Planning glossary

Used in: Stock Planner

MRP uses expedite and reschedule messages to show when open supply no longer matches when it's needed.

Core Planning Concept

MRP systems calculate projected inventory balances across future time buckets.

As projected supply and demand conditions change over time, some open supply orders will no longer arrive when they're needed.

Exception messages are generated when the MRP calculation finds that open supply no longer matches what the plan needs, for example an order that should be brought in, pushed out or cancelled.

Examples may include:

• Projected shortages
• Inventory projected below safety stock
• Supply arriving later than required
• Supply arriving significantly earlier than required
• Excess projected inventory

Expedite Logic

An expedite recommendation is triggered when projected inventory falls below safety stock before an existing open supply order arrives.

In simple terms:

• Projected inventory drops below safety stock
• An open supply order is due in a later period
• MRP recommends bringing that order forward

Reschedule Logic

A reschedule-out recommendation is triggered when an open supply order is due before it's needed.

In simple terms:

• Projected inventory stays above safety stock without that order
• An open supply order is due in an earlier period than needed
• MRP recommends moving that order out to the period when projected inventory would otherwise drop below safety stock

Noise

Expedite and reschedule messages can create a lot of noise. Depending on your time buckets, MRP may recommend moving an open order by just a day. That's logically correct, but it isn't worth acting on, and dozens of small messages can hide the ones that matter. Find the sweet spot: set your time buckets or a minimum change so that only meaningful moves show up.

Relationship to MRP

Expedite and reschedule recommendations are outputs of the MRP calculation.

The planning system first calculates projected inventory balances over time, then evaluates whether existing supply timing still supports planning objectives.

The timing of supply is often just as important as the quantity of supply.

A company may have sufficient total inventory overall, but still experience shortages if supply arrives too late.

Important Operational Consideration

Frequent expedite activity may indicate underlying planning issues such as:

• Poor forecast accuracy
• Incorrect lead times
• Supply variability
• Inventory inaccuracies
• Capacity constraints

Many organizations attempt to reduce unnecessary expedite activity because expediting often increases operational cost, planning instability, and supply chain variability.

See also: Supply order, Planned order, Past-due

Used in: Stock Planner

The required ordering increment used after the minimum order quantity (MOQ) is met.

Example

• MOQ = 1,000 units
• Lot Size = 250 units

Valid order quantities would be:

• 1,000
• 1,250
• 1,500
• 1,750

Why Suppliers Use Lot Sizes

Suppliers often define lot size increments due to:

• Manufacturing batch sizes
• Packaging standards
• Carton or pallet quantities
• Material handling efficiency
• Freight optimization

Why It Matters

Widgets come in cartons of 24, and you can't buy half a carton. If you need 30, you have to order 48, which is 2 cartons.

Lot sizes come in many forms: a full pallet (say 40 cartons), a set number of drums that fit on a truck (say 80 drums of 200 litres), or a full container load (FCL) from an overseas supplier.

The key question is how much demand one lot covers. If a carton is a few days' or a week's worth of demand, rounding up makes little difference. If one lot covers months of demand, every order leaves you with extra stock, which ties up cash and space and raises the risk it becomes slow-moving or obsolete.

To check, divide the lot size by average weekly demand. For example, cartons of 24 with demand of 12 a week means each carton is 2 weeks of cover.

Relationship to MOQ

• MOQ defines the minimum starting order quantity
• Lot Size defines the required increment above MOQ

In many supplier environments, orders must satisfy both conditions simultaneously.

See also: Minimum order quantity (MOQ), Planned order, Inventory turns

Used in: Stock Planner

The minimum quantity a supplier allows per purchase order.

Example

A supplier may require a minimum order of 1,000 units per SKU.

Why Suppliers Use MOQ

Suppliers often establish MOQ requirements due to:

• Manufacturing batch/run sizes
• Material handling constraints
• Pallet or carton quantity standards
• Production setup costs
• Freight efficiency requirements

Why It Matters

The key question is how much demand the MOQ covers. If it's a week or two of demand, it has little effect. If it's months of demand, for example 6 months, you end up holding far more stock than you need, which ties up cash, takes up space, and raises the risk the stock becomes slow-moving or obsolete.

To check, divide the MOQ by average weekly demand. For example, an MOQ of 1,000 with demand of 50 a week is 20 weeks of cover.

