MRO inventory management balances technical availability, working capital, obsolescence risk, and replenishment capability. The portfolio is difficult because a rarely used spare may carry a severe operational consequence, while another item moves substantial value without being critical to continuity. One classification cannot represent both dimensions. The public ISO 55000:2024 overview frames asset management around value, context, and organizational objectives, a useful basis for keeping service, risk, and capital in the same decision.
MRO covers materials used for maintenance, repair, and operations, such as bearings, seals, motors, filters, lubricants, tools, electrical parts, and workshop consumables. Production inventory includes raw materials, work in process components, and finished goods associated with the regular manufacturing flow. Purpose and the organization's accounting model define the boundary. The same lubricant can be a process input in one application and an MRO material in another.
Official and neutral references were consulted on August 23, 2026. Formulas, bands, and policies require calibration with local data, costs, service targets, and governance. Formulas and examples do not replace an inventory policy, financial decision, SAP configuration, demand forecast, or availability commitment.
Technical criticality and usage value answer different questions
Technical criticality evaluates the consequence of not having the item in the asset context: safety, environment, production, quality, redundancy, recovery time, contingency, and failure reach. ABC classification measures the economic concentration of usage. A rarely issued spare motor can be technically critical and appear as B or C in one year. A high volume lubricant can be A by value while having several approved alternatives.
Master data should keep these dimensions separate. Treating A as a synonym for critical focuses attention only on spending. Requiring stock for every critical item ignores redundancy, repairability, preservation, demand probability, lead time, and sourcing options.
The asset criticality matrix makes consequence and redundancy explicit. The spare parts policy adds material and supply chain characteristics.
ABC starts with annual usage value
For each material code, calculate annual usage value = quantity consumed during the period × comparable unit cost. Consumption should use actual issues for use, adjusted for returns, reversals, transfers, and movements that do not represent demand. Finance and materials teams define one consistent cost basis.
Sort items from highest to lowest value, calculate each share of the total, and build the cumulative percentage. The organization defines A, B, and C boundaries based on the observed concentration. Bands such as 70% to 80% of value for A are common references rather than a universal rule. The official Indian National Productivity Council guide states that percentage patterns are approximate and may vary widely.
| Class | Economic reading | Possible controls |
|---|---|---|
| A | Concentrates much of annual usage value | Frequent review of parameters, prices, balances, forecasts, and excess |
| B | Intermediate participation | Periodic review with risk exceptions |
| C | Small share of value, often many material codes | Simpler process without ignoring criticality or obsolescence |
A high unit price alone does not make an item A. The formula uses annual quantity and cost. A $40,000 instrument with no issue during the period has zero annual usage value, although its balance, criticality, and financial exposure still require another view.
XYZ describes demand variability
The XYZ method used in this policy classifies consumption regularity over time. For a monthly series, calculate the mean, standard deviation, and coefficient of variation: CV = standard deviation ÷ mean. Lower values indicate relatively stable demand associated with X. Intermediate values form Y. Higher values and erratic patterns form Z.
XYZ boundaries should reflect the portfolio distribution. An open study on ABC XYZ segmentation uses CV for demand variability and also shows that boundaries belong to the selected method. Applying 0.5 and 1.0 without examining the data may place almost every spare in Z. An item with a zero mean has no useful CV. Intermittent demand also requires the intervals between events, quantity per event, projects, shutdowns, and fleet changes.
| Class | Behavior | Analytical implication |
|---|---|---|
| X | More stable consumption | Statistical forecasts generally have a stronger basis |
| Y | Moderate variation, trend, or seasonality | Separate explainable patterns from noise |
| Z | Erratic or intermittent demand | Give more weight to criticality, events, and scenarios |
The observation period must cover relevant cycles. Twelve months may hide a major overhaul occurring every two years. Thirty six months can help when technology changes, decommissioned assets, and exceptional consumption are marked.
