Replenishment Methods and When to Use Them

Continuous review (reorder point and Min–Max)

• Reorder point (ROP): Place an order when inventory position ≤ ROP.

• Min–Max: When at or below Min, order up to Max.

• Best for: High-value or fast movers (A/X, A/Y), items with steady demand, and where real-time inventory is available.

Periodic review (P or T systems)

• Review at fixed intervals; order up to a target level covering review period + lead time.

• Best for: Lower-value items, group buying, supplier calendars, or when inventory visibility is limited.

EOQ and lot-sizing rules

• EOQ balances ordering and holding costs to set economical batch sizes.

• Modify with MOQ, case packs, production capacity, and price breaks.

• Best for: Stable demand items and made-to-stock manufacturing.

Kanban and two-bin systems

• Card/bin signals replenish a fixed container size; assumes short, reliable lead times.

• Best for: Parts supermarkets, workstations, and highly repetitive consumption.

Vendor-managed inventory (VMI)

• Supplier monitors downstream inventory and triggers replenishment to agreed targets.

• Best for: High-velocity consumables with strong supplier partnerships and data sharing.

Hybrid approaches and triggers

• Push for pre-season builds and launches; pull for in-season flow.

• Dual trigger: Periodic review for planning plus continuous emergency ROP.

• Demand-driven top-ups of pick faces, decoupled from purchase replenishment rules.

Browse our latest articles on inventory forecasting, replenishment planning, and Shopify operations in the Verve AI blog.

Calculations and Worked Examples

Safety stock formulas by service level

• Inputs:

  • Average demand per period (d), standard deviation of demand (σd)

  • Average lead time in periods (L), standard deviation of lead time (σL)

  • Service level (cycle), expressed by z-score (e.g., 95% -> z ≈ 1.65)

• If demand varies and lead time is constant:

  • Safety stock = z × σd × sqrt(L)

• If both demand and lead time vary (independent):

  • σDLT = sqrt(L × σd^2 + (d^2 × σL^2))

  • Safety stock = z × σDLT

Quick calculator steps:

• Choose service level and get z.

• Compute σDLT using your d, σd, L, σL.

• Multiply z × σDLT. Round up to case pack.

Reorder point under variable demand and lead time

• Demand during lead time (mean) = d × L

• Reorder point (ROP) = (d × L) + Safety stock

For periodic review (review interval R):

• Target level = d × (R + L) + Safety stock for R + L

• Safety stock period = z × σd × sqrt(R + L), or use σDLT over R + L if lead time varies.

Order quantity (EOQ, MOQ, capacity limits)

• EOQ = sqrt((2 × D × S) / H)

  • D = annual demand units

  • S = ordering/setup cost per order

  • H = annual holding cost per unit

Practical adjustments:

• If EOQ < MOQ, order MOQ.

• Snap to case/pallet multiples.

• Cap by storage or cash.

• Use price break analysis: compare total annual cost across breakpoints.

Retail SKU example

• Given:

  • d = 20 units/day, σd = 6/day

  • L = 7 days, σL = 2 days

  • Service level = 97.5% (z ≈ 1.96)

  • Case pack = 12, MOQ = 36

• Compute σDLT:

  • σDLT = sqrt(7 × 6^2 + 20^2 × 2^2) = sqrt(252 + 1600) ≈ sqrt(1852) ≈ 43.0

• Safety stock = 1.96 × 43.0 ≈ 84.3 → round to 84 (nearest multiple of 12 is 84)

• Mean demand during LT = 20 × 7 = 140

• ROP = 140 + 84 = 224 units

• EOQ (assume D = 7,300 units/year, S = $25/order, H = $2/unit-year):

  • EOQ = sqrt((2 × 7,300 × 25)/2) = sqrt(182,500) ≈ 427 → round to 432 (36 × 12)

• Policy:

  • Continuous review with ROP = 224; order 432 units per trigger, respecting MOQ/case.

Spare part or long-tail example

• Given:

  • d = 3 units/week, intermittent; σd = 3/week

  • L = 3 weeks, σL = 1 week

  • Service level = 90% (z ≈ 1.28)

  • MOQ = 1, case = 1

• σDLT = sqrt(3 × 3^2 + 3^2 × 1^2) = sqrt(27 + 9) = sqrt(36) = 6

• Safety stock = 1.28 × 6 ≈ 7.7 → 8 units

• Mean demand during LT = 3 × 3 = 9

• ROP = 9 + 8 = 17 units

• Method choice:

  • Periodic review monthly with target covering 4 + 3 = 7 weeks, or consider on-order-only with substitution if extremely slow.

Step-by-Step SOP (Roles, Triggers, SLAs)

Parameter setup and master data governance

Roles:

• Planner (Accountable): Sets service levels, methods, and parameters by SKU-location.

• Data steward (Responsible): Maintains master data (lead time, MOQ, packs, UOM).

• Buyer (Responsible): Supplier terms, MOQs, calendars.