See also: Lot size, Minimum order value (MOV), Planned order

Used in: Stock Planner

The minimum total dollar value required for a supplier purchase order.

Example

A supplier may require each purchase order to total at least $5,000.

MOV usually applies to each PO on its own, not to several POs added together. This can be a problem if you raise separate POs for different delivery locations but want them to count towards the same MOV. That normally has to be arranged manually with your account manager, and a supplier's system may reject a PO below the MOV.

Why It Matters

MOV constraints can force planners to:

• Purchase more inventory than immediately needed
• Consolidate orders
• Delay replenishment timing
• Increase inventory carrying costs

See also: Minimum order quantity (MOQ), Planned order, Supply order

Used in: Stock Planner

MRP (Material Requirements Planning)

MRP is a time-phased inventory planning process that projects future inventory balances over time.

Manufacturing ERP systems often include MRP that also explodes bills of materials (BoMs), calculating the components needed for finished products.

Some modules and apps advertised as "MRP" only perform BoM explosions, without time-phased planning. Check whether yours also projects stock against dated demand and supply to calculate what to order and when.

Core MRP Concept

At a simple level, MRP does this calculation for each item at each location, for each time period, such as each week:

• Starting stock on hand (SOH) at that location
• Plus incoming supply due to that location in that period
• Minus expected demand from that location for that period
• Resulting in projected ending inventory at that location for that period

The projected ending inventory from one period then becomes the starting inventory for the next period. That ending figure is the projected stock, also called the projected balance.

Large ERP systems usually calculate in daily buckets, while simpler planning tools often use weekly or monthly buckets.

Each location is planned on its own. Unless a source-destination relationship is set up in MRP, stock at one warehouse doesn't cover demand at another.

A general MRP principle is that planned supply orders are only created for dates after your open supply orders. That keeps the plan easy to follow, because you never see a new suggested order arriving ahead of an order you've already placed.

Simple Example

Week 1:

• Starting SOH = 500
• Supply = 200
• Demand = 400
• Ending SOH = 300

Week 2:

• Starting SOH = 300
• Supply = 100
• Demand = 450
• Ending SOH = -50 shortage

Time Buckets

MRP calculates across repeating time buckets.

Different systems may use:

• Daily buckets
• Weekly buckets
• Monthly buckets

Why Time Matters

Inventory may appear sufficient overall, but shortages can still occur if supply arrives after demand is required.

The timing of supply is often just as important as the quantity of supply.

Typical MRP Outputs

MRP generates:

• Planned supply recommendations
• Inventory projections
• Shortage visibility
• Time-phased supply views
• Inputs to capacity planning
• Inputs to cash planning
• Inputs to working capital planning

Obsolete SKUs

For SKUs marked Obsolete, the MRP rules specify that no further supply orders are to be generated.

Existing inventory can therefore be consumed by demand until it reaches zero.

• No planned supply orders are generated for Obsolete SKUs
• Projected SOH cannot fall below zero
• Once available stock is exhausted, projected SOH remains at zero
• This represents the SKU being run down and eventually finished

Example

Starting SOH = 100

Demand = 150

Supply = 0

Projected SOH = 0, not -50.

If demand continues in later weeks, projected SOH remains 0 because no further supply is planned.

Important Operational Consideration

MRP outputs are only as good as the input data, so it's worth making sure that data is as accurate as possible.

Wrong stock balances, lead times, supply dates, forecasts or transaction timing flow straight through into wrong order recommendations.

See also: Planned order

Used in: Stock Planner

Definition

Demand or supply that should already have happened by its date but hasn't.

Example

A purchase order due on 1 July that hasn't arrived by 2 July is past-due.

Sales orders can be past-due too. They're usually called back orders.

Why it matters:

The key question is whether a past-due order will still happen, and when.

• Purchase orders: any supplier, local or overseas, might have lost your order or never received it. Overseas orders also have more genuine reasons to run late, such as a delayed ship, a busy port, or goods held in customs. So a late overseas order is more likely to still be on its way, while a late local order is a stronger sign that something has gone wrong. Either way, check with the supplier and get a new date.
• Sales orders: in some industries, a customer will wait weeks for a back order. In others, the sale is lost after a short delay. Check whether old back orders are still wanted before planning stock for them.

