it character by character against the code issued for that run, not just that ink is present. Failures reject automatically, with a downstream sensor confirming the container actually left the line.
“The QA binder is the system. If an auditor wants March, we go digging.”
We replace the binder with electronic records captured where the check happens: CCP readings, sanitation sign-offs, swab results, and CIP cycle data, each timestamped and attributed to a user. An audit request becomes a filter, not an afternoon in the file room.
“We cannot keep people on the line long enough to train them on the checks.”
Turnover is a design input here. We automate the checks the equipment can make and guide the ones people still do: the screen shows what to sample, what the limit is, and will not accept a skipped step. A new hire follows the same procedure on day one that the veteran did, and the record proves it.
How this industry actually runs
The operation as we usually find it.
Food plants run on sanitation as much as on production. A line runs a shift, then takes a wet clean with caustic and sanitizer at temperature and pressure that ruins anything not rated for it. Changeovers are allergen events: a run containing peanut, milk, egg, soy, or wheat must be followed by documented cleaning and verification before the next SKU starts. FSMA 204 pushes lot-level traceability down to key data elements captured at every critical tracking event, so a recall has to be boundable in hours. Margin lives in giveaway — overfilling every container to stay legal on net weight is a real annual number, decided by how tightly the checkweigher and filler talk to each other.
Machine signals to the people who decide
Components:
- PLCs & sensors (counts, states, current)
- Legacy machine (dry contact / clamp)
- Edge gateway (normalize, buffer)
- Production dashboard (downtime, OEE)
- Alerts & reports (who acts, when)
Connections:
- PLCs & sensors to Edge gateway (EtherNet/IP, Modbus)
- Legacy machine to Edge gateway
- Edge gateway to Production dashboard (MQTT)
- Edge gateway to Alerts & reports
What we build
Starting projects that fit Food & Beverage.
- Washdown-rated vision stations for fill level, cap and seal integrity, and label placement
- OCR and OCV date and lot code verification with confirmed reject
- Lot traceability from receiving through batch, fill, case, and pallet with FSMA 204 data capture
- Checkweigher and metal-detector data capture with SPC trending and giveaway reporting
- Allergen changeover workflows with sanitation checklists, swab results, and electronic sign-off
- CIP monitoring logging time, temperature, conductivity, and flow per circuit
- HACCP critical control point monitoring with automated alarms and retained records
- Line monitoring and downtime capture for fillers, cappers, and case packers with reason codes by SKU and changeover
Capabilities we bring
What we'd build first
Three places a first project usually starts.
01
Washdown-rated vision stations for fill level, cap and seal integrity, and label placement
02
OCR and OCV date and lot code verification with confirmed reject
03
Lot traceability from receiving through batch, fill, case, and pallet with FSMA 204 data capture
Scoped in writing before any work starts — how engagements work.
Working in Food & Beverage?
Tell us the line.
What runs by hand, what is not connected, what you are trying to build. An engineer replies within one business day with whether and how we would approach it.
Common questions
What Food & Beverage teams ask first.
Can a vision system really handle daily washdown?
Yes, if it is specified for it from the start: IP69K stainless housings, connectors and cable jackets rated for your chemistry, air purge on windows, and mounting the sanitation crew can clean around without aiming a wand at a seal. Retrofitting a dry-environment camera into a wet room is what fails.
Does FSMA 204 mean we need to buy new software?
Not necessarily. The rule cares about capturing key data elements at critical tracking events and producing them quickly on request. If your ERP already holds most of it, the gap is floor-level capture and fast genealogy queries, a much smaller build than replacing your system of record.
Our product varies naturally. Will inspection reject good product?
It will if it is tuned on ideal samples. We image real production across the range you actually ship, including seasonal and supplier variation, and build tolerance windows around that. False-reject rate is measured before handover and treated as an acceptance criterion.
Can you integrate with the checkweighers, metal detectors, and coders we already have?
Usually. Most current checkweighers, metal detectors, X-ray units, and coders expose data over Ethernet, serial, or a documented protocol, and older ones typically give at least a reject signal and a count we can capture at the controller. What we get from each unit varies by make and firmware, so we confirm it against your actual equipment during the survey rather than promising it from a spec sheet.
How do we install anything without losing production?
Most of the work happens off the line. Enclosures, mounts, and cabling are built and tested in advance, and software is commissioned against recorded samples before it sees live product. The on-line portion is typically scheduled into a sanitation window or a planned changeover, and inspection systems usually run in a monitor-only mode for a period before the reject is enabled, so the line never depends on an untested system.
Strategy. Software. Systems.
Engineering for Food & Beverage.
Describe the problem in your own words. An engineer reads it — not a sales script — and tells you plainly what it would take.