Resiliency

"Resilient" describes a system comprised of the user and computer joint team (Joint Cognitive System) that is able to identify, react, and succeed in the face of unforseen or unpredictable circumstances. The following are examples of patterns in the world that make decision making difficult. The RCS approach, Applied Cognitive Systems Engineering (ACSE), produces systems that overcome these patterns.

It is inevitable in some domains that the amount of data available to practitioners is virtually impossible to parse through piece by piece in any timely fashion. In these types of domains, the user is most often looking for patterns in the information, or anything unusual in the information. The resilient solution to this type of problem often involves providing the analyst with an overview of their entire information space (for broadening checks), and in parallel providing the analyst the chance to focus (for narrowing), on particular interesting bits of information.

In some applications decisions must be made quickly. And it is often in these types of domains that the consequences of a poor decision could be extremely damaging, if not fatal. The resilient solution to time pressure provides the information needed for the decision explicitlly on the screen, within the context that is relevant for the decision making. This reduces the cognitive work of the decision maker, as the information is displayed exactly as an expert decision maker in that domain envisions it.

Over contrained resources force practitioners to make judgement calls about priority of demands, time availability, and cost in dynamic and forward moving situations. The resilient solution allows users to view the whole set of available resources in context of the set of competing demands. Users make decisions based on satisficing the SET of demands, so that a reasonable solution is achieved.

Automation performs well in nominal cases. However, when the situtation becomes more complicated, users often shut-off auto and prefer to solve the problem on their own (Woods and Hollnagel 2005). The resilient solution to clumsy automation is to provide users observability of the automation, so that the "thought process" of the automation is evident to the users. If the user does not agree with the direction auto is going, he/she can direct the automation to perfom in a more reasonable way. The user is then able to supervise the automation and work in tandem with the auto, like he/she would with any other team member.

Once a "plan" is determined, users are usually tentative to change it. However, the resilient solution provides users with insight into the consequences of their changes, so that users can react to changes in demands or guidance in a confident, and well informed manner.