• IT/Systems (Consulted): WMS/ERP configuration and integrations.

• Finance (Consulted): Carrying cost, working-capital constraints.

• Operations lead (Informed): Warehouse capacity and slotting constraints.

SLAs and controls:

• Lead times reviewed monthly for A, quarterly for B/C.

• Forecast and service levels refreshed monthly; promotional overrides weekly.

• Parameter changes require dual approval for A items.

Trigger logic and review cadence

• Continuous review items: System evaluates inventory position every transaction; triggers when ≤ ROP or Min.

• Periodic review items: Weekly or biweekly review; order up to target.

• Kanban: Visual or system cards trigger immediate top-up.

• Review cadence:

  • A/X: Weekly parameter review; daily exception monitoring.

  • B/Y: Biweekly; C/Z: Monthly or event-based.

Approval and exception queues

• Auto-approve orders within thresholds; escalate if:

  • Exceeds budget or capacity

  • Breaks MOQ/price-break policies

  • Requires expedites or supplier deviation

• Exception queues to manage:

  • Stockout risk within lead time

  • Supplier delays and past-due POs

  • Data anomalies (sudden demand spikes, negative on-hand)

Execution in WMS/ERP

• Plan: System generates POs/TOs; buyer confirms dates and quantities.

• Receive: Dock-to-stock within SLA; capture lot/expiry if applicable.

• Putaway and slotting: Reserve and pick-face assignments; respect FEFO/lot rules.

• Pick-face replenishment: Trigger when pick bin ≤ Min or before waves.

• Close loop: Update on-hand, backorders, and allocations in real time.

Audit and continuous control

• Monthly audit: Parameter drift (z, σ, L) vs actuals.

• ABC/XYZ recalc quarterly.

• Cycle counting results drive root-cause fixes (adjust ROP if shrinkage/system lags).

• Governance logs for parameter changes and approvals.

Warehouse Execution and Slotting

Primary vs reserve locations and pick-face top-ups

• Keep a small, ergonomic pick-face and bulk in reserve.

• Set pick-face Min–Max based on daily pick velocity and replenishment SLA.

• Proactive top-ups during off-peak to avoid picker delays.

Wave vs demand-driven replenishment

• Wave-based: Pre-stage top-ups before batch waves; suits high-volume retail.

• Demand-driven: Trigger immediate top-up when pick task risks shorting.

• Use hybrid: Pre-wave plus near-real-time safety for fast movers.

Task interleaving and labor planning

• Interleave putaway, replenishment, and picking to reduce dead-heading.

• Labor plan from forecasted picks and top-up work; protect capacity for peak hours.

Perishables: lot, FEFO, and expiry

• Enforce FEFO; set dynamic pick-face Max to minimize aging.

• Parameterize shelf-life at receipt; trigger early depletion strategies for short-dated stock.

Cycle counting impacts on replenishment

• Freeze replenishment for locations under count to prevent discrepancies.

• After adjustments, recheck ROP/Min triggers to avoid false orders.

Multi-Location and Omnichannel Policies

Store/DC/3PL replenishment rules

• Stores: Periodic review with presentation minimums; limit backroom accumulation.

• DCs: Continuous review with ROP/EOQ; align with supplier MOQs.

• 3PLs: Codify SLAs, visibility, and parameter ownership in contracts.

Transfer orders and inventory pooling

• Use TOs to balance across nodes; set lateral transfer thresholds and costs.

• Pool inventory centrally when demand variability is high; decouple for region-specific seasonality.

Multi-echelon considerations

• Assign service levels by echelon, not everywhere at max.

• Upstream buffers sized to protect combined downstream variability.

• Propagate demand realistically (sell-through, not shipments) to avoid bullwhip.

Drop-ship and cross-dock flows

• Drop-ship: Supplier holds buffer; you set allocation and ATP rules.

• Cross-dock: Time orders to wave departures; minimal safety stock at the cross-dock node.

Monitoring, KPIs, and Diagnostics

Service level, fill rate, and stockout rate

• Cycle service level: Percent of cycles without stockout.

• Fill rate: Percent of demand fulfilled immediately.

• Stockout rate: Percent of SKUs or order lines shorted.

Target ranges (typical starting points):

• A/X: Fill rate 97–99%, stockout <2%

• B/Y: Fill rate 95–97%, stockout <5%

• C/Z: Fill rate 90–95%, stockout <10%

Inventory turns and carrying cost

• Turns = COGS / Average inventory.

• Carrying cost = Capital + warehousing + risk (% of inventory value).

• Set targets by class; e.g., A: 8–12 turns, C: 3–6 turns.

Replenishment accuracy and on-time top-ups

• Replenishment accuracy: Percent of orders created per policy (right qty/time).

• On-time top-ups: Percent of pick-face replenishments completed before pick start.

