How Much Information?

We know that it is possible to have too little information for disaster assessment and decision making.  But is it possible to have too much?  Don't we want to know everything that is happening, in as much detail as possible? 

 

The answer depends on who you are, where you sit, what you do, and for what purpose you need the information.  For example, if you are managing response by a public utility in a small sized city or large town, you really do want to know if there is an outage impacting one block in one neighborhood, and you want to know the house numbers involved.  That information is important to getting the right repair crew to the right address at the right time (considering all the other demands for service) to fix the problem.

 

But as we move up the chain of increased responsibility for broader problems:

 

... if you are the chief elected official of the jurisdiction, you may want to only know the general impact and the general neighborhood,

 

... if you the person responsible for assistance to utilities in the state emergency operations center, you only want to know specifically about this outage if there are unusual technical problems involved - otherwise the total outages in the city and whether assistance is needed from other resources is enough information.

 

... and if you are the governor you just want the numbers, how many outages in the state (unless there is a really good human interest story here that can be used to make a policy or budgetary point).

 

Each of the people involved has different information needs for fine details, which generally get increasingly less relevant as you go higher up the chain of responsibility and authority.  As you go further up the chain the data is consolidated with data from other locations to form a more general and broader picture of the event.  However, there will be incidents in which fine detail becomes important for any of a wide variety of issues.  These may be technical, resource allocation, media attention driven, or political issues that can disrupt the response.  In these cases the levels that deal with details have to identify and highlight the problems for attention by higher levels of management, and where the higher levels have to be able to reach down the chain to recover the detailed data.

 

So why not simply push all the house numbers in our case up to the Governor as a routine matter?  That way he has everything he or she needs when it is needed.  The truth is that too much information slows decisions, and may result in wrong decisions by individuals without specific training in the technical details.

 

And there is a related problem.  The reality is that getting more and more information until you have everything is just not very easy to do.  Some of this is initial uncertainty; some of it is the size of the problem.  The key is to decide when good enough is good enough, and to realize that getting to perfect is time and resources intensive while not making a major impact on the quality of the decisions. 

 

All of this means that at each level we have to define the key elements of information the decision makers at that level will need.  Then the emergency management system has to be able to provide that information in a format that the decision maker can easily use.  In some cases this will be specific numbers; in other cases a RED-YELLOW-GREEN color coded continuum is not only adequate but preferable.  Although the old Homeland Security color coded terrorist threat conditions proved to be an inefficient irritant, the color coding is not inherently bad.  Its application must be tied to decision needs, as must all of the information presented for decision making.