The ABC XYZ matrix guides management attention without automating policy
The combination produces nine cells. AX joins high annual value and stability, supporting rigorous review of forecasts, prices, and lots. AZ joins high value and variability, requiring investigation of events and excess risk. CZ can include inexpensive unpredictable consumables and rare parts that stop critical assets.
Each cell opens a question set rather than issuing a fixed answer. A critical CZ item with a fourteen month lead time may justify one preserved unit. An AX item with a nearby supplier and predictable usage may operate with frequent replenishment. A repairable AZ item may need a pool and repair agreement instead of several new units.
Add technical criticality, observed lead time, substitutability, repairability, shelf life, preservation requirements, minimum order quantity, supplier risk, obsolescence, and shortage cost to the matrix. The final decision records assumptions, owner, and review date.
Lead time covers the complete cycle
Total lead time can include requisition preparation and approval, quotation, manufacturing, transportation, customs, receiving, inspection, testing, and release. For a repairable item, it also includes removal, shipment, diagnosis, approval, repair, return, and acceptance testing. A supplier promise represents only part of the cycle.
Use observed history and retain dispersion instead of one average. The United States Government Accountability Office documented how gaps between estimated and actual acquisition lead times for spare parts affect readiness and investment. A 60-day average can hide deliveries ranging from 30 to 180 days, which matters for safety stock.
When lead time changes, reorder point, safety stock, maintenance schedule, and alternatives need a coordinated review. Contract improvements, local repair, standardization, or an approved second source can free capital with less exposure than a direct balance cut.
Service level, reorder point, and safety stock
A service target needs a defined measure. Cycle service level is the probability of avoiding a stockout during a replenishment cycle. Fill rate is the proportion of demand supplied immediately. Asset availability is a different outcome that also depends on diagnosis, labor, access, and execution.
A basic continuous review expression uses reorder point = expected demand during lead time + safety stock. If average demand is 4 units per month and total lead time is 2 months, expected lead time demand is 8. Safety stock depends on demand variability, lead time variability, selected service level, and the statistical assumptions.
For intermittent demand, normal distribution and square root of time assumptions can produce fragile parameters. Simulation, failure scenarios, interval data, and criticality decisions may be more suitable. The method should be documented and tested against actual shortages and excess.
Maximum stock, economic lot, purchase multiple, and review frequency also influence the balance. The matrix classifies management attention, while the replenishment parameter belongs to the model selected for that item.
Repairable items need a separate loop
A repairable unit moves through available, reserved, installed, removed, awaiting diagnosis, under repair, condemned, and returned states. Required quantity depends on failure events, repair turnaround, condemnation, logistics, and installed positions. Counting warehouse balance alone misses units trapped in the loop.
Control serial number where applicable, condition, location, report, repair supplier, due date, warranty, accumulated cost, and configuration. A removed core needs to reenter the process. Buying another unit because the repairable was forgotten in receiving increases capital and hides loop performance.
Pool compatibility requires technical evidence. Two motors with the same power may differ in voltage, mounting, ingress protection, shaft, encoder, brake, or environment. Engineering approves substitution and interchangeability.
Obsolescence and duplicates start in master data
Duplicates arise from free text, language, abbreviation, unit, manufacturer, or part number differences. Before creating a code, search technical attributes, manufacturer, part number, drawing, equipment, and approved alternatives. Cleansing should preserve history and cross references instead of merging items because descriptions look similar.
For obsolescence, record installed asset, installed base, last purchase, last issue, supplier, discontinuation notice, alternative, repair option, and replacement plan. A nonmoving item can be a strategic spare, excess, scrap, or a part for a retired asset. Balance alone cannot decide.
Master data quality indicators include potential duplicate codes, inconsistent units, balance without linked equipment, missing part numbers, incomplete descriptions, and blocked materials with open reservations.
SAP MM and SAP PM connect material and work
In the SAP design, MM supports material master data, inventory, goods movements, and procurement. PM organizes technical objects, plans, maintenance orders, operations, components, execution, and maintenance costs. Product version and configuration determine available behavior.
