# XYZ Demand Variability Classification Worksheet > Source: https://netray.co/tools/xyz-demand-variability-classification-worksheet > Publisher: Netray (netray.co), AI and integration engineering for ERP and manufacturing systems. > Section: Free Tools ## Short answer Classify items into X, Y and Z by coefficient of variation of demand. Edit the rows, add your own items and export to Excel. ## Category supply-chain ## Audience - Materials planner - Demand planner - Inventory manager ## Tool **Kind:** worksheet **Mode:** editable **Columns:** #### Item **Key:** item **Type:** text #### Average monthly demand **Key:** mean **Type:** number **Unit:** units **Help:** Mean demand per period over at least 12 periods. #### Std deviation of demand **Key:** sd **Type:** number **Unit:** units **Help:** Standard deviation of the same periods. #### Coefficient of variation **Key:** cv **Type:** number **Expr:** if(gt(mean,0), sd / mean, 0) **Format:** decimal **Help:** Std deviation divided by mean. #### Class (1=X, 2=Y, 3=Z) **Key:** cls **Type:** number **Expr:** if(lte(cv,0.5), 1, if(lte(cv,1), 2, 3)) **Format:** integer **Help:** X: CV up to 0.5. Y: above 0.5 up to 1.0. Z: above 1.0. **Rows:** **Item:** BRG-6204-2RS **Mean:** 1200 **Sd:** 180 **Item:** SEAL-KIT-A14 **Mean:** 340 **Sd:** 210 **Item:** HYD-HOSE-3/8-2M **Mean:** 95 **Sd:** 120 **Item:** PCB-CTRL-REV-C **Mean:** 600 **Sd:** 260 **Item:** GASKET-FLG-DN50 **Mean:** 2100 **Sd:** 350 **Item:** MOTOR-0.75KW-4P **Mean:** 18 **Sd:** 31 **Totals:** #### Total average demand **Key:** mean **Agg:** sum **Format:** number #### Average CV **Key:** cv **Agg:** avg **Format:** decimal #### Highest CV **Key:** cv **Agg:** max **Format:** decimal **Notes:** Thresholds follow common practice (X up to 0.5, Y up to 1.0, Z above 1.0). Many companies tune them. Use at least 12 periods of history and remove one-off spikes you can explain before judging variability. ## Content **Intro:** XYZ analysis sorts items by how predictable their demand is, the way ABC sorts them by value. Enter the average and standard deviation of demand per item and the sheet returns the coefficient of variation and the X, Y or Z class. **Sections:** #### How to use the worksheet **Body:** Export demand history by item from your ERP, one row per item, for the same period length (months or weeks). Compute the mean and standard deviation per item, then type or paste them into the sheet. The coefficient of variation and class calculate on their own. Add rows with the add button and delete the sample rows. When finished, export to Excel and join the result with your ABC classification to form a nine-cell matrix. **Bullets:** - Use the same period length for every item - Use at least 12 periods where possible - Strip out explained one-offs such as a single large project order - Re-run quarterly, classes drift #### The method behind X, Y and Z **Body:** The coefficient of variation (CV) is standard deviation divided by mean. It is unit-free, so a 50-unit item and a 5,000-unit item can be compared. X items have stable demand (low CV), Y items fluctuate with trend or season, Z items are lumpy or sporadic. Common cut-offs are 0.5 and 1.0, but they are conventions, not law. A Z item with a high CV is also the one where a statistical forecast adds the least and where safety stock policy matters most. #### How to read the result and what to do next **Body:** Combine with ABC. AX items are high value and predictable: tight min-max, frequent replenishment, low safety stock. AZ items are high value and erratic: protect with capacity or make-to-order, review weekly, avoid blanket high service targets because stock is expensive. CZ items are low value and erratic: carry a generous buffer or stock a single big lot, the carrying cost is small. Use the class to pick the planning method, for example a statistical forecast for X, judgement plus customer input for Z. #### Common mistakes and ERP notes **Body:** The most common error is mixing periods, for example some items on weekly and some on monthly buckets, which changes CV. Another is using shipments instead of true demand, so stockouts hide variability. Intermittent items with many zero periods get a high CV by nature, for those consider Croston-type methods or demand-frequency measures as well. In most ERPs, history comes from issue transactions or sales order lines, for example SAP consumption values MVER, Oracle item demand history, or SyteLine item transaction history. **Cta:** If you want this classification refreshed automatically from live ERP history and pushed back as a planning attribute for approval, Netray can build that over your ERP data, read-only by default. **Faq:** #### What CV thresholds define X, Y and Z? **Answer:** There is no universal standard. Many planners use X up to 0.5, Y above 0.5 up to 1.0 and Z above 1.0. Some use 0.25 and 0.5 for very stable ranges. Pick thresholds that split your own item base sensibly and document them. #### Should I do XYZ on units or on value? **Answer:** Use units, or quantity in a common base unit. XYZ describes demand behaviour, not money. Value is the job of ABC analysis, and the two are combined afterwards. #### How much history do I need? **Answer:** At least 12 periods, ideally 24 if there is seasonality. With fewer than six periods the standard deviation is unreliable and the class will flip from run to run. #### How do I treat items with many zero periods? **Answer:** Keep the zeros, they are real demand behaviour. Such items will naturally land in Z. For planning them, look at average demand interval and order size separately rather than relying on a normal-distribution safety stock formula. ## Search terms xyz analysis inventory, xyz demand variability, xyz classification excel, coefficient of variation demand, abc xyz analysis template, demand variability worksheet, xyz inventory classification, forecastability classification --- Talk to an engineer: https://netray.co/contact?intent=erp_ai | Machine catalog: https://netray.co/agent/catalog.json