The Lean manufacturing, PPT Plus,
"Back-to-Basics" e-Training Package 
Lean Manufacturing Solutions

Increase the effectiveness of your
Lean Manufacturing Training Program
Lean Manufacturing Simulation Game 

Cycle counting has two purposes: first, to find errors so that their causes, not just the errors, can be fixed, and second, to measure inventory record accuracy. It is totally unacceptable to use cycle counting to find only the errors and correct those. This is exactly the same logic bank tellers use each day when reconciling transactions with balances. They cannot simply "make up" a seventy-five cent shortage without first determining its cause. They must absolutely determine its cause, as next week's short­age could be $75.

Sampling

Cycle counting is sampling. Sampling is a term that has a specific mathematical meaning. It is a technique, in which certain members of a population are selected—called a "sample"—and a feature of that sample is measured. It is then inferred that this measurement is a characteristic of the population. If we were to look at a population that consists of a group of people, a sample might be selected and their height measured. If the sample has an average height of 5'9", it can then be inferred that the average height of the entire population is 5'9".

Sample Validity

Two characteristics of the sample determine how well the sample represents the population. The first is the size of the sample with respect to the population. If the sample is extremely large, it is generally more reliable than if the sample is extremely small. That is, if the sample size is 90% of the population, it would be an excellent representa­tion of the entire population. On the other hand, if the sample was only 1% of the population, it would not be as representative and thus less reliable.

The second, and equally important characteristic of the sample, is its stability. That is, if a sample produces a particular result and by increasing the sample size it continues to produce the same result, it can be assumed that the sample reliably represents the population. This is particularly important when the population size is un­known or extremely large. In the United States, this application of sample size is illustrated every four years during the Presidential election. The national television networks predict election outcomes with only a very small percentage of the ballots counted.

Sampling and Inventory Records

Whenever inventory records are being measured, and a program is being put in place to maintain inventory record accuracy, "sampling" is referred to as "cycle counting" or "cycle auditing," with cycle counting being the dominant term. The term "cycle counting" refers to the practice of counting a small portion of the inventories on a continuing and repeated basis. That is, if the inventory consisted of 10,000 part numbers and the company wants to count each part once a y*ear, 40 parts would be counted every day. This is distinctly different from the physical inventory in which all the parts are counted in a short time period.

Two Inferences

There are two important inferences that can be made when sampling or cycle counting inventory. The first and most obvious is: it can be inferred that the accuracy of the sample is the accuracy of the population. The size of the sample with respect to the population and the stability of the sample are used to determine how reliable this inference is.

The second and more important inference that can be drawn is: the type of errors and their causes found in the sample will also be found in the balance of the population. For example, we open a couple of boxes from a specific supplier that were each labeled one gross (144). We find, however, that one box contains 118 items and the other 120. We could then make the inference that other items from that supplier would probably also be short packed.
These two inferences make cycle counting the key tool in maintaining inventory record accuracy.

Types of Cycle Counting

All cycle counting techniques are exactly the same, with only one exception: how the sample is selected from the total population. Mathematicians tend to stress this selec­tion process. However, with regard to creating and main­taining inventory record accuracy, selection does not seem to make any significant difference, provided it is not tremendously biased towards any one small group of parts. Four different types of cycle counting techniques will be discussed: Control Group Cycle Counting, Random Sample Cycle Counting, ABC Cycle Counting and Process Control Cycle Counting.

To be Continued