Big data!

(Thanks to The Planet for the picture of the working side of their data center)

Supply Chain Managers are awash in more data than ever before.  Each system implementation or process change is likely to create more granular information than previously thought possible.  This explosion of data comes as we may still be managing through legacy system issues remaining from acquisitions, 'not quite perfect' system implementations, poor master data management, or several other data challenges (do you have hardware that wasn't scaled to handle the amount of data you're now creating?  You're not alone)  - welcome to the Age of Big Data.

At the eyefortransport 14th Annual Logistics CIO & Supply Chain Technology Forum this afternoon I heard six Supply Chain thought leaders participate in a panel discussion on "Visibility, Technology and Methodology for achieving End to End Visibility".  Jeff Jones, VP of IS at UPS conveyed that UPS has petabytes of data (I think it was petabytes, it might have been exabytes or zettabytes, I know it was a very large number).  Try to imagine the amount of data UPS creates every day - every container they move, every order they pick, every parcel they sort, ship and deliver.  There are datapoints collected from their own systems, their transportation partners (subcontracted ocean, air, rail and truck freight), GPS tracking information on their rolling assets, structured data and unstructured data (images of bills of lading in addition to electronic file transmissions), they must literally be drowning in data.

Well, maybe not drowning in it.  Jeff also said that UPS is investing just under a billion dollars (billion with a B) in IT this year and much of it is focused on supply chain visibility - using all of that data to drive supply chain improvements.

Are you thinking that UPS is in a different league?  Your organization doesn't have a billion (with a B) dollars to invest in IT this year?  Well, neither does mine but that can't stop us from making progress.  I draw your attention to the City of Chicago's data portal which has a wealth of datasets easily accessible for ad hoc reporting via browser as well as API.  The City of Chicago isn't bursting with cash to spend on frivolous IT projects and it's just about the last place I'd expect to see implementation of systems that drive data visibility but they've done it.

What are your organization's plans for harnessing big data?  What types of data challenges do you commonly run into as a Supply Chain Manager?

Is more better?

I spent a morning this past weekend with a fellow Supply Chain Manager who recently arrived at his organization here in Chicago.  He and I commiserated about the vast amounts of capital tied up in inventory (and although my organization's inventory is largely non-perishable, his inventory was not and the clock was ticking on some of it).

It reminds me of a visit this past Fall to a vendor that has a significant business line building ASRS's (automated storage and retrieval systems - essentially very large, very costly warehouses) and the software that supports them.  Some of these warehouses are certainly necessary (a recent install of theirs was for storing library books at a major research library - if you don't believe that is a 'good' type of inventory you may want to read Nicholson Baker's Double Fold, a compelling book about the deacquisitioning effort that our libraries have undertaken).

As I wandered through this vendor's offices, conference rooms and rest rooms and was awed but the beauty of their architectural and engineering drawings of their installations (yes, the artwork in the restrooms consists of very nicely framed engineering drawings) I was also thinking about the dollars invested in the storage of inventory.  Many of these installations were built for storing tens of thousands of pallets and represent a significant investment to store inventory, which itself likely represented a significant investment on the balance sheet.  In other words, those drawings represented certainly tens and likely hundreds of millions of dollars of inventory investment.

The APICS Body of Knowledge tells us that the only reason to hold inventory is when doing so is less costly than not holding the inventory.  I question whether the business cases used to justify the construction of those many ASRS's hold water today.  Can the toolset of the Supply Chain Manager - Lean, JIT, Agile Manufacturing, substituting information for inventory - eliminate some of the inventory in these warehouses or even eliminate the need for the warehouse entirely?  Very likely.

My friend's inventory challenge was interesting in that as the product aged it could no longer be used for it's original, intended use and had to be downgraded for less value-added uses.  Although my organization's inventory is not perishable (at least not on the type of timeline that my friend faces) we still have opportunities to reduce the accumulation of inventory and always will - the Supply Chain Manager is on an never ending quest to balance minimal inventory investment and customer service.

Cutting your Inventory and Postponement

My local Costco added retail gasoline sales recently and I noticed that they carry only 87- and 93- octane (regular and premium), as opposed to most service stations that carry 3 grades of gas (or petrol for my friends on the other side of the pond).  I usually purchase premium for my own car (and regular whenever I have a rental) - I've never seen any statistics on it but it would be interesting to see what percentage of sales each of the typical 3 grades represent.

In addition to reducing the inventory items Costco needs to monitor and replenish they have reduced by 50% the number of underground tanks they install and maintain.  I'm not very familiar with gasoline retailing but there must be several other areas of payback that I'm missing.  I wonder why other gas retailers haven't followed suit.

I  noticed another major company undertaking a new inventory strategy recently.  Coca Cola is rolling out new soft drink dispensers that have 106 flavors, which you would think would cause their inventory to surge.  That's likely not the case as they are using a strategy of Postponement where the final mixing of recipe/parts is done as late as possible (in this case, literally when the consumer selects "lime diet coke" at the dispenser).  They can use generic parts for several final products - in this case that same lime flavoring is used for Coke, Coke One, Sprite etc.   I've heard that the machines even send electronic alerts when they need service or flavor refills.  If that's the case it still needs some fine tuning as several of the flavors are not available at my local lunch place more often than not .

Have you seen any interesting examples of inventory strategies recently?  Please let me know in the comments below.