ON-THE-FLY PALLETIZING
Viroteq delivers AI on-the-fly palletizing for production lines that refuse to stand still. The runtime recomputes pallet patterns mid-cycle, adapts to changing SKU mixes in real time, and recovers gracefully when production is disrupted. As a result, on-the-fly palletizing cells keep stacking through every variant change, sensor blip, and order reshuffle — without operator intervention, without robot re-teaching, and without scripted downtime between products.

Real-time recompute
Per-case decision latency
Mixed SKUs supported
Live production tested
Live production is messy. Cases arrive out of sequence, an SKU drops mid-shift because of a packaging stockout, a sensor reads a carton dimension that does not match the master data, and the order list reshuffles after a customer call-off. Conventional palletizing software, which depends on a known SKU sequence and a pre-validated pallet pattern, simply stalls in this environment. The line stops, an operator intervenes, and throughput drops. On-the-fly palletizing exists precisely because the production floor never matches the spreadsheet.
However, raw flexibility alone is not enough. On-the-fly palletizing must compute decisions inside the cycle window of a real production line — typically under 100 milliseconds for cadences of 600 to 1200 cases per hour. The AI engine has to balance pattern density, weight stacking rules, overhang limits, and pallet stability all at once, every single placement, every single time. Anything slower, and the robot waits on the software instead of the other way round.
Furthermore, on-the-fly palletizing has to recover from sensor noise without halting. A misread barcode, a partially occluded carton, a vision frame dropped due to lighting — all are routine on a live floor. The runtime degrades to the most reliable input available, recomputes from the new state, and continues building the pallet. According to industry guidance from the Material Handling Industry of America, real-time adaptability is now a baseline expectation for end-of-line automation rather than a premium feature. Viroteq’s edge-first technology platform closes the gap with deterministic response and graceful degradation built into the core runtime.
As a result, on-the-fly palletizing wins exactly where pre-programmed stacking falls over: co-packers running ten variants per shift, FMCG producers with seasonal SKU rotation, e-commerce fulfilment lines with order-driven sequencing, and beverage plants juggling promotional packs alongside core ranges. One software stack, one operator workflow, every production reality.


On-the-fly palletizing begins the moment the next case arrives at the cell. Vision sensors and dimensioning scanners feed live geometry to StackrBrain, which compares the actual carton to the SKU master record. If the case matches, the AI updates the pattern. If the case is unexpected — wrong dimensions, missing data, or a substitute pulled by the line — the AI recomputes the remaining pallet from scratch in under 100 milliseconds.
Next, RobotStackr OTF issues the placement command to the robot controller. Communication runs over REST API and WebSocket, so the robot brand is irrelevant — FANUC, ABB, KUKA, Universal Robots, Yaskawa, and Stäubli all work natively. The robot lands the case, sensors confirm placement, and on-the-fly palletizing immediately computes the next move.
Additionally, the runtime carries pattern state forward across the entire pallet. Stability rules, weight distribution, and overhang constraints are evaluated continuously, not just at start. If a previous placement was slightly skewed, on-the-fly palletizing adjusts the next placement to compensate. As a result, the finished pallet is stable enough for a 1500 km truck ride, even though no two cases followed the same path through the cell. For complementary inbound automation, RobotDepalr handles depalletizing of supplier loads with the same vision-driven approach.
Three purpose-built products cover the on-the-fly palletizing scope — from real-time mixed-SKU recalculation to consistent single-SKU lines and inbound supplier depalletizing. All three share the StackrBrain AI engine, run on the same Industrial PC hardware, and integrate with major MES, WMS, and PLC platforms via REST API without bespoke development.

RobotStackr OTF is the dedicated runtime for on-the-fly palletizing. Every pallet is computed mid-cycle in under 100 ms — no pre-sequencing, no manual programming, no downtime between SKUs. Ideal for high-mix consumer goods, co-packing operations, and order-driven fulfilment lines.

RobotStackr OS powers consistent single-SKU production-line palletizing with optimal pre-computed patterns. Ideal companion to on-the-fly palletizing for plants that mix uniform high-volume runs with variable mixed-SKU sequences from one shared runtime.

