Bad things happen. Natural events, such as hurricanes, wildfires, floods, and earth-quakes, kill, injure, and create destruction over significant areas. Human-caused incidents, ranging from industrial accidents to deliberate acts of terrorist or criminal violence, can similarly injure or kill people, damage property, and disrupt daily life.

Recognizing that disasters will occur, we make investments in emergency preparedness. We train firefighters to deal with everything from everyday kitchen fires to wildland firefighting operations that may involve hundreds or even thousands of responders. We store relief supplies in warehouses, for delivery to flood victims who have lost their homes and are temporarily...

In this chapter, we introduce and demonstrate reliability analysis. We begin with a brief introduction to the way reliability analysis is done in the technical field, the source of the techniques and concepts we are drawing on and adapting for this work. Making the transition to preparedness, we then review how concepts similar to those we explore here have been applied to the analysis of emergency response systems.

The heart of the chapter is a demonstration of a reliability analysis on a highly simplified example response system. In this illustrative analysis, we show how answers to the three questions posed...

To develop an approach for measuring the reliability of a response operation, it is necessary to examine a realistic emergency scenario requiring more complex response operations than the simple example described in Chapter Two. The characteristics of an incident define what the requirements are for response operations and what “success” at responding would mean. This includes the specific capabilities¹ that are necessary to respond and the scale of those capability requirements. The specifics of a particular emergency or disaster also shape the variety and importance of potential complications that could hinder response efforts and get in the way of responders...

The first step of the FMECA analysis that we are adapting to examine preparedness for large-scale response operations is to define the individual parts of the system that are being assessed and identify their functions. Doing so requires building a block diagram of the system elements and their linkages to one another, and articulating what it means for the pieces of the system to work well. This chapter does both these things for the chlorine response scenario described in Chapter Three.

To address the requirements of an emergency situation, response organizations deliver capabilities that can accomplish the necessary tasks.

To...

Having mapped out the response system and identified the linkages among different activities and components, we come to the second step in our adapted FMECA, which is to identify failure modes. Failure modes are “the observable manners in which a component fails” (Ebeling, 1997, p. 168). Identifying failure modes for an entire system requires systematically walking through “what could go wrong” at each point in the system that would observably affect response performance. But failure mode analysis is more than developing a list of potential problems that could get in the way of response.

In contrast to the example in...

Though simply identifying failure modes and describing their impact on different parts of a response provide some insight, the picture is incomplete without the information needed to determine which failure modes are more important than others. That understanding requires estimating the probability of different failure modes occurring and assessing the consequences for response performance if they do occur.

For an actual jurisdiction evaluating the reliability of its response plan for a large chlorine release, the assessment of the probability, effects, and severity of different failure modes could potentially be done in a number of different ways. Planners in the area...

The premise of this work was that adapting techniques from reliability engineering and risk analysis for evaluating the performance of technical systems could contribute to better ways of evaluating preparedness and anticipating the likely future performance of emergency response systems for large-scale events. We believe that premise has largely been proven out, with both the process of such analyses and their results potentially contributing to preparedness planning and evaluation in different but complementary ways.

The methods that we have described are not entirely novel, nor is the use of concepts like response reliability unknown in previous analyses of response performance....