See also: Supply order, Expedite / reschedule, MRP

Used in: Stock Planner

A planned order is an MRP output: a recommended quantity and date to buy, make or transfer, depending on the supply source set up. It is calculated from projected stock, lead times, MOQ, lot size and other order rules. It isn't placed yet. The plan raises it so the projected balance stays at or above safety stock. Also called a planned supply order.

In larger ERP systems, a planner releases a planned order, turning it into an open purchase, production or transfer order.

Why it matters: planned orders show future buying, production and transfers ahead of time, helping with supplier forecasts, cash and capacity planning, and confirming lead times early. Review them before they become supply orders.

See also: MRP, Supply lead time, Safety stock

Used in: Stock Planner

Supply lead time is the total elapsed time between placing a supply order and the inventory being received into Stock on Hand (SOH).

Typical Lead Time Components

Lead time may include:

• Supplier order processing
• Manufacturing or production time
• Picking and packing
• Freight transit time
• Customs clearance
• Receiving and warehouse processing

Example

A supplier may quote:

• 10 days manufacturing
• 7 days ocean transit
• 3 days receiving and processing

Total supply lead time = 20 days

Important Operational Consideration

It is important to understand exactly what activities are included within the agreed lead time.

Different supplier agreements may define responsibility differently.

Examples include:

• Supplier ships from their warehouse only
• Supplier delivers to your receiving dock
• Freight arranged by customer vs supplier

Supplier Agreement Differences

Terms such as:

• Ex-Works (EXW)
• Free Into Store (FIS)

can significantly change the practical lead time experienced by the planner.

The same supplier manufacturing time may result in very different inventory arrival dates depending on who manages freight and receiving activities.

Terms like these are often based on Incoterms, the international rules that set out who pays for and manages each step of shipping. To learn more, see the ICC Incoterms rules.

The date on the PO isn't always the date stock arrives in your warehouse. For example, on FOB terms with 1 month to manufacture and get to port, then 1 month on the vessel, the PO date is the FOB date at the port, but the MRP lead time to your warehouse is 2 months. Plan on the date the stock is due into your warehouse, not the PO date.

Planned vs Actual Lead Time

The planned lead time is what the supplier quotes or what's set in your system. The actual lead time is how long orders really take to arrive. These can differ a lot. If the planned lead time is 20 days but orders take 35 days on average, plans built on 20 days will keep running short. It's worth tracking actual arrival times and deciding which number to plan with. You can also ask the supplier whether the quoted time is realistic.

DIFOT (Delivered In Full, On Time) helps check whether planned lead times are realistic. The "On Time" part measures whether deliveries arrive by the agreed date. If orders placed with the planned lead time regularly arrive late, compare actual lead times with the planned figure and discuss the difference with your supplier.

Understanding Supplier Supply Models

Effective planners benefit from understanding how their suppliers actually replenish and produce inventory.

Different suppliers may operate very differently:

• Make products only after receiving customer orders
• Wait to accumulate enough demand before starting production
• Hold finished stock ready for immediate shipment
• Purchase inventory only after receiving customer demand

For manufacturing suppliers, additional delays may occur due to:

• Production scheduling queues
• Limited manufacturing capacity
• Availability of raw materials or components
• Internal supplier lead times

Even when a quoted lead time appears fixed, the underlying supply process may create variability.

Developing a conceptual understanding of how suppliers operate can significantly improve planning decisions and shortage risk management.

Why It Matters

Incorrect lead times can result in:

• Inventory shortages
• Excess inventory
• Expediting costs
• Missed customer demand
• Incorrect MRP recommendations

See also: Planned order, MRP, Supply order

Used in: Stock Planner

A supply order represents incoming inventory intended to satisfy future demand.

Depending on the business model, supply orders may include:

• Purchase Orders (POs) from external suppliers
• Production / Manufacturing Orders created internally
• Transfer Orders bringing stock in from another internal location

Unique Identifiers

Each supply order typically contains a unique identifier used for:

• Tracking
• Receiving
• Supplier communication
• Shipment matching
• Planning visibility
• Audit/history purposes

Supplier System Relationship

Your purchase order (PO) is usually entered into the supplier's system as a sales order (SO).

