1. Discuss the advantages and disadvantages of using analysis trees. Do you think using an analytical tree approach would be of value at the facility where you work (or a facility where you have worked if you are not working presently)? Or do you think a less elaborate method of analysis would suffice?

It is said that individuals can hold between five and nine related facts in their minds at a time. Most investigations have many more facts than that, and therefore using an analysis tree is advantageous to investigators.

A disadvantage of the analysis tree is that many people have added connections and relationships as they see them. This can possible mislead others into following an erroneous connection that another person might feel is a solid one.

While MORT (Management Oversight and Risk Tree) is said to be effective and was used extensively during the 1970’s and 1980’s there is a shortage of individuals who know how to use it properly” (Oakley, 2003, p. 102). No matter how great the tool, if people don’t know how to use it, it is useless.

We currently utilize flow charts for a myriad of uses. These are typically offshoots of the analysis trees, but don’t limit us to solving a problem or placing blame for an incident. Early flow charts were sometimes cumbersome, but today there are formats that are adaptable that we use. I believe that an analysis tree can grow too many “branches” to make it viable, but I do like the very simple idea of the MORT with color coding for adequacy. Colors immediately alert individuals to areas that need additional attention, whether it be a causal chart or a departmental responsibility chart. However, the idea that it needs to be simple enough to be effective should be first and foremost in design.

2. Describe the process of tier diagramming and how it relates to root cause analysis.

Basically tier diagramming is a method of breaking down broad categories into smaller ones, then breaking those smaller ones into finer detailed ones until all possible root causes can be identified. It represents a step by step thinking process.

First, and possibly the most difficult step is developing the overall desired accomplishment. What is it that needs to be understood? For example, if we want to know what caused a scaffolding to fall, we would begin with a simple question … what causes a scaffolding to fail or why does a scaffolding fail. Following this, investigators would brainstorm all possibilities following investigation of the incident involved. Each possibility would be written underneath or to the right of the main question. This is the first “tier”. Any relationships between the causes should have arrows linking them.

Once the first tier has been done, it should be analyzed to be certain that each item is necessary for the goal to be accomplished (the scaffolding to fail). Now each item on this tier becomes the new “goals”, and each item should have a tier below it. Then these items are checked for sufficiency to accomplish the goal above them, and so on. This continues until the item is so basic that it doesn’t have a multiple number of causes. These will be the root causes and can be used in the root cause analysis.

The last two steps are to validate the tree and develop corrective actions (Oakley, 2003, p. 105). Validation involves making certain that every bit of the information has been analyzed, and developing corrective actions will be the basis of correcting the event.

One of the positive aspects of diagramming, is quite simply, the visibility of thinking processes of several investigators combined in one chart. Written reports and verbal conversations are rarely so organized and linear. An extremely important aspect is understanding each step of each tier, and being certain that all technical information has been incorporated.

3. Compare and contrast the analytic tree approach with the fault tree analysis approach. Consider the pros and cons of the two types of tree analysis approaches.

The fault tree process asks the question just how did this happen as opposed to asking what causes a situation like this to occur. Using the scaffolding example, the analytic tree begins with how can a scaffolding fail? while the fault tree analysis asks what caused this scaffolding to fail? Like the name, the fault tree places blame as opposed to the analytic analysis which is just descriptive logic involving technical information. A fault tree analysis is quick and usually reliable. It produces meaningful data which allow for evaluation leading to improvement of any situation that is being investigated. Cons for the fault tree analysis is that all contributors to the event must be anticipated. Some investigations may end up being expensive and take a lot of the investigators time. This method also depends upon the accuracy and expertise of the analyst.(Fault Tree Analysis, n.d.)

Analytic tree analysis allows for ruling out basic information that is not less than adequate. For example, the scaffolding is created with rubber non-skid feet. However, this has nothing to do with our scaffolding collapsing. The rubber non-skid feet were not less than adequate in causing the fall of the scaffolding.

While fault tree analysis is very specific, analytic tree analysis is kind of like throwing things against the wall and seeing which ones will stick. It is certainly more thorough than the fault analysis, however the fault analysis is more direct.

Both methods are structured, easy to create and easy to understand. (Oakley, 2003, p. 112).

References

Fault Tree Analysis, (n.d.). Retrieved August 2012 from

Oakley J.S. (2003) Accident Investigations: Basic Theories, analytical methods, and applications. American Society of Safety Engineers: Des Moines IL., Cengage Learning