Play Odds In Ordering

Probability influences our lives. Insurance companies base their rates on it. Gambling meccas are built on it. Life expectancies are plotted by it. And parts department stocking decisions are driven by it. Because you are dealing with an unknown (If I phase a part into stock today what is the probability that it will sell in the next 12 months) one would surmise that the higher the probability the

Probability influences our lives. Insurance companies base their rates on it. Gambling meccas are built on it. Life expectancies are plotted by it. And parts department stocking decisions are driven by it.

Because you are dealing with an unknown (“If I phase a part into stock today what is the probability that it will sell in the next 12 months”) one would surmise that the higher the probability the better the setting.

This is true to a point. So much rests on the criteria we use, departmental profitability, inventory movement, obsolescence levels and customer service. Accordingly parts managers continually search for the holy grail of phase-in settings.

As a parts manager, I looked at phase-in from a probability standpoint. We used the standard three in 12 (three months of sales in a 12-month period) as our phase-in setting, meaning there must be at least three separate months of sales within 12 months for that part, or it should not be stocked. From my own observations and calculations I concluded that it provided better than a 90% probability of actively reselling the part again over the next year. This is the setting I recommend as a starting point.

What is the probability of resale if various other settings were used for phase-in? The questions piqued my interest. For the answers, I sought professional help.

I contacted Frantisek Marko, an assistant professor of mathematics at Pennsylvania State University's Hazleton campus. He subsequently created a formula that calculates the probability of resale of a part in an ensuing 12-month period.

Here are the results of calculated phase-in scenarios based on his formula. I would note that the probability results are completely objective and mathematically calculated and therefore irrefutable.

Phase-In Criteria % Probability of Resale in the Next 12 Months
3 in 12 94.3%
2 in 12 87.5%
3 in 11 95.3%
2 in 11 89.1%
3 in 10 96.3%
2 in 10 90.7%
3 in 9 97.1%
2 in 9 92.2%
3 in 8 97.9%
2 in 8 93.7%
3 in 7 98.6%
2 in 7 95.1%
3 in 6 99.1%
2 in 6 96.4%

As you can see by the sample scenarios, all except the two in 12 and two in 11 yield better than a 90% probability. So the three-in-12 criterion stands as a good recommended starting point.

The remaining question is: Beyond the probability of resale potential over the next 12 months, given parts proliferation, what setting is best for phase-in?

For example, when I plug a three in four (three months of sale in a four-month period) into our probability calculator the result is a 99.8% probability of resale in the next year.

However, since the test period is only four months and three months of sales would have to occur within that time period, there would be many missed stocking opportunities due to the restrictive criteria.

The resale probability is very high but phase-in for the majority of parts, except those with extremely high sales velocity, would be impossible. On the other hand, the two-in-six criterion produces a high probability of resale, but is also a very aggressive phase-in.

The best approach would be to separate your inventory into as many sources as possible in order to apply varying settings. All parts departments are unique. Although everyone can begin from a similar setting like the three-in-12 standard, eventually your marketplace and customer demands will influence your phase-in criteria decisions.

You may decide to phase in everything at two in six, then further test some parts with three in nine.

For those instances where parts are forced into stock, either because they are unsold special orders, special-order returns, speculative purchases or the like, the calculated probability of resale over the next 12 months is a low 48%.

The parts manager is the final filter for determining what parts will ultimately be placed in stock. So there you have it!

Gary Naples is a parts consultant to dealers and manufacturers. He's authored two books on parts management. He's at 570-824-1528/[email protected].

Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish