The supplier said ten days. The planner entered ten days. The MRP ran clean for six months. Then the eleventh shipment arrived on day fourteen, the twelfth on day twelve, the thirteenth on day sixteen, and suddenly the production schedule that looked solid on Monday is a fire by Thursday. The plant manager is calling the supplier directly, the procurement officer is paying expedite fees, and the planner is staring at an MRP report that says everything was fine because the system still believes a ten-day lead time is the truth.
This is what happens when planning systems treat lead times as constants. Suppliers slip. Carriers reroute. Customs holds a container. Raw material vendors juggle priorities, and your order moves down their queue without anyone telling you. The number in the supplier record stays at ten days, but the actual world starts running on twelve, fifteen, sometimes twenty. The gap between the recorded lead time and the realised lead time is where every expedite, every overtime shift, and every angry customer call gets manufactured.
Building an MRP system that survives this reality requires a different mental model. It requires planning across horizons, calculating order-by dates that account for variance, and treating the supplier lead time as a working assumption rather than a fact.
Why Static Lead Times Make MRP Fragile
Most MRP implementations treat the supplier lead time as a single number stored against a supplier or item record. The planner enters ten days. The system subtracts ten days from the production date and prints an order-by date. If the order is placed on or before that date, the material is assumed to arrive on time. The whole logic chain rests on one number being correct.
The problem is that the number is rarely correct in the way the system assumes. It is an average, a best case, or a quote from the supplier when business conditions were calmer. Real lead times are distributions, not points. A supplier with a stated ten-day lead time might deliver in eight days half the time, in twelve days a quarter of the time, and in fifteen or more days the remaining quarter. If your MRP plans against the ten-day average, you will be late on at least a quarter of orders by definition.
The fragility shows up when slippage compounds. A material arrives two days late. The production order it was tied to slips. Downstream finished goods orders slip with it. The next planning cycle inherits the delay, and the system starts generating recommendations that assume the new normal is the old normal. Within weeks, the planner stops trusting the MRP output and starts running parallel spreadsheets to second-guess every order-by date the system produces.
This is the moment most teams decide MRP does not work for them. It is also the moment when the real fix is not abandoning MRP but rebuilding it around horizons and explicit lead time variance MRP logic that acknowledges suppliers slip.
How Horizon-Based Planning Absorbs Slippage
The first structural defence against supplier variance is planning across multiple time windows rather than one. A system that only looks at the next seven days has no chance of catching slippage early. By the time a fourteen-day variance becomes visible, the order is already past due. A system that looks across seven, fourteen, thirty, and sixty day windows surfaces the same shortfall at four different urgency levels, and gives the planner four different opportunities to intervene.
FalOrb structures its MRP engine around exactly this layered view, with configurable horizons at seven, fourteen, thirty, and sixty days. A material shortfall driven by a supplier who is consistently slipping shows up first in the sixty-day horizon as a low-urgency signal. The planner sees it during routine review, has time to talk to the supplier, qualify an alternate, or adjust a production date. By the time the same shortfall appears in the seven-day horizon, the structural fix is already in motion. The seven-day view exists to catch genuine surprises, not to discover problems that should have been visible weeks earlier.
This approach is explored in more depth in our piece on MRP planning horizons, which walks through the operational logic of each window. The key insight for handling lead time variance is that horizons act as a buffer in time. A supplier who slips by three days does not break the thirty-day horizon, because the horizon was built to accommodate uncertainty. The same slip would break a single-window MRP that was tuned tightly to the stated lead time.
Horizons also create natural escalation tiers. A shortfall visible only at sixty days is a planning conversation. A shortfall visible at thirty days is a procurement action. A shortfall visible at fourteen days is an expedite candidate. A shortfall visible at seven days is a crisis. The system does not need to invent urgency, because the horizon itself encodes it.
Order-By Dates That Reflect Real Constraints
The second piece of the puzzle is replacing static lead times with order-by date calculations that work backwards from the production date. An order-by date is the latest day on which a purchase order can be placed and still arrive in time for the production it supports. It is calculated by taking the required-on date from the production schedule and subtracting the supplier lead time, with a buffer for variance if the planner has set one.
