1)    Clues on finding causes…

When describing causes, you are searching for words to adequately describe the cause. Clues for finding the causes and for the description of these causes can be taken form the “effect” that you are questioning.

Example:  “Negative publicity”

We are looking for at least two causes here (as in each effect should have at least two causes). Whilst searching for the reason of why we have any publicity at all, which is probably the first cause, we also need to understand what makes that publicity negative, because not all publicity is negative. This would then be the second cause (or perhaps there may be more than one cause for the negativity so then second, third and/or fourth causes may follow).

It should be noted that whilst searching for at least two causes it is not restrictive to just two causes. This is the basic requirement and yet if diligently challenged more causes may still be found. It is the exhaustive search for causes that will provide the most opportunity for control of your problem, as the more causes you find, then the more opportunities you will have to control change or mitigate aspects of the problem.

Example:  “Hand bleeding”

If we take “hand bleeding” as an example, we are trying to understand what has happened. “What were you doing?” is a good question to ask and then “what happened?”. But if I am really perceptive and I examine the word “bleeding” then I should understand that I must have “blood” to bleed. So for the hand to bleed something must have happened, ie …you cut your hand and you bleed because there is blood in your hand.

Example:  “Delayed shipping”

If a cause has the word “delay” in its description, as in “delayed shipping”, then what does this mean? Delay is a reference to time so in the causes of the shipping delay there must be some reference to time otherwise we haven’t really understood what we are looking for. So for “shipping delay” the causes could be something to the tune of “a shipment is due” (time reference) and “we didn’t ship it in the time frame required”. If you never have a time frame then you can never be late so you need to establish the time frame in the causes of “delayed shipping”.

This basic understanding could then be applied to all causes.

2)      Understanding conditional causes….

If you follow the Apollo methodology of causal analysis, you need to find at least one action and one condition for each effect, but typically there are more conditions. The context of many conditional causes is that it, something must exist. Now what does this mean? E.g. To be able to break a rule firstly there must be a rule. The rule must exist. The same can be said of not following a procedure, or not meeting a standard, not following a protocol. In each case the rule, the procedure, the standard or protocol must exist otherwise you can never say that you failed to follow it or achieve it.

Example:  Radiator Hose

Take a radiator hose for example. Your problem would be that you have lost your coolant. Why? Because the radiator hose has burst.  So? Why does this mean that you lose all of your coolant? Because the coolant flows through the hose …..otherwise why would you lose any coolant at all?

For a hose to rupture there must be a hose. The hose has to exist. Without the hose there would be no ruptured hose. So now you have …….. a “hose connects motor to radiator”(the hose exists), “the hose transports coolant” (it has a purpose)and “that hose has burst”(this is what happened).

There must be a relationship between the hose and the coolant otherwise the hose rupturing will have no impact on losing coolant. The relationship is the condition in each and every case as is the existence of the item in question. The happening, the event, would be the “action”

Example:  Bolt fell out

For a “bolt to fall out” for example, you must first “have a bolt”. The bolt has to exist but what does the bolt do? It performs a function. It has a purpose.

Rather than saying that “you have a bolt” as one cause of a “bracket failing” for example, simply recognising the fact that the bolt exists as your causal description, it would be more precise to say what purpose the bolt has, that the “bolts secure the bracket” (thereby recognising that you have a bolt) and that that “bolt has fallen out”. What will happen if the bolt falls out? Will it produce the effect you are trying to understand? If it does then you have got it right. So if the “bracket is secured by bolts” and those particular “bolts fall out” won’t the bracket “fail to perform its function”? Unless you can establish the relationship of the bolt to the bracket, then the bolt falling out will cease to have any impact on the failure of the bracket.

Example:  Pipe dropped

In another example the “pipe dropped” because the “scaffold collapsed”. For this to make any sense we must first need to understand the relationship between the two. So unless you put the pipe on the scaffold then the scaffold collapsing would have absolutely no bearing on the pipe dropping. So to clarify then the “pipe dropped” and this was caused by the “pipe being on the scaffold” when “the scaffold collapsed”.  A way of testing the logic of this connection is to work the connection from right to left. i.e….. if the “pipe is on the scaffold” and the “scaffold collapses” will the” pipe drop”? Now this statement is, to all intents and purposes, quite logical but there is a catch here. It will only fall if it is unsupported by anything else. You have just identified another cause. So in reality what you are really saying (working from right to left) is that if the “pipe is on the scaffold” and is “unsupported by anything else”, then when the “scaffold collapses” the “pipe will drop”. Does this make sense? Yes it does. So you now know that the logic of that connection is sound. You have tested the logic of the connection.