This means:

• Your incoming supply is the supplier's outgoing demand
• Both systems may reference different document numbers
• Cross-referencing identifiers is important during issue resolution

Why It Matters

Supply orders are critical inputs into MRP and inventory planning because they determine:

• Future inventory availability
• Expected replenishment timing
• Shortage risk
• Supplier execution performance

Common Risks

Supply orders may become unreliable due to:

• Supplier delays
• Incorrect dates or quantities
• Duplicate orders
• Unprocessed receipts
• Cancelled or held orders
• Production constraints

See also: Planned order, Past-due, MRP

Used in: Stock Planner

A Unit of Measure (UoM) defines how inventory, purchasing, production, or sales quantities are represented and tracked.

Why It Matters

Your unit of measure and your supplier's may not match. You might count stock in single units while your supplier sells by the carton or the box. An order for "100" could mean 100 units or 100 cartons, so always say which unit you mean on the purchase order. Then check that the supplier's confirmation uses the same one.

Your own units may not match inside the business either. You might buy and store a liquid in drums but use it in manufacturing in fluid ounces, or buy and store steel in set lengths but use it in linear metres. The planning system needs a conversion between each unit so stock, usage and orders all line up.

Common UoM Examples

• Each (EA)
• Carton (CTN)
• Inner Carton
• Outer Carton
• Pallet (PLT)
• Kilogram (KG)
• Liter (L)

Why UoM Matters

Different businesses and suppliers may transact the same product using different units of measure.

Incorrect assumptions can create major ordering and inventory issues.

Real-World Example

A planner intended to order inner cartons of Easter chocolate.

The supplier interpreted the order as outer cartons.

Since each outer carton contained 6 inner cartons, the actual order quantity became 6× larger than intended.

UoM Conversion

UoM conversion defines the mathematical relationship between different units.

Example:

• 1 Outer Carton = 6 Inner Cartons

Regional and Supplier Differences

Different countries, industries, and suppliers may use different measurement systems.

Examples include:

• Metric vs Imperial measurements
• Liters vs Fluid Ounces
• Kilograms vs Pounds
• Metric tons vs Short tons

UoM assumptions should always be validated when working across suppliers, regions, or other business systems.

Important Operational Consideration

Equivalent quantities are not always operationally interchangeable.

Example:

• 1 × 10L paint tin = 10 liters
• 10 × 1L paint tins = 10 liters

Both represent the same total volume, but they may not satisfy the same operational requirement.

A production process may specifically require the single 10L container format.

Why It Matters

Incorrect UoM handling can result in:

• Over-ordering or under-ordering
• Inventory inaccuracies
• Production disruptions
• Freight inefficiencies
• Receiving discrepancies
• Planning errors

See also: MRP, Minimum order quantity (MOQ), Forecast

Used in: Stock Planner

Customer service level measures whether stock is available for the sales order when the customer wants it, against the service standard you've set. If the standard is a 2-week lead time, the question is whether the stock was available for the sales order within 2 weeks. If the standard is supply from stock, the question is whether it was available for the sales order on the day the customer ordered, not merely physically present and still awaiting processing. It is measured as the share of orders, lines or units supplied in full and on time against that standard.

Common Examples

Service expectations may include:

• Immediate stock availability
• Same-day shipment
• Delivery within several days
• Backorder acceptance
• Scheduled production lead times

Customer Expectations Can Differ

Different customers may have very different expectations.

Examples include:

• Some customers expect products immediately from stock
• Some customers are comfortable waiting for replenishment
• Some products are expected to always be available
• Some specialized products may naturally require longer lead times

Customer expectations influence your company's service policy, but customer service level measures performance against the rules your company has set. If a customer wants same-day supply but your policy is next-day supply, you measure whether the sales order was supplied in full by the next day—not the same day.

That policy might cost you customers who need faster service, but that is separate from meeting your service standard. Your service level target sets how often you aim to meet that standard—for example, 95 times out of 100.

Working Capital Trade-Off

Higher service levels generally require higher inventory investment.

Additional inventory may increase:

• Working capital requirements
• Warehouse space usage
• Inventory carrying costs
• Obsolescence risk

Example

Service level and inventory investment are not linearly related.

Achieving very high availability levels will probably require disproportionately larger increases in safety stock and working capital. A 100% customer service level would need infinite inventory, because you can't plan for every outlier.

Balancing Service and Inventory

Effective planning involves balancing:

• Customer expectations
• Inventory investment
• Supply lead times
• Demand variability
• Business profitability

Industry and Market Expectations

The industry and market shape what customers expect: supply from stock, or within an agreed lead time.

