Breaking it Down
by Anthony J. Nicolella
You are a buyer for a large buying activity. Your installation continues to grow substantially because of a recent Base Realignment and Closure decision. To keep up with the energy demands of the base, your installation commander was able to secure funding to build a state-of-the-art, high-voltage electrical substation. You have been given the task to buy all theequipment for this substation.
You received the purchase request packet for the substation equipment from your customer; it includes a large bill of materials (bOm), such as in figure 1. This is just one of many procurements you are working on, and you are wondering how you will ever find the time to develop your government objective for such a large buy.
Well, you are in luck. You can develop your government objective using stratified sampling.
A SEVEN-STEP PROCESS
Before getting started, let’s answer some basic questions regarding stratified sampling: What do we mean by stratified sampling? Where can i find more information about it? What steps are involved?
According to the
stratified sampling is a method used for cost/price analysis. it allows you to concentrate your efforts on the items with the greatest potential for cost/price reductions, while using random sampling to identify any pattern of overpricing smaller-value items. stratified sampling is not only a sampling method, but also
http://www.acq.osd.mil/dpap/dars/dfarspgi/current/index.html) list this technique in relation to spare parts and support equipment, to break down items into high-dollar value and low-dollar value strata.
There are seven major steps used in stratified sampling:1. Identify a stratum of items requiring 100 percent analysis. 2.Group the remaining items into one or more strata to facilitate analysis. 3. Determine how many items will be sampled in each stratum. 4. Select items to be analyzed. 5.Analyze all items identified for analysis-specifically, summing prices for the 100 percent analysis stratum while developing a decrement factor for any stratum being randomly sampled. 6. Apply the decrement factor to the total proposed cost of items in the stratum.
ANALYZING STRATUM 2
7. Sum the pre-negotiation positions for all strata.
Let’s apply all seven steps in analyzing this bOm. for simplicity purposes, we will only use two strata in our analysis.
in figure 2, transformers (xfrms) make up $375,000 of the total proposed cost of $413,900. This stratum alone represents 90.6 percent of the total proposed cost, which is significant. because of this, Stratum 1 will require 100 analysis,as indicated in step 1. Our should-cost objective for the transformers in Stratum 1 would be $368,000.
For the randomly sampled items (copper wire, capacitors, etc.) in Stratum 2, application of a decrement factor would be required as indicated in Steps 5 and 6. This stratum makes up only 9.4 percent ($38,900) of the total proposed cost of $413,900, which is not significant. Thus we do not want to spend an inordinate amount of time analyzing this stratum.
Our analysis of stratum 2 will be totally different than for Stratum 1, in that it will include two key components: computing and assigning random numbers to our items, and applying a decrement factor to come up with a range and target objective. Let’s see how this works.
Steps 3 and 4 in Figure 3 (determining how many items will be sampled in each stratum and selecting items to be analyzed) require us to generate and assign
random numbers to the items in Stratum 2. The easiest way to generate random numbers is by using the random numbers function in microsoft Excel. Just follow these five easy steps: 1. Go to formulas. 2. Select Math & Trig icon. 3. Select RAND. 4. Generate random numbers. 5. Grab bottom right-hand corner of first random number cell and drag it down until each item has a random number assigned to it.
Now that we have assigned random numbers to Stratum 2, we can start our analysis. To determine which items to analyze in Stratum 2, we randomly selected items that ended in a number of “54” and “31.” In our Stratum 2 example, only two items had these random numbers: copper wire and fuses, with totalproposed costs of $20,000 and $2,500, and should-cost amounts of $17,500 and $2,250, respectively.
For Stratum 2, total the proposed prices for all sampled items. Total the dollar differences between should-pay and proposed prices ($2,500 + $250), and divide that total difference ($2,750) by the total of proposed prices ($22,500). This technique gives more weight to the higher-priced sampled items in establishing the decrement factor, which is computed as follows:
We can take this decrement factor of 0.12 and come up with a recommended objective price of $34,232, or approximately 88 percent of the proposed $38,900. However, the confidence interval would be from $34,232 (88 percent of $38,900)to $38,900 (100 percent). Based on this, a target objective of $36,566 (94 percent of $38,900) would be reasonable.
A 100 percent analysis was performed on Stratum 1 (the high-dollar items), and a random analysis was performed on Stratum 2 (the remaining items). Now that this has been completed, we can move on to our final step: to sum the pre-negotiation positions for all strata.
For Stratum 1, that pre-negotiation position would be $368,000, and for Stratum 2 it would be $36,566 (the 94 percent confidence interval), giving us a total prenegotiation position of $404,566.
This article demonstrates how quick and easy it would be for a buyer to develop a government objective using stratified sampling. The next time your agency has a large buy, try using stratified sampling. it will provide high-quality results in a short period of time, and save the buyer and his or her organization valuable resources.
For more information, contact the author email@example.com.ANTHONY J. NICOLELLA, a retired Ar my of ficer who hel d numerous pre- and post-award contracting positions, is a Professor of Contract Management for the Defense Acquisition University (DAU) – South in Huntsville, AL. He holds a B.S. in logistics management from Pennsylvania State University and an M.S.A. in general administration from Central Michigan Universit y. Nicolell a is Level III certified in contracts and is a member of the U.S. Army Acquisition Corps. Before joining DAU, Nicolella was a Senior Buyer/Planner for NV Energy Inc. and a Supervisory Contract Administrator for the University of Nevada, Las Vegas.