Official SAP S/4HANA Cloud documentation states that adding stock components to a maintenance order operation can generate a reservation. Nonstock components can generate a purchase requisition. The requirement date indicates when the material should be available. A reservation asks the warehouse to keep material ready for a future purpose.
Reservation, availability commitment, and goods issue represent different moments. A reservation does not prove withdrawal. Goods issue records actual consumption and supports assigning quantity and cost to the order. Separated material that is not used needs a return or cancellation process. Consumption without an order loses the link to asset, failure, and work.
The maintenance work order preserves that connection. Understanding SAP PM technical objects helps prevent consumption against the wrong equipment or functional location.
Worked case: utility line U-4
Teaching scenario: utility line U-4 has duty pump P-430 and standby P-431. One 75 kW motor fits both positions, and no approved local substitute exists. The plant analyzes four material codes using twelve months of net issues, comparable average cost, and monthly consumption. The figures demonstrate the method and are not policy recommendations.
| Item | Annual use | Unit cost | Annual value | CV | Class |
|---|---|---|---|---|---|
| Seal kit M-8821 | 24 | BRL 3,200 | BRL 76,800 | 0.29 | AX |
| Repairable motor M-5170 | 2 | BRL 32,000 | BRL 64,000 | 1.73 | AZ |
| Bearing M-6312 | 48 | BRL 420 | BRL 20,160 | 0.62 | BY |
| Gasket M-1402 | 120 | BRL 35 | BRL 4,200 | 0.88 | CY |
Total annual value is BRL 165,160. The seal kit represents 46.5%. The motor raises the cumulative share to 85.3%. The local rule keeps in A the item that crosses 80%. The bearing takes the cumulative share to 97.5% and remains B under the rule that includes the item crossing 95%. The gasket closes C. Declared boundaries prevent result manipulation.
Technical and supply decisions
The AX seal kit has stable consumption, two approved suppliers, a 25 to 40 day observed lead time, and pump redundancy. Its reorder point starts with lead time demand and the selected service target. Procurement negotiates lot size, while planning identifies major shutdown demand separately from normal history.
The AZ motor remains critical despite erratic use. A new unit needs 120 days, repair usually needs 75, and the historical condemnation rate, meaning the share rejected as beyond economical repair, is 10%. The plant keeps one preserved unit in the pool, a diagnosis agreement, and monthly review of units under repair. Criticality, repair loop, and lead time support the quantity.
The BY bearing has an approved alternative and a 15 day lead time. The CY gasket is inexpensive, small, and shelf life controlled. Each receives a distinct policy. Old code M-3319 appears as a possible motor duplicate. Engineering confirms documentary equivalence, materials blocks new purchasing on the old code, and history remains available through a cross reference.
From the order to actual consumption
An inspection finds vibration and suspects damage in the P-430 motor. The maintenance order includes the motor as a component and generates the reservation under the configured process. The warehouse stages the unit without posting goods issue before diagnosis. Misalignment is confirmed, and the installed motor remains fit. The reservation is canceled and the spare returns to available status without fictitious consumption.
During a later shutdown, electrical inspection condemns the installed motor. The pool unit is reserved, withdrawn, and issued to the order. The removed motor enters the repair loop with serial number, diagnosis, quotation, approval, and deadline. The order receives material, cause, measurements, labor, and result. Its history improves the next ABC, XYZ, criticality, and pool sizing review.
Day 180 result
The closeout uses the same population and cutoff rule as the baseline. Immediate fulfillment for critical items rose from 76 fulfilled requests out of 84, 90.5%, to 92 out of 95, 96.8%. For other items, it rose from 188 out of 205, 91.7%, to 224 out of 235, 95.3%. The case targets were at least 95% for critical items and 93% for the others. Both results support retaining the tested policy without turning either percentage into a universal availability promise.
