K↗KASYNETSOPERATIONAL INTELLIGENCE
Engineering prototypeKA
LEGACY SAMPLE PROJECTION · A short traffic sample supplies average access costs to a separate shift calculation. This is not a continuously simulated shift. Open the coupled Picking Lab →

The picking floor.Down to the pallet.

30 picking trucks. 15 voice. 15 scanner. WMS hours, unfinished work, high-bay putaway and shared aisles.

SIMULATION PROTOTYPE

Synthetic WMS orders and stock, four synchronized congestion scenarios, and assumed device/lift timings. No production system is connected.

Exact counts · schematic scale
Sorter picking
100rows
44,000pallet positions · 4 levels
Classic picking
50rows
22,000pallet positions · 4 levels
Each row
110 × 4slots
440positions across all levels
Total layout capacity
66,000slots
55 | 55cross-aisle after position 55
INTERACTIVE 3D FLOOR

Sorter picking ROW S001

ROW DETAIL · S001
DETAILED GEOMETRY
Drag to orbit · Shift-drag to pan · Scroll to zoom · Select a row
110 positions / level · L1–L4 · split after 55
Picker / movingPedestrianHigh-bay / corridorsHold / STOP6 / 3km/h · proposedAll 150 rows exist in the model
Row S001 · positions 01–110 · all four levelsSchematic dimensions · draft traffic plan
SELECTED ROW / ALL FOUR LEVELS

S001 · 440 pallet positions

55 + 55
POSITIONS 01–55MAIN CORRIDORPOSITIONS 56–110
Click a position to inspect its address.S001-001-L1
GEOMETRY CHECK

Built from your floor brief

Sorter picking100 rows × 110 × 4 = 44,000
Classic picking50 rows × 110 × 4 = 22,000
Every row01–55 / corridor / 56–110
STOP signs3 per row · 450 in the model
Proposed speed limits6 km/h corridors · 3 km/h aisles
Active fleet30 pickers · 40 pallets + 20 Bins

110 is interpreted as positions along each level. No measured aisle widths or rack heights were supplied; this is a layout review, not a certified traffic plan.

PHASE 01 / LAYOUT SPECIFICATION

A precise count.
A schematic footprint.

The current geometry follows the latest picking-zone brief. It replaces the earlier small illustrative rack layout.

100 sorter + 50 classic rows

Rows S001–S100 and C001–C050 are modeled individually. Their side-by-side zone arrangement is provisional.

110 positions on each of 4 levels

This means 440 locations per row and 66,000 total. Location addresses identify zone, row, position, and level. Stock boxes are visual examples, not inventory records.

A real break after position 55

Each rack is built as two separate 55-position sections. The central corridor is continuous across all 150 rows. A front corridor serves both zones.

Traffic signage is in the model

STOP signs are placed at the front and on both sides of the central corridor. Ground arrows indicate two lanes in main corridors and provisional alternating directions in row aisles. Detailed signs become visible when you zoom in or inspect a row.

Speed limits: proposed demo policy

6 km/h in main corridors, 3 km/h in picking aisles and on junction approaches. STOP still requires a full stop before joining. These values are selected for this prototype, not prescribed by a universal European warehouse standard. Real limits depend on the site risk assessment, visibility, vehicle and load, pedestrian separation, and manufacturer instructions. Fleet speed uses schematic distances and simulated time; it is not a calibrated real-world vehicle model.

Guidance: KomNet / NRW workplace safety authority ↗ · EU-OSHA: in-house transport ↗. These sources support risk-based site rules, not the specific 6/3 values.

Dimensions need confirmation

Pallet pitch, aisle widths, rack heights, zone separation, and vehicle dimensions use schematic proportions. No exact warehouse footprint or regulatory compliance is claimed. Traffic animation is illustrative and does not validate safe movements or clearance.

WMS hours and interrupted work

The shift event model offers editable WMS hours per picker: 7.4 means 7 hours 24 minutes, with presets for 7, 6 and 5 hours. Its default workload is 600 jobs / 12,000 lines. Linked mode runs a fresh traffic sample for the selected dispatch, driving rules and high-bay demand. For each zone, observed picker time minus WMS service is divided by confirmed lines to obtain the travel/access/wait cost charged to shift lines. This assumes the sampled flow continues; it is a stationary sample projection, not a continuous full-shift 3D simulation or a measured forecast. Interrupted jobs do not feed back into the traffic sample. Stalled or zero-confirmation samples pause the projection. Manual mode uses an independent editable average travel/access time. Both modes use the WMS device timing and packaging assumptions.

0, 5 or 10 selected pickers interrupt their first job after 10 of 20 confirmed lines, then start a fresh job on fresh carriers. Confirmed goods stay on the original carriers in handover staging; unpicked lines remain reserved at source. Manual policy leaves the remainder awaiting release; recovery policy queues it for the next available picker in the same zone, with configurable delay and carrier collection time. The ledger records every original line ID, old/new owner, interruption, reassignment and completion time. Unconfirmed work at shift end remains backlog; no quantity is deleted or counted twice. These are simulated policies, not verified behavior of the real WMS.

