Transaction is defined
The rate states exactly what completes a move, divert or order line, including pickup and drop confirmation.
A useful warehouse robotics RFQ states what the equipment carries, how often each route is demanded, where constraints occur, which control layer owns each decision and how performance will be accepted. This page reuses a compact compliance-style structure to capture those engineering facts—not sustainability slogans.
Use one row per operating case when loads or routes differ materially. Attach source files, timestamps and owners so changes can be traced during design review with Tompkins Robotics.
| Input group | Minimum information | Why it changes the concept | Preferred evidence |
|---|---|---|---|
| Load unit | Length, width, height, mass, center of gravity, underside, overhang, orientation and stability | Controls payload interface, braking, pickup tolerance and safe transfer | Dimensioned drawing, weight range, representative samples |
| Material flow | Origins, destinations, moves by interval, SKU/order profile, replenishment, peak duration and cutoff | Determines mission demand, travel, queueing, buffer occupancy and fleet quantity | Timestamped transaction extract and route matrix |
| Site geometry | Clear aisle width, turns, gradients, floor joints, doors, lifts, crossings, safety zones and charging areas | Defines feasible paths, speed restrictions, traffic rules and intervention access | Current CAD, measured survey, marked photographs |
| Control interfaces | WMS/WCS/WES/PLC/fleet-manager ownership, messages, acknowledgements, timeouts, alarms and network policy | Prevents duplicated decisions, silent failures and unresolved exception states | Responsibility matrix, interface specification, sequence diagrams |
| Acceptance | Design and peak rates, transaction definition, availability boundary, test duration, fault cases and evidence owner | Turns commercial expectations into repeatable FAT and SAT decisions | Approved protocol, scenario data, result and deviation template |
The rate states exactly what completes a move, divert or order line, including pickup and drop confirmation.
Demand identifies duration, recurrence, replenishment overlap and whether backlog is allowed.
Every station and lane has a capacity, blocking rule and response when full or unavailable.
Commands, data state, retries, alarms and manual decisions belong to named systems and teams.
Pedestrian crossings, restricted areas, maintenance access and intervention points appear on the layout.
Blocked routes, host silence, network loss, low energy, lost load and disabled stations have expected outcomes.
If the project cites ISO 3691-4, ISO 12100, ISO 13849-1 or a regional machinery regulation, specify whether each item is a design reference, product requirement, site integration obligation or requested third-party certificate. Do not use a framework name as a substitute for model-specific documentation. Tompkins Robotics will not treat a logo row as acceptance evidence.
For brownfield deployments, include cutover zones, allowed downtime, rollback steps, inventory reconciliation and temporary manual flow. For expansion, include the next expected destinations, workload range, charging capacity and network growth so the initial phase does not create avoidable constraints.
Freeze the route map, robot configuration, interface version and test dataset before a run. Execute a named transaction mix for a stated duration, record completed and failed moves, queue time, buffer occupancy, energy state and recovery time, then repeat the same scenario after each material change. An example protocol requires a 120-minute nominal run followed by blocked-path, host-timeout and unavailable-station scenarios; the numbers are test conditions to approve, not Tompkins Robotics performance claims.
Fixed automation versus flexible robotics. Fixed conveyors and sorters often deliver higher deterministic capacity for stable SKU mixes and known divert accuracy targets, but changeovers can be slower when destinations or packaging change weekly. Flexible AMR and AGV fleets usually phase into brownfield aisles faster and adapt layouts under volatile demand, yet they depend more on traffic rules, charging windows and exception recovery. State which outcome you prioritize: peak lines per hour under a frozen order profile, or reconfigurable routes with a documented system availability boundary.
Peak rate versus operational resilience. Headline moves-per-hour can conflict with buffer occupancy, redundancy and graceful degradation. A design that maximizes short peaks may starve recovery when a station is unavailable. Your RFQ should name the peak window length (for example 90 minutes), whether backlog is allowed, and the mean-time-to-repair assumption that still keeps outbound cutoffs intact.
Architecture comparison dimensions (not vendor scores). Compare AMR, AGV and sortation paths on the same axes: payload envelope (kg and mm), clear aisle width (mm), peak factor versus average, WMS/WCS/PLC ownership, charging or energy window, and recovery time after a blocked path. Tompkins Robotics uses these axes to decide whether a flexible fleet, a guided AGV loop or a hybrid with fixed induction is the honest concept—not a brochure feature list.
Tompkins Robotics evaluations assume floor friction and joint quality allow the stated navigation mode; steep grades, soft floors or unmarked temporary obstacles can force speed derates. Shared aisles with forklifts require written priority rules before claiming concurrent throughput. Cold or wet zones may restrict battery chemistry, sensor performance and charging placement. These limits belong in the RFQ so SAT cases can exercise them instead of discovering them after go-live.
Attach the route matrix, load evidence, site drawing, interface responsibility and draft acceptance scenarios. Mark unknowns clearly; they are inputs to resolve, not gaps to hide.
Submit project requirements