RobotDepalr automates inbound supplier pallet handling with the same vision-driven AI used in on-the-fly palletizing. 3D vision identifies layers on incoming pallets and singulates components onto a conveyor for line-side feed. Closes the loop on full plant automation.
On-the-fly palletizing recalculates the remaining pallet pattern at any point during the build. When an unexpected SKU arrives, an order updates, or a sensor flags a deviation, the AI generates a new placement plan in under 100 milliseconds without pausing the robot or interrupting the line.
Live vision and dimensioning data drive every placement decision. On-the-fly palletizing fuses inputs from 3D cameras, barcode scanners, and weight cells, then degrades gracefully when one source fails. As a result, sensor noise and partial outages no longer halt the line or trigger operator escalation.
On-the-fly palletizing accepts any sequence of cases without pre-programming. Mixed cartons, promotional packs, substitutes, and rush items all flow through the same cell. Therefore, co-packers, FMCG plants, and e-commerce fulfilment lines avoid the planner cycles and robot re-teaching that dominate legacy stacking workflows.
On-the-fly palletizing pays back fastest in environments where production variability, mixed SKUs, and order-driven sequencing dominate. The six segments below represent the highest-value applications across the Viroteq customer base. Furthermore, all six share a single runtime and a single operator HMI, which keeps the engineering burden inside the plant team. As a result, on-the-fly palletizing scales from a single co-packer cell to a multi-line FMCG plant without parallel software stacks.
Contract packers running ten to twenty SKUs per shift across multiple brand owners. On-the-fly palletizing eliminates per-product setup time and absorbs late changes from clients without scripted downtime.
Fast-moving consumer-goods plants with seasonal SKU rotation and frequent promotional pack changes. On-the-fly palletizing sequences the live order list without robot re-teaching between variants.
Order-driven fulfilment lines where pallet contents reflect the live customer queue. On-the-fly palletizing pulls each next case from the order and stacks it without a fixed pattern, supporting parcel-grade shipping density.
Beverage producers running multipacks, multipack variants, and promotional bundles alongside core SKUs. On-the-fly palletizing handles the size and weight variation without per-pack pattern programming.
Industrial bakeries with daily mix changes for retail and food-service customers. On-the-fly palletizing accommodates short-run product schedules and absorbs last-minute sequence reshuffles per outbound truck.
Distribution-centre outbound where each pallet matches a specific store, route, or customer order. On-the-fly palletizing reads the live shipping list and stacks accordingly, no per-store pre-programming required.

Contract packing line running fifteen variants per shift for three brand owners, each pallet mapped to a specific customer order with full traceability metadata.

FMCG line switching between core packs, multipacks, and seasonal promotional bundles every two hours, with no robot re-teaching or planner downtime between rotations.

Distribution-centre outbound matching each pallet to a specific store route, with sequence pulled from the live shipping list and stacked without per-store pattern templates.
Per-case recompute time
Cases per hour throughput
Typical deployment timeline