Target setting and alert thresholds

• Alerts when:

  • Projected stockout within lead time window

  • On-hand > 1.5× Max or > 60 days of supply

  • Supplier OTIF < 95% for A items

  • Forecast error MAPE > target by class

Root-cause analysis playbook

• If stockouts rise:

  • Check lead-time drift, supplier OTIF, and demand spikes.

  • Recompute σDLT and safety stock; apply temporary expedite.

• If overstock rises:

  • Validate forecast bias, MOQ constraints, and price-break buys.

  • Reduce service level for low-impact SKUs; enable transfers/markdowns.

• If fill rate lags but stock is available:

  • Investigate pick-face mins, wave timing, and system reservation rules.

• If accuracy issues persist:

  • Audit master data, counting accuracy, and unit-of-measure conversions.

Exceptions, Seasonality, and New Products

Handling supplier delays and backorders

• Triage by ABC: Expedite A, re-promise B, backorder or substitute C.

• Split shipments and partial receipts; adjust ATP and notify customers.

• Trigger temporary service-level uplift for affected SKUs if variability spikes.

Promotions and peak season overrides

• Create promo uplift factors and blackout dates for parameter freezes.

• Temporarily raise ROP/Target level using forecasted uplift and longer lead times.

• Pre-build upstream, protect pick-face capacity, and add labor for top-ups.

• Post-event burn-down plan: Reduce targets, accelerate transfers, and markdowns.

New product introduction and end-of-life

• NPI:

  • Set initial buy using analog SKUs and launch calendar.

  • Use tighter review cadence and conservative service targets for first cycles.

• EOL:

  • Freeze reorders, deplete via substitutions and promotions.

  • Protect high-priority customers; avoid returns/obsolescence.

Substitutions and allocation rules

• Define acceptable substitutes and hierarchy.

• Allocation by channel priority or customer tier during constraints.

• Track lost sales vs substituted sales to refine parameters.

Implementation Roadmap and Technology

Selecting WMS/ERP/APS capabilities

• Must-have features:

  • Continuous and periodic review policies

  • Safety stock/ROP calculators with variability

  • MOQ, pack, price-break, and capacity constraints

  • Multi-location ATP and transfer planning

  • Pick-face replenishment triggers and FEFO control

• Selection criteria:

  • Data latency and API integration

  • Parameter governance and audit trails

  • Usability for planners and operators

  • Scenario testing and what-if analysis

Data migration and parameterization

• Cleanse master data: UOM, case/pallet, lead times, MOQ, supplier calendars.

• Initialize ABC/XYZ and carry into policy templates.

• Backfill history (orders, POS) for at least 12–24 months where available.

Pilot, A/B testing, and stabilization

• Pilot with a representative SKU set and one location.

• A/B test old vs new parameters; track fill rate, turns, and inventory value.

• Iterate weekly until KPIs stabilize; then scale in waves.

Change management and training

• SOP documentation and RACI clearly communicated.

• Role-based training: planners, buyers, warehouse leads.

• Daily huddles for exceptions; weekly KPI reviews with action owners.

FAQs

How do I choose between Min/Max, reorder point, and Kanban?

• Reorder point: Best for high-value, steady-demand SKUs with real-time visibility.

• Min/Max: Similar to ROP but “order-up-to”; good when suppliers require larger batches.

• Kanban: Best for short, reliable lead times and repetitive consumption, often in factories or parts supermarkets.

What’s the difference between replenishment and restocking?

• Replenishment sets the policies and triggers for ordering or transferring stock to meet service targets.

• Restocking is the warehouse execution of moving goods to pick faces or shelves based on those policies.

How do I calculate safety stock with variable lead times?

• Use σDLT = sqrt(L × σd^2 + (d^2 × σL^2)), where d and σd are demand mean and standard deviation per period, L and σL are lead time mean and standard deviation (in the same periods). Safety stock = z × σDLT.

Which KPIs matter most and what targets should I set?

• Fill rate, stockout rate, inventory turns, carrying cost, replenishment accuracy, and on-time top-ups. Start with A/X fill rate 97–99%, stockout <2%, turns 8–12; tune by class and business goals.

How often should I review and update replenishment parameters?

• A/X weekly, B/Y biweekly, C/Z monthly. Always review after major changes in lead time, supplier performance, or demand patterns (promos, seasonality).

How do I plan for seasonality and promotions?

• Apply uplift factors to demand, extend assumed lead times, temporarily raise ROP/targets, and pre-build upstream. Freeze parameters close to the event and run a burn-down plan after.

What changes for perishables or regulated items?

• Enforce FEFO and lot controls, track expiry, and use smaller pick-face max to prevent aging. Ensure compliant storage, documentation, and recall readiness.

When is vendor-managed inventory a good fit?

• When a supplier can see your inventory and consumption reliably, lead times are consistent, and both parties agree on service targets and governance. It’s ideal for high-velocity consumables and shared logistics gains.

Ready to replace spreadsheets entirely? See how Verve AI helps Shopify merchants forecast demand and create smarter purchase orders.