DEV Community

Cover image for The Kanban Card Needs the Supplier’s Next Milk-Run Time, Not an Average Lead Time
Techleadsit
Techleadsit

Posted on

The Kanban Card Needs the Supplier’s Next Milk-Run Time, Not an Average Lead Time

Fusion SCM Cloud Online Training becomes easier to apply when a replenishment signal is tied to a real receiving rhythm. Consider a bicycle plant that consumes brake calipers from a line-side bin. Its nearby supplier runs a milk route that reaches the dock at 7:00, 11:00, and 15:00 on working days. When an operator empties a bin at 10:20, a generic eight-hour lead time points toward the evening, when the dock is closed and no route is planned. The useful promise is the next scheduled delivery opportunity, adjusted for the plant calendar and the time needed to replenish. Connecting the Kanban pull sequence to that cadence gives production, receiving, and the supplier one operational expectation instead of three interpretations of an average lead time.
Averages Hide the Shape of the Day
A lead time can describe elapsed effort without describing when transport actually moves. Suppose the caliper loop uses six containers, each supporting forty minutes of assembly. The supplier may need only ninety minutes to pick and stage a replacement, yet a truck departing at fixed intervals determines arrival. A card triggered just before 11:00 behaves differently from one triggered just after it. Applying the same duration to both can produce an expected time that no truck can meet. The mismatch affects more than a timestamp. A supervisor may expedite unnecessarily, receiving may staff the wrong period, and a planner may conclude that container quantity is inadequate. A recurring delivery schedule preserves the discrete windows that govern the physical flow.
Build the Schedule Around Observable Commitments
The design conversation should begin with the route agreement, not a convenient set of times typed from memory. The plant and supplier need to confirm working days, dock cutoffs, staging duration, holiday treatment, and whether several deliveries can occur on one day. They should also decide which time represents the commitment: truck departure, gate arrival, or material availability at the receiving point. For this loop, gate arrival is the clearest shared event. The pull-sequence lead time still matters because it represents replenishment work before the schedule is consulted. The organization calendar matters because a Friday afternoon signal must not silently target a closed Saturday dock. Clear definitions make the calculated expectation testable against actual events.
Oracle’s current readiness note Associate Delivery Schedules with Kanban Pull Sequences explains that supplier and interorganization pull sequences can use recurring delivery schedules. A schedule can contain multiple delivery times for every working day, while a specific-days pattern supports one time for each selected day. One schedule can be associated with a pull sequence, and an optional default can support newly created sequences. The expected delivery date and time is calculated with the Kanban replenishment time, pull-sequence lead time, organization calendar, and associated schedule. The note also cautions that changing the default does not update schedules already attached to existing pull sequences. That boundary is important during rollout because a new default is not a bulk conversion mechanism.
One Association Must Represent the Whole Pull Sequence
A pull sequence with multiple approved suppliers deserves special review because it still carries one associated delivery schedule. If two suppliers serve different routes, attaching one supplier’s cadence to the shared sequence could create an expectation that the other cannot meet. The team may need to align the suppliers to a common milk run, reconsider the sequence design, or leave the schedule unassociated until the operating model is coherent. The same reasoning applies to interorganization replenishment. A shuttle between plants may run predictably, but the schedule should represent its actual destination arrival pattern. Configuration cannot reconcile a commercial sourcing choice that implies incompatible delivery rhythms. The schedule is most useful where one sequence truly has one dependable cadence.
Defaulting also requires restraint. Marking a common schedule as the default can reduce setup effort for new pull sequences, but it should not become evidence that every new part belongs on that route. Brake calipers may travel three times daily, while painted frames arrive only on Tuesdays and Thursdays. A sequence owner should verify the schedule during creation, and changes should follow controlled ownership. Because an associated schedule cannot simply be deleted while in use, retirement planning should identify affected sequences before removing obsolete definitions. Naming conventions can help administrators distinguish routes, but the operational test remains more important: does the selected pattern match the supplier or internal transport commitment for this material flow?
Test Signals on Both Sides of a Delivery Window
A useful test set triggers cards at 10:20, 10:55, and 11:05, then compares the expected results with the schedule and configured replenishment time. Another test should trigger late on Friday and cross a nonworking day. Review the delivery expectation on in-process cards and in the administration experience, not only the setup page. Then change a default schedule and confirm that an already associated sequence retains its prior assignment. Include a pull sequence without a schedule to verify that scheduled and unscheduled methods can coexist as intended. These cases reveal boundary behavior that a midday happy-path test misses. They also give warehouse and supplier representatives concrete timestamps to validate before production depends on them.
After deployment, compare calculated delivery times with gate arrivals and line-side availability. Repeated misses before the truck reaches the dock may indicate a supplier or transport problem; misses between the dock and the line suggest an internal receiving delay. Cards consistently generated just after a window may justify changing container quantity or route frequency rather than editing lead time until the numbers look favorable. The expected timestamp is therefore both a promise and a diagnostic reference. Owners should review schedules when shifts, calendars, suppliers, or routes change, and should communicate exceptions such as a temporary holiday timetable. A precise configuration becomes stale if no one owns the physical cadence it represents.
Conclusion
The brake-caliper loop works when its digital expectation follows the truck that can actually carry the replacement. A recurring delivery schedule turns a broad lead-time estimate into the next feasible window while retaining replenishment time and the organization calendar. Sound implementation verifies the meaning of each time, limits one association to a genuinely shared cadence, handles defaults deliberately, and tests signals immediately before and after route boundaries. Teams studying Oracle Fusion SCM Training can use this case to connect configuration with a visible factory outcome: an operator releases a card, receiving anticipates a credible arrival, and the line gets another container without avoidable expediting. Ongoing comparison with gate and line-side events keeps the schedule useful as the route evolves.

Top comments (0)