Live-production environments share one defining trait: the schedule never matches reality. A co-packer signs a brand owner on Friday and runs the new SKU on Monday. A consumer-goods plant pulls forward a promotional bundle to capture a retail window. An e-commerce fulfilment line sequences pallets directly from customer orders that arrive minute-by-minute. In every case, the pre-programmed pallet pattern is obsolete before the first case lands. On-the-fly palletizing exists to absorb that variability without slowing the line.
For co-packers, on-the-fly palletizing collapses the per-product setup window. Where legacy systems require a robot re-teach and a pattern validation cycle for each new SKU, RobotStackr OTF ingests dimensional data, pulls the order list, and starts stacking. As a result, contract packers absorb client changes in real time rather than overnight, and the cell utilisation rate rises sharply across the shift.
FMCG and beverage plants face a similar challenge in a different shape. Seasonal SKU rotation, promotional pack changes, and multipack variants create a constantly shifting product mix. On-the-fly palletizing pulls each next case from the live order list and stacks it without a fixed pattern, so a Monday core run looks the same to the operator as a Friday promotional sequence. Furthermore, brand-agnostic robot support means plants with mixed FANUC, ABB, KUKA, and Universal Robots fleets keep one software stack across cells. Pair the deployment with the broader palletizing solutions portfolio for full plant coverage.
On-the-fly palletizing integrates with your existing line through three modern protocols. REST API is the primary front door for order, batch, and pallet data — well-documented and easy for IT teams to reason about. WebSocket carries live cycle events, sensor feedback, and operator alerts at low latency. OPC-UA bridges the runtime to plant historians, MES, and SCADA without ripping out the data architecture you already have. Furthermore, native PLC handshakes for Siemens S7, Rockwell ControlLogix, and Beckhoff TwinCAT keep the controls team in full ownership of the cell.
As a result, on-the-fly palletizing deployments coexist with installed fieldbus, ladder logic, safety relays, and conveyor PLCs — no proprietary middleware, no replacement of working infrastructure. Robot brand support spans FANUC, ABB, KUKA, Universal Robots, Yaskawa, and Stäubli. In addition, the runtime sits on Industrial PCs inside the cell, so latency is bounded at the controller, not the cloud, and the line stays operational during external network outages or audits.
Standard request/response over HTTPS for order, batch, and pallet data exchange.
Low-latency event streaming for real-time cycle data, sensor feedback, and HMI updates.
Bridge to plant historians, MES, and SCADA dashboards via the industrial data standard.
Native handshakes for Siemens S7, Rockwell ControlLogix, and Beckhoff TwinCAT.
Book a personalised demo and see how on-the-fly palletizing absorbs SKU variability, recovers from sensor noise, and keeps your line running through every production change.
Bring your line specs, SKU mix profile, and PLC stack — Viroteq specialists will map an on-the-fly palletizing deployment that fits your existing cell footprint and keeps the line running through every production change.
| Cookie | Duration | Description |
|---|---|---|
| cookielawinfo-checkbox-advertisement | 1 year | Set by the GDPR Cookie Consent plugin, this cookie records the user consent for the cookies in the "Advertisement" category. |
| cookielawinfo-checkbox-analytics | 11 months | This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Analytics". |
| cookielawinfo-checkbox-functional | 11 months | The cookie is set by GDPR cookie consent to record the user consent for the cookies in the category "Functional". |
| cookielawinfo-checkbox-necessary | 11 months | This cookie is set by GDPR Cookie Consent plugin. The cookies is used to store the user consent for the cookies in the category "Necessary". |
| cookielawinfo-checkbox-others | 11 months | This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Other. |
| cookielawinfo-checkbox-performance | 11 months | This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Performance". |
| CookieLawInfoConsent | 1 year | CookieYes sets this cookie to record the default button state of the corresponding category and the status of CCPA. It works only in coordination with the primary cookie. |
| elementor | never | The website's WordPress theme uses this cookie. It allows the website owner to implement or change the website's content in real-time. |
| viewed_cookie_policy | 11 months | The cookie is set by the GDPR Cookie Consent plugin and is used to store whether or not user has consented to the use of cookies. It does not store any personal data. |
| Cookie | Duration | Description |
|---|---|---|
| _lscache_vary | 2 days | Litespeed sets this cookie to provide the prevention of cached pages. |
| Cookie | Duration | Description |
|---|---|---|
| CONSENT | 2 years | YouTube sets this cookie via embedded YouTube videos and registers anonymous statistical data. |
| _ga | 1 year 1 month 4 days | Google Analytics sets this cookie to calculate visitor, session and campaign data and track site usage for the site's analytics report. The cookie stores information anonymously and assigns a randomly generated number to recognise unique visitors. |
| _ga_* | 1 year 1 month 4 days | Google Analytics sets this cookie to store and count page views. |
| Cookie | Duration | Description |
|---|---|---|
| VISITOR_INFO1_LIVE | 5 months 27 days | YouTube sets this cookie to measure bandwidth, determining whether the user gets the new or old player interface. |
| YSC | session | Youtube sets this cookie to track the views of embedded videos on Youtube pages. |
| yt.innertube::nextId | never | YouTube sets this cookie to register a unique ID to store data on what videos from YouTube the user has seen. |
| yt.innertube::requests | never | YouTube sets this cookie to register a unique ID to store data on what videos from YouTube the user has seen. |