Examples include:

• FMCG (Fast-Moving Consumer Goods, called CPG in the US) customers expect products on the shelf, supplied from stock
• Industrial or specialised product customers often accept an agreed replenishment lead time
• Spare parts customers expect critical items from stock, while non-critical items can be supplied to a lead time

Once your company has chosen its service standard, the customer service level target is how many times out of 100 you aim to meet that standard.

Important Operational Consideration

Not every product or customer necessarily requires the same service level.

Businesses often apply different inventory and replenishment strategies depending on:

• Customer importance
• Product criticality
• Demand predictability
• Supply risk
• Commercial value

See also: Safety stock, Inventory turns, Sales history variability

Used in: Stock Planner

Inventory Turns / Stockturns

Stockturns show how much working capital is needed to support your sales. More turns means less cash tied up in stock for the same sales. Fewer turns means more cash tied up in stock.

Stockturns measure how many times a year your average stock is used up and replaced to meet customer demand, at the service level you've chosen. For example, if your cost of goods sold for the year is $600,000 and you hold $100,000 of stock on average, that's 6 turns a year.

Simple Concept

If a business replenishes inventory:

• 52 times per year → inventory turns faster
• 12 times per year → inventory turns slower
• 2 times per year → inventory turns much slower

Why It Matters

Faster inventory turns reduce:

• Working capital requirements
• Inventory carrying costs
• Warehouse space usage
• Obsolescence exposure

Relationship to Customer Service

Inventory exists to support customer service expectations.

However, the amount of inventory required depends heavily on:

• Replenishment frequency
• Supply lead times
• MOQ and lot size constraints
• Demand variability
• Supplier reliability

Ordering smaller quantities more frequently can often reduce average inventory investment — if operationally and commercially possible.

Example

With the same annual demand, and other factors unchanged:

• Receiving one week's demand each week generally needs less average cycle stock than receiving one month's demand each month
• Receiving one month's demand each month generally needs less average cycle stock than receiving six months' demand twice a year

The difference is how much stock is received each time, not how often a purchase order is placed on its own. The schedule of receipts can differ from how often purchase orders are raised.

In practice, ordering frequency may be constrained by:

• MOQ requirements
• Freight economics
• Supplier production models
• Lead times
• Operational capacity

Relationship Between Turns and Working Capital

Inventory turns and working capital are closely related.

All else being equal:

• Higher inventory turns generally reduce average inventory holdings
• Lower average inventory holdings generally reduce working capital requirements

As a simple conceptual example:

• If inventory turns double, average inventory investment may roughly halve.

Aggregate vs Individual Product Performance

Inventory turns are often measured at an aggregate business or warehouse level.

This means overall inventory performance may hide important product-level differences.

Slow-moving or obsolete inventory (SLOB) can negatively affect overall inventory turns because capital remains tied up in inventory with low or no consumption.

Businesses often analyze inventory turns at multiple levels such as:

• Company-wide
• Warehouse/location
• Product category
• Individual SKU

Balancing Efficiency and Service

Effective inventory planning balances:

• Customer service expectations
• Working capital investment
• Supply reliability
• Operational practicality
• Inventory efficiency

Lot Size and Stockturns

All else being equal, with no safety stock, your average stock is half of each order lot. If you halve the lot size, average stock halves and stockturns double. The working capital you need halves too. It's a direct relationship. For example, ordering 2 weeks of demand at a time instead of 4 takes you from 13 turns a year to 26.

See also: Lot size, Safety stock, Customer service level

Used in: Stock Planner

Safety Stock / Safety Time

Safety stock and safety time are planning buffers used to protect against uncertainty in supply and demand.

Safety Stock

Safety stock is additional inventory intentionally held above expected demand requirements.

Safety stock is held in your base unit of measure, the same unit you use for stock on hand. If you count stock in single units, set safety stock in single units too, not in cases or pallets. Otherwise the plan will compare numbers that don't match.

Safety Time

Safety time represents a time-based protection buffer.

Instead of storing a fixed quantity, planners define additional coverage time such as:

• 3 days
• 1 week
• 14 days

Converting Time Into Units

For MRP, safety time is converted into an inventory quantity using expected demand rates.

