October 09, 2005

Scenario: Chaos in New Orleans

A few weeks back I promised a longer rant about the New Orleans disaster and scenario planning. So here goes. Scenario planning is basicaly a tool for identifying possible futures and exploring how one might arrive (or avoid arriving!) at such futures.

Typically, scenario planning involves identifying key drivers of the future (social, political, technological, etc.) that have some level of uncertainty around them. You then map out different futures based on combinations of those drivers* (possible futures can of course be mapped out less rigorously -- just based on a good brainstorm, for example).

Now the key to scenario-planning is the reverse-engineering bit. Once you identify plausible futures, you ask yourself "How could we have arrived at such a future?" and, depending on how you feel about said future, "what steps can we take to increase or decrease our chances of getting there?"

The details of what a New Orleans hurricane disaster scenario-planning exercise might have looked like are not important at this point. It's not clear, for that matter, whether traditional scenario planning would have been the best forecasting and planning tool in this particular case (particularly late in the game). Nevertheless, one fairly constant characteristic of scenario planning is that, when done well, it gets you looking at worst- (and best-) case scenarios and thinking hard about how those come about.

So what astounds me about New Orleans is this: did no one envisage a thoroughly ravaged, anarchic New Orleans, ask themselves how such a scenario would come about, and then take even the most basic necessary steps to try to avoid it?

Well, almost. FEMA actually did run this type of exercise back in 2004, contracting a private firm to run a multi-agency exercise involving hundreds of emergency planning officials and a hurricane named Pam. There is a reason why the last step of scenario planning -- the steps you can take to increase the probability of getting to the future you like -- is so critical. All the evidence seems to suggest that it was on this last step that emergency planning officials failed miserably and inexcusably, seemingly neglecting to follow through sufficiently on what they learned from this exercise.

Bottom line: avoiding what took place in New Orleans should not have been rocket science. Grim scenarios about New Orleans were no secret. Even tourists learned about them. This was a case of too much scenario and not enough planning, and it didn't need to have been that way.


On a related (New Orleans) note, check out my "readings" side bar for two interesting tales of "private sector to the rescue" from Forbes (thanks to Nathan for the tip). Fedex and Wal-Mart apparently mobilized their infrastructure and supply chains to deliver goods and services that the governments involved couldn't.

[*]  The simplest version of this looks like a 2x2 grid based on two drivers (one per "axis" -- with the endpoints of the axis representing strong desirable and undesirable states for that driver).

Posted by anatole at October 9, 2005 03:40 PM

What astounds me even more is the lack of (planning? or maybe just reporting) of how to rebuild/build a better levee system. What's the point of rebuilding a city only to let such an event happen again?? Argh.

Posted by: Lana at October 12, 2005 10:31 AM