Q
In our small DC, we have set up 27 cycles based on concrete dates. However, shouldn’t other factors influence the cycle? And which products should we count? Randomly picking products does not seem efficient.
A
BILL KUIPERS
Principal, Spaide, Kuipers & Co.
It’s easy to get tangled up in a complex cycle counting program that is so difficult to manage and execute that it doesn’t have anywhere near the desired results. It sounds to me like you’re trying to slice and dice with a little too much precision. I prefer a somewhat more general approach.
You should start with the fundamental assumption that aggressive cycle counting is treating the symptom rather than the problem — the horse is already out of the barn. Accurate inventory doesn’t come from frequent cycle counting; it comes from good controls and inventory handling practices. Anybody can count. The hard part is knowing what to count and coordinating the timing of the events. So that’s the place to start.
Ranking items into A, B, and C groups (either by value or volume) should be the basis for cycle counting schedules. In general, A items should be counted once or twice a month, B items once or twice per quarter, and C items once or twice per year.
The trick to actual counting is a combined approach of counting specific items as well as counting by location. You can’t trust cycle counting by just counting where the system thinks the item is located: By counting an entire rack or range of racks, you will pick up stock that is not in the recorded location. Specifically, I like to count the active pick location of an item according to the ABC schedule, and at the same time count whole rows of the reserve stock area on a similar schedule. FYI, most cycle counts find more stock than the system expected. This is often caused by receipts, transfers, and returns never entered.
The secret of a productive cycle count is good identification of stock, easy and disciplined data collection, and sound systems.
The newsletter Distribution Digest (Jan. 8, 2004) recommends taking these six steps to start and maintain a cycle counting program:
-
Measure the accuracy of your current inventory reports. This will help you find areas on which to concentrate your efforts.
-
Determine what level of accuracy is acceptable. This level can be different for various SKUs.
-
Review and design. Evaluate how your current inventory practices affect accuracy. If necessary, revise and document them in a reference guide.
-
Develop initial inventory balances. Start with an accurate inventory. This may require a physical inventory, or you could simply use the first round of cycle counting as your baseline figure.
-
Begin cycle counting. Once you start, it is important to identify and review any trends or issues. For example, are certain items always off? If so, investigate why.
-
Perform the inventory counts just before or after normal working hours, when inventory isn’t moving. This greatly reduces the chance of counting errors.
A
DEBRA ELLIS
Principal, Wilson & Ellis Consulting
The most common methods for cycle counting are geographic and ranking. They may be exclusively implemented or combined to create a customized solution.
The geographic method starts at one end of the warehouse and systematically continues until every item has been counted. It is a simple process and relatively easy to implement. It counts every item an equal number of times each year.
Multiplying the items by the counts and dividing the total by the count days determine the count units. For example, 2,000 SKUs (stock-keeping units) counted three times a year with 27 count days yields 222 SKUs to be counted. The idea of cycle counting is to maintain accurate inventory with minimal disruption. The 27 days may need to be expanded if the SKU volume is large.
The ranking method establishes the count pattern based on product movement and/or sales volume. Some companies choose either movement or volume. I recommend a combination of the two. High-volume movement creates more opportunities for inaccuracies. High dollar volume creates a higher financial risk.
Create two ranking reports: one by dollar volume, the other by units. Assign the SKUs that account for 80% of the volume an A ranking; the next 15%, a B; and the final 5%, a C. Inventory typically follows Pareto’s Law, with 20% of the SKUs accounting for 80% of the volume.
Using the SKU count from the geographic example:
- 400 A SKUs counted four times per year = 1,600 counts
- 600 B SKUs counted two times per year = 1,200 counts
- 1,000 C SKUs counted once per year = 1,000 counts
- 3,800 total counts/27 count days = 141 SKUs counted per day
The ranking method increases the counts for the high-volume items while decreasing the items counted per day.
Choose the best method for your product lines and movement, and then adjust it as needed. Rather than choosing 27 targeted days for cycle counting, consider incorporating the process into your daily routine. It will create a consistent pattern and improve the overall success rate.
A
JEFF ZEILER
York Consulting Group
You are correct — there are more effective ways to set up a cycle counting program than randomly selecting products, which means that you treat all products equally, and that is not reality. Typically, 20% of your products represent 80% of your shipments. The remaining 80% of your products are progressively less and less important to your company’s bottom line. So, how do you construct a more real-world cycle counting program?
First of all, you need to know your product and the value you put on counting it. You do this by grouping products into categories based on several criteria. Usually the number of times the product ships and the dollar value of the product itself are the critical criteria used in the analysis.
Classify products into at least three groups and assign a code to each (A, B, and C are the industry practice). Fast-moving and/or high-value products are A items, followed by medium-moving products that are B items, and finally slow-moving, low-value products that are C items.
Then, with agreement from your finance department, agree to a counting cycle for each group. Again, the typical count cycles are monthly for the A products, quarterly for the B products, and annually for the C products.
Another way to create counting tasks is to do a cycle count whenever the on-hand inventory becomes zero in a location. This counting can be very fast, since the person picking the location empty can perform a count while he is at the location. If you have a real-time warehouse management system, the counts can be entered at the time of counting; otherwise, the counts will need to be recorded on paper and entered later.
As products come and go, be sure to periodically reanalyze the product movement for changes to the ABC classifications.
There are more complicated cycle counting programs, but this presents the basic method for setting and determining cycles. After this, you need to set up procedures for performing counts and re-counts, adjusting inventory as needed, and measuring inventory accuracy as a benchmark for your continuous improvement.
By getting proficient at cycle counting, it is possible to completely eliminate the need to do a physical inventory, which may force you to temporarily shut down your operation. For that, a dollar value can be assigned to provide an added justification for the cycle counting effort.