Example:

• Average demand = 100 units/day
• Safety time = 5 days

Result:

• Safety stock requirement = 500 units

Why Safety Stock Exists

Safety stock protects against two things:

• Variability in customer demand, for example when the forecast is wrong or a large order arrives unexpectedly.
• Variability in supply timing, for example supplier delays or transport disruptions.

Lead Time and Variability

Longer lead times generally create greater uncertainty and variability.

The further planners must look into the future, the less predictable supply timing and customer demand may become.

Supplier Operating Model Matters

Supplier operating models can significantly affect supply variability.

Examples include:

• Manufacturers producing only after orders are received
• Suppliers waiting to accumulate enough demand before production
• Wholesalers holding finished inventory ready for shipment
• Distributors purchasing stock only after receiving customer orders

Manufacturing suppliers may also experience variability due to:

• Production scheduling delays
• Raw material shortages
• Component availability
• Capacity constraints

Safety Stock as an MRP Trigger

In MRP, safety stock isn't meant to sit unused. It's the level that triggers replenishment.

MRP calculates projected stock for each period. When projected stock drops below safety stock, MRP plans a new supply order to bring it back up.

Customer service level:

This is the share of demanded units you aim to supply straight from stock, without back orders or lost sales. For example, a 95% service level means that for every 100 units customers order, you aim to ship 95 from stock on hand.

The higher the service level you need, the more safety stock you have to hold. The cost also climbs steeply near the top: going from 95% to 99% takes much more extra stock than going from 90% to 95%. So it's worth setting higher targets for your most important items and lower ones for the rest.

Normal vs Abnormal Variability

Safety stock is generally designed to absorb normal operational variability, not extreme or exceptional events.

Example:

• Expected demand = 100 units
• Typical variability may range between 80–130 units

Safety stock may help protect against this type of normal fluctuation.

However, safety stock is usually not intended to fully absorb highly abnormal events such as:

• Demand surges from 100 to 600 units
• Major supplier shutdowns
• Port closures
• Natural disasters
• Systemic supply chain disruptions

In most planning environments, "normal" variability is estimated using historical operational behavior.

Past demand patterns, supplier performance, and lead time variability are commonly used to determine appropriate safety stock levels.

Why It Matters

Insufficient safety stock can increase shortage risk.

Excessive safety stock can increase:

• Inventory carrying costs
• Obsolescence exposure
• Warehouse space requirements
• Working capital usage

See also: MRP, Planned order, Customer service level

Used in: Stock Planner

Slow-moving inventory refers to products with relatively low inventory usage or sales activity over time.

SLOB commonly stands for:

• Slow Moving
• Obsolete Inventory

Slow Mover vs Obsolete

• Slow movers still have ongoing demand, but consumption occurs infrequently or at low volumes
• Obsolete inventory generally has little or no expected future demand

Why Slow Movers Occur

Slow-moving inventory may result from:

• Demand declines
• Forecast inaccuracy
• Large MOQ or lot size constraints
• Long lead times
• Safety stock policies
• Product lifecycle changes
• Customer behavior changes

Example

A product averaging only 2 units of demand per month while holding 500 units of inventory may be considered a slow mover.

Important Operational Consideration

Slow-moving inventory is not always "bad" inventory.

Some products intentionally carry low turnover because they are:

• Critical spare parts
• Service items
• Long lead-time products
• Insurance stock for important customers
• Required for regulatory or contractual obligations

Why It Matters

Excessive slow-moving inventory can increase:

• Inventory carrying costs
• Warehouse space usage
• Working capital requirements
• Obsolescence risk
• Write-off exposure

SLOB Management

Businesses often review SLOB inventory to determine whether actions are required such as:

• Inventory reduction
• Transfers between locations
• Supplier return negotiations
• MOQ reduction discussions with suppliers
• Product phase-out planning
• Promotional sales activity
• Disposal or write-off

See also: Inventory turns, Forecast, MRP

Used in: Demand Planner

Exceptions

An exception is any event that causes recorded sales to differ from the normal range of customer demand.

Exceptions may increase recorded sales, reduce recorded sales, or prevent genuine customer demand from being captured. If these events are not identified, they can distort the demand pattern used for forecasting.

Why are exceptions important?

Statistical forecasting assumes that past demand is a reasonable indicator of future demand. When unusual events are included in the sales history, they can distort this relationship and produce forecasts that are either too high or too low.