The power of an order-by date is that it converts a static lead time into a moving target that recalculates every time the production schedule changes. If a production order moves from week four to week three, the order-by dates for every material on its bill of materials shift accordingly. The planner does not have to remember to recalculate. The system surfaces the new dates the moment the schedule changes.
FalOrb generates order-by dates for every purchase recommendation produced by its MRP engine. The recommendation includes the suggested quantity rounded to the supplier's minimum order quantity, the order-by date computed from the production schedule and supplier lead time, and an urgency tier that shifts as the order-by date approaches. When an order-by date is breached, an alert fires. The alert is not a generic stock notification. It is a specific signal that a particular material for a particular production order is now at risk because the order window has passed.
This is the level of specificity that makes dynamic reorder timing useful. A planner facing a list of fifteen items below their minimum threshold has to triage manually. A planner facing a list of three items with breached order-by dates and twelve items still within their order window has a clear action queue. The system has done the prioritisation that the planner would otherwise do mentally, and it has done it consistently across every item in the catalogue.
Urgency Tiers and the Expedite Signal
Even with horizons and order-by dates, suppliers will still slip in ways that require active intervention. The question is not whether expedites will happen. The question is whether the system surfaces them as an explicit signal or buries them in a list of routine alerts that the planner has learned to scroll past.
FalOrb's restock intelligence engine assigns an urgency tier to every recommendation: critical, soon, or monitor. The tier is derived from how close the situation is to breaching the production schedule, taking into account current stock, in-transit inventory, scheduled receipts, and the order-by date for any new orders. An item with healthy stock and a comfortable order-by date sits in monitor. An item where the order-by date is within the next planning window moves to soon. An item where the order-by date has been breached or where stock will hit zero before any plausible receipt sits in critical.
This tiered approach turns the expedite signal into something operational rather than reactive. When a supplier confirms a delay, the planner does not have to manually calculate whether the delay will cause a stockout. The system recalculates the projected available balance with the new arrival date, reclassifies the urgency tier, and updates the alert. If the new tier is critical, the planner knows to call alternates. If it is still in soon, the planner knows there is room to absorb the delay without expediting.
The same logic applies in reverse. When a supplier confirms an early shipment or when consumption slows because a production order was delayed, items that were sitting in soon can drop back to monitor. Alerts auto-resolve. The action queue shrinks. The planner spends time on the items that genuinely require attention rather than on items where the situation has already corrected itself.
Treating Lead Time as a Living Variable
The deeper shift behind all of this is treating supplier lead time as a living variable rather than a fixed attribute. Every receipt against a purchase order is a data point. The actual lead time, measured from order placement to receipt, is information the system can use to refine its assumptions over time.
A mature lead time variance MRP setup tracks actual versus expected lead time per supplier and per item. When the actual consistently exceeds the expected, the system flags the gap. The planner can update the supplier lead time to reflect the new reality, or they can leave the recorded value in place and add an explicit variance buffer. Either way, the decision is informed by data rather than by a phone call to the supplier where they insist the original quote is still accurate.
This kind of feedback loop is only possible when every receipt creates an immutable record that links back to the original purchase order. Our piece on the immutable audit ledger explores why this matters for traceability, but the same property is what makes lead time analytics possible. You cannot measure variance against a moving target. You can only measure it against a record that captures what was promised, what was ordered, and what arrived. When all three are preserved, the supplier lead time stops being a guess and starts being a measurement.
The shift from reactive expedites to predictive procurement is not a single feature. It is a stack of design decisions: horizons that absorb variance, order-by dates that recalculate dynamically, urgency tiers that prioritise without manual triage, and lead time data that learns from every receipt. Suppliers will still slip. The MRP just stops breaking when they do.
FalOrb helps manufacturers build MRP that absorbs supplier lead time variance through layered planning horizons, dynamic order-by dates, and urgency-tiered restock recommendations. Book a 30-minute walkthrough or email us at [email protected] to see how it applies to your operation.