Is The Air Force's Billion Dollar Flop the Biggest Supply Chain Failure in History?
Six years ago, Supply Chain Digest published a piece on The 11 Greatest Supply Chain Disasters in history, which was updated in a blog post on The Top Supply Chain Disasters of All Time by Editor-in-Chief Dan Gilmore back in 2009 which added five new ones to the list, bringing the total to 16.
The top three were:
- the failure of Foxmeyer's "Lights Out" Warehouse,
which was the top disaster in the original report and wiped out the 5 Billion dollar company almost over night;
- the Boeing outsourcing fiasco,
which led to massive 2-year plus delays in the production and delivery of the long-awaited 787 Dreamliner and some 2 Billion in charges to fix supplier problems; and
- GM's Robot Mania,
in the 1980s when CEO Robert Smith pent 40 Billion on robots that didn't work for the most part
But SI thinks the Recent Air Force Modernization Effort should top the list. As per this great article over on the New York Times Site on the Billion-Dollar Flop, the six-year old effort that had already eaten up more than 1 Billion didn't even achieve a quarter of the planned capabilities -- with another Billion required to achieve that minimal target. This says that the effort, supposed to cost $628 Million, would require over 8 Billion to complete! This easily dwarfs the 2 Billion in charges plus losses due to delayed sales suffered by Boing and the 5 Million Foxmeyer failure.
Does it dwarf the GM failure? The failed gamble cost GM a lot, but they are still in business, and posted almost 1.5 Billion in profit last year. And they were able to fix their processes and technology and improve over time.
In comparison, the Air Force is stuck relying on legacy logistics systems, some of which have been in use since the 1970s. And it turns out that this failure is just the tip of the iceberg, with the Institute for Defense Analyses noting that modernization of the department's software systems, which has been a priority for 15 years, has cost over 5.8 Billion as of 2009 and most large operational software system efforts are still behind schedule. So now we're up to six billion.
And the losses mount for every year a legacy system (way) past it's prime has to remain in production. With today's rapid pace of software, and hardware, refresh cycles, it's often difficult to find a replacement part for a piece of hardware that is only 3-years old, and if you do find it, it's costly. The Air Force has to find replacement parts for systems that are 13 and 30 years old! And lets not forget energy and support costs! Older systems often consume way more power and require more support hours than newer systems. Plus, over time, the expertise in supporting such systems goes from relatively common to extremely rare as more and more people retire or move to different systems and technologies and no new people learn the antiquated systems. As a result, the expertise that remains becomes very costly as the few people left demand a premium and expenses mount when they have to be flown in from halfway across the country.
Plus, the failure has instilled a fear of future technology fiascos, causing them to impose an across-the-board deadline of 18 to 24 months for future upgrade projects. While this sounds good in theory, and an upgrade project for most systems generally shouldn't take longer, there are some systems where the requirements analysis is going to take 6-12 months and the migration plan, which will involve a lot of data mappings, development, and testing, will take just as long. Add a staged implementation plan, quality assurance, and user testing, as well as time for any customizations the COTS (Commercial Off The Shelf) Vendor has to make to the core system, and the project could take longer. So, this is going to prevent some upgrades from happening until COTS technology in certain area improves or a vendor is willing to bite the bullet and create the mapping middleware without a contract in the hopes it will get one. In the mean time, losses mount.
While SI does not have the data to calculate, it would bet that if you did a total loss analysis over all delayed and failed projects leading up to, revolving around, and including the modernization initiative, over the last decade, the number would be 5 times higher, just like the license cost of an on-premise software solution amortized over five years turns out to often be 1/10th of the total cost of ownership.
It might not add up to a 40 Billion loss yet, but by the time the Air Force recovers and modernizes all of the systems that need modernizing, it will likely get there.