Identifying exceptions helps separate one-off events from the underlying demand pattern that the forecast is trying to predict.

Common Examples

Examples of exceptions include:

• Promotions
• Customer stock builds
• Large project orders
• Product launches
• End-of-life clearance
• Supply shortages
• Data entry errors
• Natural disasters or industrial action

For example, a customer's usual buyer may be away on holiday or vacation, with no trained person to cover. Sales that normally run at 40,000 units a month can fall to zero because nobody placed the order, not because the underlying demand disappeared.

Not every unusual sale should be treated as an exception. The objective is to identify events that are unlikely to occur again under normal trading conditions.

See also: Sales history, Intermittent demand, Forecast

Used in: Stock Planner / Demand Planner

Forecasting

Forecasting is the process of estimating future demand for a product.

Forecasts help organizations plan inventory, purchasing, production, capacity, cash flow, and supply requirements before actual customer orders occur.

Good to know

A forecast is your best estimate of what customers will want to buy. It isn't a promise of what you can supply. Limits on working capital, manufacturing capacity, or other constraints can mean you're able to supply less than the forecast.

Why Forecasting Matters

Most supply chains require time to respond to demand.

Because materials often have lead times, companies must make inventory and supply decisions before actual demand is known.

Forecasting helps organizations:

• Anticipate future demand
• Position inventory appropriately
• Improve customer service
• Reduce shortages and excess stock
• Support purchasing and production planning

Common Forecast Types

Organizations may use different forecast methods depending on the business.

Common examples include:

• Statistical forecasts based on historical demand
• Forecasts from customers
• Forecasts from account managers / product managers
• Promotional forecasts
• Consensus or collaborative forecasts

Forecast Accuracy

Forecasts are estimates and will rarely be perfectly accurate.

Some level of normal variation between forecast and actual demand is expected in nearly all supply chains.

Demand variability, seasonality, promotions, market changes, and unexpected events can all affect forecast accuracy.

The objective is not perfect accuracy, but forecasts that are directionally useful and operationally reliable.

Consistent forecast bias is often more problematic than normal forecast variation.

Repeatedly over-forecasting may create excess inventory, while repeated under-forecasting may increase shortages and expedite activity.

Forecast bias is a persistent tendency to forecast above or below actual demand. Over-forecasting leads toward excess stock. Under-forecasting leads toward shortages and expediting. Normal variation is different: it moves in both directions and is not persistently high or low.

At an aggregate company level, forecasts are generally more accurate in percentage terms than for an individual SKU at a location. Over-forecasts and under-forecasts can offset. That is not guaranteed. Correlated changes in demand, and systematic bias, may still move the total. An accurate total can also hide errors in the mix or at particular locations.

Aggregation is not the same as a longer forecast horizon. Aggregation combines many items or locations. A longer horizon looks further ahead in time.

Forecast Value Add (FVA) is how much demand-planning effort improves accuracy compared with a simple baseline, such as a simple moving average. Compare them on the same actual demand, the same periods, and the same forecast horizon, using the same accuracy measure. A positive result means the extra work improved accuracy. A negative result means the baseline did better. FVA shows whether the analysis, the meetings, and the manual adjustments earn their keep.

Accuracy measures are generally more stable, and more useful, when volumes are larger and there are enough observations. At low volumes, a miss of one or two units can create a very large percentage error, and zeros can make some percentage measures undefined. Read the percentage with the unit error, any bias, and the effect on stock and customer service. A small number of sales does not make every accuracy measure meaningless.

Forecast Consumption

In many planning environments, forecasts are gradually replaced by actual customer orders as demand becomes known.

Important Operational Consideration

Poor forecast quality can lead to:

• Excess inventory
• Stock shortages
• Expediting costs
• Reduced customer service
• Increased working capital

While normal forecast error is expected, forecasts that are consistently inaccurate or heavily biased become far less useful for planning purposes.

In most supply chain environments, it is often more important that a forecast is operationally useful than perfectly accurate.

Forecasts are commonly more accurate in percentage terms at higher levels of aggregation than at detailed item or customer levels.

For example, aggregate forecasts may still be highly useful for:

• Revenue projections
• Capacity planning
• Workforce planning
• Procurement planning
• Financial planning

Forecast performance should therefore be monitored and continuously improved over time, with particular attention given to reducing systematic forecast bias.

Forecast Generation

Forecasting is fundamentally the process of separating signal from noise.

