MT437 Unit 8 Cycle Count Plan Example

About 397 Counts a Day: An MT437 Unit 8 Cycle Count Plan for 31,200 Active Parts

[Student Name]

Purdue University Global

MT437: Strategic Warehouse Management

Unit 8 Cycle Count Plan

[Instructor Name]

[Date]

Lookout Parts Supply, its parts and their annual dollar movement are composite; counting rates and tolerances are stated assumptions, and the sources cited are real.

Why Cycle Counting

Today, Lookout Parts Supply shuts the building for two days each January for a full physical inventory, and in between, the system's quantities drift from what is on the shelves. Unit 7 found pickers spending 11 percent of direct time searching and scanning, much of it at locations whose stock did not match the system. DeHoratius and Raman (2008) found that inventory records at a large retailer were wrong for about 65 percent of items, often because of errors that accumulate quietly between counts. Cycle counting replaces the annual shutdown with a steady schedule that finds and fixes errors while they are small.

ABC Classes

Parts are classed by annual dollar movement, units shipped times cost, because errors in high-movement parts cost the most. About 2,500 A parts, 8 percent of the range, carry about 64 percent of annual dollar movement. About 6,860 B parts, 22 percent, carry about 26 percent. About 21,840 C parts, 70 percent, carry the remaining 10 percent.

Count Frequencies and Workload

A parts are counted monthly, twelve times a year, 30,000 counts. B parts are counted quarterly, 27,440 counts. C parts are counted twice a year, 43,680 counts. The total is 101,120 scheduled counts a year. Over 255 counting days, that is about 397 counts a day.

Trigger counts are added whenever the system shows a location at zero, a quantity goes negative or a picker reports a location short. Based on last quarter's exceptions, these add about 60 counts a day.

At about 45 locations an hour, including travel and recording, about 457 counts a day need about 10.2 counter hours, a little over one and a third people on a 7.5-hour shift. The plan staffs one full-time counter and a second counter for about three hours a day, drawn from the receiving team freed by the Unit 3 two-track process.

Counting every A part every month costs about two and a half counter hours a day; skipping it costs searches every night.

Tolerances and Accuracy Targets

A count is accurate if it falls within the tolerance for its class. A parts have no tolerance: the count must match exactly, because they move fastest and are worth the most. B parts are accurate if within one unit or 2 percent of the system quantity, whichever is larger. C parts are accurate within 5 percent.

The accuracy targets are 99.5 percent of A counts, 98 percent of B counts and 95 percent of C counts, measured monthly.

The Counting Process

Counts are blind: the counter sees the location and part but not the system quantity, so the count cannot be anchored to what the system expects. Any count outside tolerance triggers an immediate recount by a second counter. If the recount confirms the difference, the inventory control lead investigates the last transactions at that location, a missed putaway, a mis-pick or an unrecorded transfer, before adjusting the system. Adjustments over $500 on an A part need the inventory manager's approval. Richards (2018) stresses that the value of cycle counting comes from finding why records are wrong, not only from correcting them.

The Counting Window

Counts run on the day shift, between the end of night picking and the afternoon putaway wave, when the night pickers have gone and the forward zones are quiet except for rush orders. Locations being replenished or put away are skipped and rescheduled the next day, because counting a location while stock is moving creates false variances.

What the Plan Will Cost and Save

The plan needs about 10.2 counter hours a day, about 2,600 hours a year, roughly $62,700 at $24.10 an hour, most of it from receivers freed by the Unit 3 change rather than new hires. Against that, the January physical inventory takes the whole building for two days, about 1,100 labor hours plus lost shipping, and the nightly searching recorded in Unit 7 costs far more over a year. The plan is not free, but it replaces costs the company already pays.

Ending the Annual Shutdown

If the plan meets its accuracy targets for twelve months, the company's auditors can be asked to accept cycle counts in place of the January physical inventory, saving two days of closed operations a year. Frazelle (2016) notes that a well-run cycle count program is usually more accurate than an annual count, because errors are found within weeks rather than months.

What Could Go Wrong

Cycle counting fails when counts are skipped on busy days and never made up. The plan therefore tracks counts completed against counts scheduled each week, and any shortfall is carried into the next week rather than dropped.

Recommendation

Start cycle counting with A parts in the first month, add B and C parts in the second, report accuracy by class monthly and review the class boundaries every six months as movement changes.

References

DeHoratius, N., & Raman, A. (2008). Inventory record inaccuracy: An empirical analysis. Management Science, 54(4), 627-641. https://doi.org/10.1287/mnsc.1070.0789

Frazelle, E. H. (2016). World-class warehousing and material handling (2nd ed.). McGraw-Hill Education.

Richards, G. (2018). Warehouse management: A complete guide to improving efficiency and minimizing costs in the modern warehouse (3rd ed.). Kogan Page.