Inventory replenishment policies in Dynamics 365 SCM

Inventory replenishment is the day-to-day question of when to order how much. Dynamics 365 SCM answers it through coverage groups — named policy bundles linked to items (and optionally per warehouse) that drive the master planning or Planning Optimization engine. Choosing the right coverage group per item category is the foundational supply chain decision.

Coverage codes. A coverage group specifies one of:

• Period — group demand into periods of N days; create one planned order per period covering that period's net demand.
• Requirement — create a planned order per demand transaction. Smallest order sizes but most order overhead.
• Min/Max — maintain inventory between minimum (reorder point) and maximum levels. Planning replenishes to max when min is breached.
• Manual — planning makes no suggestion; planner manages by exception.

Period coverage. Best for items with steady demand and reasonable lead time. Parameters:

• Coverage period — e.g. 7 days.
• Negative days — how far past the demand date the system can still consume from existing supply before suggesting more.
• Positive days — how far future supply can satisfy current demand.

A weekly period order means each Monday the system suggests one order covering the week's net demand. Reduces order count vs requirement-based.

Requirement coverage. Best for engineer-to-order, large-value, or low-volume items where pegging demand to supply matters more than order consolidation. Every demand creates its own supply suggestion.

Min/Max coverage. Best for fast-moving, low-value items where the cost of carrying safety stock is less than the cost of stockouts. Parameters:

• Minimum quantity — reorder point.
• Maximum quantity — order-up-to level.
• Minimum / Maximum keys — reference to date-dependent profiles for seasonality (e.g. higher min during holiday).
• Multiple — order quantity rounding (cases of 12).

The engine watches on-hand + on-order; when projected balance dips below min, an order suggestion brings it back to max.

Coverage group fields beyond coverage code.

• Safety stock — extra buffer above the minimum.
• Reorder margin — additional days of lead time buffer.
• Lead time — purchase or production lead time used in plan calculation.
• Calendar — working calendar to compute delivery dates.
• Time fence — how many days into the future planning produces suggestions.
• Action message time fence — when planning produces "expedite" or "delay" messages on existing supply.
• Forecast plan — which forecast plan the item consumes from.

Item coverage. At the item level, you can override the default coverage group per warehouse or even per dimension combination. A SKU might be Min/Max in the central DC and Requirement in regional DCs.

Allocation keys. When forecast is at family level but planning is at SKU level, allocation keys distribute the family forecast to SKU forecast based on historical percentages. Coverage group dictates how the SKU forecast then drives replenishment.

ABC analysis and coverage choice. A common heuristic:

• A items (top 20% by value) — Min/Max or Period with tight parameters; high attention.
• B items — Period; moderate parameters.
• C items — Min/Max or kanban with loose parameters; minimal attention.

Periodically re-classify; items move between categories as demand patterns shift.

Planning Optimization vs classic master planning. Coverage group parameters apply to both. Planning Optimization (the modern in-memory engine) processes them faster and supports incremental runs; classic master planning batch-processes the full plan. Both yield similar suggestions; the engine choice is a performance and architecture decision.

Common pitfalls.

• One coverage group for everything. Coverage groups should be a thoughtful taxonomy (3–10 named groups), not a one-size-fits-all.
• Lead times wrong. The single biggest cause of bad suggestions. Audit lead times annually; reality drifts from the system.
• Min/Max levels never adjusted. A reorder point set at go-live and never reviewed becomes wrong as volumes change. Build a quarterly review.
• Safety stock everywhere. Stacking safety stock on top of period coverage and lead time buffers inflates inventory without clear benefit.
• Time fences misunderstood. A short time fence creates churn (suggestions every day); too long misses real-time changes. Tune per item category.

The art beyond the parameters. Replenishment policy choice is half the story; the other half is the planner's discipline to act on the system's messages (expedite, delay, postpone). A coverage group that produces good suggestions only matters if someone executes them on time.