Historical sales contain both genuine demand patterns (the signal) and random variation (the noise).

By focusing on consistent patterns rather than isolated events, the resulting forecast is generally more representative of underlying customer demand.

The quality of the forecast is heavily influenced by the quality of the Sales History used as input. Reviewing exceptions and maintaining representative Sales History will generally improve forecast quality more than changing forecasting parameters.

See also: Forecast horizon, Sales history, Sales history variability

Used in: Demand Planner

The forecast horizon is how far ahead the forecast runs. Near-term forecasts support replenishment and scheduling, while longer-term forecasts support supplier commitments, capacity, budgets, and cash flow.

The horizon needs to be long enough to cover lead times and those commitments.

A nearer-term forecast is generally more accurate, because there is less time for conditions to change and the information is fresher. That is not an absolute guarantee.

Further out, uncertainty usually grows. Update the forecast as those periods get closer.

See also: Forecast, Supply lead time, MRP

Used in: Demand Planner

Intermittent demand has many weeks with no sales. The demand is irregular, and the size of each order may vary.

An average of 10 a week might really be an order of 100 every 10 weeks. The smooth average hides that pattern. Stock can sit idle between orders, then run short when the order arrives.

Look at both the gaps and the order quantities, particularly the demand during the supply lead time. A run of weeks with no sales does not automatically mean the product is obsolete.

Percentage accuracy measures may mislead when actual demand is zero.

See also: Exception / outlier, Forecast, Sales history

Used in: Demand Planner

Sales history for forecasting is net customer sales: customer sales less credits and returns, in units. It excludes production consumption and internal stock transfers.

Those are movements of stock, not customer sales. Mixing them into the history distorts demand and can double-count it.

Historical sales are imported from your accounting, sales or warehouse system and converted into a standard format suitable for forecasting.

Why is sales history validated?

Every business system stores sales differently. Before forecasting can begin, the imported data is validated to ensure it is complete, consistent and suitable for analysis.

Validation helps identify data quality issues before they affect forecast accuracy.

Typical Validation Checks

Validation includes checks such as:

• Required columns are present
• Dates are valid
• Quantities are numeric
• Duplicate records are identified
• SKU and Location values are populated

What should I expect from my source system?

Every source system exports sales history differently.

Some systems provide individual invoice transactions, while others export daily, weekly or monthly summaries. Depending on the system, an export may also include returns, cancelled orders, or stock transfers. Use returns and credits to reach net customer sales. Leave out stock transfers and production consumption, even when the file contains them. They are stock movements, not customer sales.

See also: Exception / outlier, Forecast, Unit of measure (UoM)

Used in: Demand Planner

Sales history variability

Sales history variability is how far recorded sales fluctuate around their usual level. The same average can sit inside a narrow range or a wide one.

An observation that is unusual for a steady SKU may still fall inside the normal range of a more volatile SKU.

Products with relatively consistent demand are generally easier to forecast than products with highly variable demand, even when the average demand is the same.

For example, the following sales histories both average 100 units per week:

Low Variability

100, 110, 90, 100, 80, 120

High Variability

50, 200, 120, 80, 0, 150

Although both examples have the same average demand, the first has a tighter normal range. A week that looks unusual in that steady pattern may still sit inside the range of the more volatile pattern.

With the lead time and the service target unchanged, higher demand variability generally needs more safety stock.

Historical sales are used to estimate that variability, but stockouts and exception events can distort what the history shows.

See also: Forecast, Safety stock, Customer service level

Used in: Demand Planner

SKU Forecasting Group

Rather than reviewing forecasts one SKU at a time, products can be grouped into meaningful business categories to simplify reporting and identify trends.

A SKU can sit in more than one grouping, such as brand, category, or product manager. Each grouping is another way to slice the same sales and forecasts.

These groupings are alternative views of the same sales. Do not add the views together where they overlap, or the same sale is counted more than once. This is a general planning idea. It does not mean a planning system provides an unlimited number of grouping fields.

Common Examples

The Forecasting Group can represent any classification that is meaningful to your business, for example:

• Product Manager
• Product Category
• Product Family
• Brand
• Supplier
• Department
• Business Unit
• Product Type

There is no single correct choice—the Forecasting Group should reflect how you naturally review and manage your product range.

See also: Forecast, Sales history, Exception / outlier