Daily Archives: December 15, 2009

106 Discrete Steps to Global Trade

Stanford University Professors Warren Hausman and Hau Lee recently decided that an end-to-end model was required for global trade management and decided to research the requirements. Analyzing imported goods from Asia to the US in the apparel sector, Hausman and Lee identified 106 discrete steps in the global trade management process. One Hundred and Six. Wow!

They also found ample opportunities for ROI for investment and improvement of global trade processes. Specifically, they estimated that importers actively using Asian sourcing had an opportunity through automation to reduce their supply chain costs by a range of 0.6-2.2% of annual sales. This is a substantial level versus average corporate net profit margins in the apparel sector. For instance, at an average profit margin of about 6%, such a decrease in costs would boost the corporate bottom line by 10% – 37%.

So, if you have IT-enabled global trade management, you:

  • have enhanced efficiency
    as you don’t have to manually execute 106 steps
  • are significantly safer
    the visibility lets you corrupt hiccups before they become costly seven or eight figure disruptions
  • have profits a-plenty
    as you’ve just increased the bottom line by 10% to 40%

So if you don’t have one, go get yourself a GTM solution today! Need a provider listing, start with the resource site.

Share This on Linked In

Inventory Accuracy Isn’t Rocket Science! Why Are You Still in the Dark Ages?

A recent article in Supply Chain Digest on Measuring Inventory Accuracy started off by nothing that the experts are saying that there’s no clear answer because there are so many ways to calculate inventory. WTF? Did I read that right?

The article then went on to say that the editorial staff at Supply Chain Digest did some informal networking and found that the answers were all over the map. Really? Who did they ask? Cave Trolls? We’re not in the supply chain dark ages anymore or, at least, we shouldn’t be.

So what did the editorial staff find in their investigation for an inventory accuracy calculation which defines the expected variance between book inventory and actual count? They found:

  • Jim Tompkins of Tompkins Associates offered the following formulas:
    • Financial: (Reported/counted Value inventory-System inventory Value)/Expected inventory value
    • Operational: Total inventory UOM Variation/Total Expected Inventory
    • Locational: Number of locations with variances/Total locations
  • Dave Piasecki of Inventory Operations Consulting noted that:
    every accuracy measurement is flawed in itself in that it can’t by itself show you a true picture of your accuracy and that you have to devise an appropriate “composite score”
  • Ken Miesemer of St. Onge recommends:
    cycle counts by location or geographic counts (an aisle or two at a time)
  • Doug Baker of Istoner states that:
    they rely on absolute and net dollar variance as well as unit variances from the cycle count processes

Ugh! I don’t get it. I know each of these experts has heard of RFID and the Internet and should know that this isn’t a hard problem anymore. At a high level, here’s what you do.

  1. Slap an RFID on each shipping unit — be it a unit, box, or palette — as it’s produced and enter it into the system.
       Now you know how much you’ve produced.
  2. Each time it enters or leaves a location, scan it.
       Now you know how much should be at each location.
  3. Use a supply chain visibility solution to link up with your retailer’s POS systems and have them upload a feed of units sold every day.
       Now you know how much is left at the retailers and you instantly know, at any time, the upper limit of how much inventory you have in the chain. Actual inventory is last count minus sales since last POS feed minus theft since last physical count.

Now, if you also use the system to track thefts and calculate average historical theft rate by SKU category (by day) and average daily sales rate, at any time you can produce an inventory count that is expected accurate within the sum of the (daily) theft variance and sales variance. Pretty easy, eh? And all you have to do is use the modern supply chain technology systems you should have been using for at least the last half decade. Any questions?

Share This on Linked In