HINT…In other words it should make sense if you were to read it from left to right which is how the “caused by” logic works, but that it should also make sense when you read it from right to left. Actively look for any exceptions that make a lie out of this statement. If you find an exception then you have effectively found another cause to add to your list.

3)     Too many words…

….another area that could be improved upon is in regards to the quantity of words that are used to describe causes.  A common pitfall is to use too many words to describe causes. What is then far more likely to happen is to combine a couple of causes into one cause box. The impact of this will be displayed in the logic of the next connection. Because the cause is now far more complex it is a lot harder to clearly identify specific causes. Often causes will be related to a part of the effect but not the other part. So the logic starts to break down and the chart becomes harder to understand.

Example:  “non-return valve failed”

Why …because it was “old and not maintained”. You could write this description in one box but then when you ask the next “caused by” question you will get some answers that relate to “Old valve” and you will have others that pertain to “not maintained”. If this were to be included in one connection then it would fail to make sense. The age of the valve and the maintenance of the valve are separate questions and the causes for each of these need to be separated for it to make any sense in the chart.

By being concise in the description, limiting the description to just 2 or 3 words, then this combining of causes becomes less possible and the chart will be easier to understand.

 

4)     Actions or Conditions

 

There is often confusion with labelling causes correctly. Some people will label a cause as an “action” whilst others will label the same cause as a “condition”. Does this matter? To some extent the answer is yes it does. This is based on the understanding that it is easier to control a “conditional” cause.  It is controllable. It tends to deal with tangibles.

The “actions”, the things that happen instantaneously, are far harder to control. They are unpredictable otherwise we would have done something about it already, and they are also largely centred around the people elements and therefore it is difficult to get reliable, consistent outcomes from controls centred on these types of causes.  So then yes it is important to try and label causes correctly.

Example:  “Shaft is worn”

If I were to say that the “shaft is worn” would this be an action or a condition? It is a condition isn’t it? It is the state the shaft is in prior to anything happening. If I were to say that the “shaft is wearing”, what would you say? Is this an action or a condition? It is the “action” cause that leads to, or causes, a worn shaft. If however it is wearing and wearing and wearing, doesn’t this become a condition over time? It is the same thing happening over and over again.

It is all related to how you see things. How causes are described by individuals will have a direct impact on how causes are subsequently labelled….as either actions or conditions. In other words there will be some variations.

Perhaps it is important to understand what is behind the labelling process. i.e. the labelling of causes as either actions or conditions. To me it is really about understanding what you have found but also then of what needs to found – to discover what is missing and then to search for it. It is about ensuring that we ask the next question – that we don’t miss anything. It is about that exhaustive search for all causes. This is the real reason behind the labelling process for me.

If you have two conditional causes for an effect essentially you should automatically be searching for an action that made use of those conditions. The labelling process prompts you to ask the next question.

By being diligent in this pursuit you will find more causes. More causes are better because you now have more opportunities of control.

 

 “What is the difference between actions and conditions”?

Well the conditions were already there. It was like that. The conditions pertain to the ability of something to happen. The action makes use of the conditions to create an effect. It tends to be instantaneous in terms of time. It is that thing which is different, that has changed.

Example:  Fire

If I were to ask “What are the causes of a fire” what would you say? Well there needs to be some fuel. There needs to be oxygen. There needs to be an ignition source. Well you can have paper, oxygen and matches but that doesn’t mean you have to have a fire. Something needs to happen. Something needs to make use of the conditions that are present, the ability of fire to happen, to actually cause a fire. This is what has changed, happened. This is the action.

It is that single moment in time when the action makes use of the available conditions to create an effect. If even one of the conditions was eliminated the fire would have no opportunity to occur.

There is tendency to see the obvious however. We see the things that happen, the things that people do. These are all the action type of causes. What we fail to see as easily are all of the conditions that are there. The benefit of finding the conditions is that conditions are easier to control.

 

5)    Why does anything happen???…..because it can!

This simple statement should bring to mind what all of the conditions are that need to exist for an event to occur. When you examine the last part of the statement you are essentially trying to understand the ability of something to occur….trying to understand what conditions are present.

Example:  Why does an object fall?

For example … why does an object (anything for that matter) fall?…It falls because it can fall.

Why can it?  Because it is at height (elevated above a surface)….because it is heavier than air, because it is in the presence of gravity. Surely all of these conditions must be true for something to fall, to be able to fall? Then something happens that makes use of these conditions….the action.

 

Written by Jack Jager

Apollo RCA Trainer for ARMS Reliability

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Post Navigation