Operational congestion / high-bay putaway

The pressure study releases the same 600 picking lines and high-bay putaway demand into four synchronized runs: concentrated versus distributed WMS dispatch, each with and without provisional LOOADS driving controls. Congested dispatch sends the selected number of pickers to S039 and C015. High-bay trucks use those same rows to place pallets at levels L2–L4, with adjustable lift/place/lower time. Pickers and high-bay trucks can queue behind each other.

Entry capacity is six active vehicles per loop, with FIFO staging and re-entry each circuit. Transfers between staging and loop entry are abstracted. Physical lift exclusion applies in both driving modes; upper-level putaway is separate from reserved ground-level picking inventory. Output, backlog, vehicle waiting, complete orders, putaways and geometric overlaps are counted at a matched chosen deadline. These are simulated operational scenarios, not site measurements or validated productivity predictions.

Zone-specific quantities and physical carriers

Only sorter picking mixes outer tasks (full packages onto a pallet) and inner tasks (individual pieces into a large open-top plastic Bin). Classic picking has only individual inner tasks, alternating between Euro and non-Euro pallets; there are no outer tasks in classic. WMS confirmation validates the target carrier type before changing stock. Current carrier loads show confirmed packages or pieces.

All carried load carriers have a 1200 × 800 mm footprint. The Bin shape references a manufacturer plastic pallet box, 760 mm high ↗. The Euro footprint follows EPAL dimensions ↗. The non-Euro pallet has the same footprint by your brief, with a different wood construction. No certification, load capacity, stacking plan or real SKU pack specification is assumed. Visible contents are capped for clarity and do not represent a validated packing layout.

Synthetic WMS: 15 voice + 15 scanner

PK01–PK30 alternate between voice and scanner, with 15 operators per cohort, balanced within the 20 sorter and 10 classic pickers. Assignments can be swapped for a counterbalanced replay. Each cohort receives 300 order lines: 300 outers and 600 individual pieces. The wave has 60 carrier orders and 600 lines. At the default assumed pack size of 12 inners per outer, this is 8,400 base units overall, or 4,200 per cohort. Changing the pack size recalculates the stock equivalents.

Tasks contain an assigned location, SKU, quantity type, base-unit equivalent, order and target carrier. Verification starts only after the matching physical pick stop is reached. The modeled sequence is verify location/SKU, pick the physical quantity, then confirm the quantity. A rejected check must retry before inventory can be deducted. Initial stock covers all reservations plus a small synthetic reserve. Stock and task-state conservation are shown in the console.

Voice and scanner verification both default to 3 seconds. Physical picking uses 8 seconds + 1 second per handled outer or individual inner, with 2 seconds for quantity confirmation. A configured rejection adds a 5-second retry and another verification. These are editable assumptions, not vendor benchmarks. There is no built-in voice productivity boost. Timing or repeat-scenario changes reset both driving runs.

Voice output is a browser speech preview of a visible transcript, when supported. Responses and barcode checks are simulated; there is no microphone, speech recognition, physical scanner, production WMS, customer order import or real inventory connection. Channel cohort outcomes also include route and traffic differences and are not an isolated hardware trial.

30 picking trucks: explicit concept rules

PK01–PK30 each carry two tandem load carriers on low long forks behind the driver. Twenty sorter pickers each carry an outer pallet and plastic Bin. Ten classic pickers each carry a Euro pallet and a non-Euro pallet of the same 1200 × 800 mm footprint: 40 pallets and 20 Bins total. Both runs use identical initial states, requested cruise speed of 6 km/h, pick tasks and pedestrian arrival intent. Active loops use ground-level positions in the front 55-position rack sections, between the two existing main corridors. Rear-section routing is outside the active fleet simulation. High-bay putaway at levels L2–L4 is included in the pressure study.

With LOOADS (concept) enforces 6/3 km/h limits, 3 km/h turning, 1.5-second STOP checks, vehicle/pedestrian holds, and full-body footprint checks that include both load carriers. The clearance buffer defaults to 3.5 schematic metres and is adjustable; this is a demo assumption, not an authenticated internal requirement. Both modes follow planned one-way routes and carry low loads; wrong-way events are not injected.

Without LOOADS disables speed, STOP and clearance enforcement and requests 6 km/h in both corridors and aisles. It deliberately permits modeled clearance breaches and geometric overlap. This is a risk counterfactual, not a recommended operating mode. No artificial throughput uplift, accident probability, or financial saving is applied.

Metrics are derived from fixed-step vehicle states and events. Minimum body gap may be negative when modeled footprints overlap; this is not a measured collision or injury. Waiting can increase and flow can stall. No safe-operation or certified collision-avoidance guarantee is made. Full internal LOOADS documentation has not been authenticated.