| Measure | Baseline | Day 180 | Decision |
|---|---|---|---|
| Critical-item fill rate | 76/84 = 90.5% | 92/95 = 96.8% | Retain the service level and review every critical shortage |
| Other-item fill rate | 188/205 = 91.7% | 224/235 = 95.3% | Retain segmentation without raising every item |
| Orders delayed by material | 18/120 = 15.0% | 5/138 = 3.6% | Review all five orders by cause and material |
| Completed repairable loop | 104-day median across 9 loops | 71-day median across 12 loops | Retain the contract and pool against the internal 80-day ceiling |
| Expired reservations | 22/96 = 22.9% | 4/112 = 3.6% | Correct the owner and reason for all four |
| Identified excess capital | BRL 248,000 | BRL 97,000 remaining, 39.1% of baseline | Revise disposition plans for the remaining items |
The BRL 151,000 addressed reconciles to the starting capital: BRL 96,000 came from avoiding one duplicate purchase and BRL 55,000 from validated transfer or use, totaling BRL 151,000. BRL 248,000 minus BRL 151,000 leaves BRL 97,000. At Day 180, the team retains the AX and AZ rules, the repairable loop, and reservation cleanup. The five delayed orders, four expired reservations, and BRL 97,000 remain under review with an owner and cause for each group. The ABC×XYZ matrix remains decision evidence rather than an automatic command to buy, issue, or dispose.
Indicators connecting service and capital
- Fill rate and stockouts by technical criticality.
- Orders delayed by material unavailability.
- Inventory value, excess, and nonmoving balance by policy.
- Inventory accuracy and count variance.
- Promised, average, median, and percentile lead time.
- Expired reservations, staged material without issue, and issue without order.
- Total repair turnaround and units by loop status.
- Condemnation rate and repair cost compared with replacement.
- Confirmed and prevented duplicate codes.
- Obsolescence with installed base and action plan.
- Parameters reviewed after criticality or supplier changes.
High turnover can coexist with repeated shortages. Low stock can hide delayed orders. The dashboard should connect capital, service, risk, and execution instead of rewarding one isolated metric.
PM Run within documented product boundaries
PM Run is a planning, mobility, and execution layer over SAP PM. It can support access to the work package and field records used in maintenance execution. SAP MM and the organization's defined processes remain responsible for material master data, inventory, reservations, goods movements, MRP, and procurement.
PM Run does not calculate ABC, XYZ, safety stock, reorder points, service levels, or forecasts. It also does not replace SAP MM, physical inventory, financial policy, spare parts engineering, or criticality decisions. Any integration or visualization depends on the scope and configuration that are actually contracted and documented.
Sources and limitations
- SAP Help, creating reservations and purchase requisitions for maintenance orders, consulted August 23, 2026.
- SAP Help, maintenance order component data, consulted on the same date.
- SAP Help, checking material availability status.
- ISO 55000:2024, asset management overview and principles.
- National Productivity Council of India, inventory control techniques and ABC.
- U.S. Government Accountability Office, acquisition lead times for spare parts.
- Open study on value based ABC and variability based XYZ, neutral academic source.
SAP pages reflect the products and versions identified in the portal at the consultation date. Transactions, fields, and behavior vary by product, release, and configuration. ISO material is copyrighted, so only its public overview was used without reproducing the standard.
Frequently asked questions
Must every critical spare be held in stock?
The decision considers consequence, redundancy, substitution, repair, lead time, probability, preservation, and alternatives. Criticality starts the analysis but does not set quantity alone.
Does class A mean technically critical?
A indicates concentration of annual usage value under the ABC rule. Technical criticality requires a separate assessment.
How should XYZ handle very rare demand?
CV becomes weak with a near zero mean and long intervals. Analyze event spacing, quantities, shutdowns, fleet, criticality, and scenarios.
Does a reservation mean the part was consumed?
A reservation requests material for a future purpose. Goods issue records actual withdrawal under the configured SAP process.
Does PM Run replace SAP MM for MRO inventory?
No. PM Run operates as a planning, mobility, and execution layer over SAP PM. Material management stays in the organization's defined systems and processes.
After SAP processes are defined, see how PM Run connects planning and execution to SAP PM.
