Problem-solving and particularly the part that focuses on root cause analysis (RCA) has always been one of the topics that has had my special interest. Specifically, two questions always slumbered in my head, viz. (1) whether you could speak of one root cause, or that you should speak of multiple (root) causes; and (2) whether you should speak of the root cause or rather the root condition. In a series of six posts, I will try to explain how rigorous problem-solving logic (using an example) can help us answer these questions. At the same time, I hope the example and the logic will be of use in your problem-solving efforts or your coaching thereof. This third post will dive into tracing back the causal event chain, introducing and applying concepts like causing events, the initial causing event, and the initial active cause.
In the first post, I set the stage by defining the starting points and some basic initial concepts (like problem, cause, agent, target, event and Tripod Beta’s causal diagramming technique). I also introduced an example that I will use throughout the series. In the second post, I added some more concepts (necessary condition, defensive and control barriers). This third post of this series of six will dive into tracing back the causal event chain, introducing and applying concepts like causing events, the initial causing event, and the initial active cause.
Tracing Back the Causal Event Chain
In a gap-type of problem we are investigating a problem that occurred in the – when done well – very recent past (see also my blog post on 5xWhy? here http://dumontis.com/2011/06/5xwhy/). So, during the RCA we are looking for causing events that are upstream in time and space. Somewhere upstream, we will also find our well-known “root cause”.
Therefore, we need to trace back the time and space path upstream looking for the initial causing event, that – through a causal or event chain leading to other causing events (and under the presence of necessary conditions) – ultimately led to the observed problem. And unless there are two causing events at the exact same moment in time (which is not very likely in most practical cases), there can only be one causing event at each level of the causal chain.
So, how do you know whether to take the route of the agent or the route of the object at every level of the event chain? To find out, you determine the cause of each causing event in the causal event chain as described above.
In our example, the prime event of the person’s head hitting the hard floor was caused by the falling person as we have seen. Now we are looking for the event that made the person fall. The problem-solving team found out that the person fell off the vehicle due to a rolling pipe that knocked the person off.
The rolling pipe (the active agent and a problem) is the cause of the causing event where the person (the target) was knocked off the vehicle by the rolling pipe and fell. The falling person (the active target and a problem) caused the event of the person hitting the hard floor (the agent) leading to the head injury (the problem).
In the next step, the problem-solving team found out that the pipe started rolling at the moment that another pipe was loaded onto the stack of pipes already being on the vehicle.
Our causal diagram now looks like this:
Figure: the preceding events leading to the prime event.
In this way, you track back the causal event chain until its origin. The well-known 5xWhy? technique (see my earlier blog post here: http://dumontis.com/2011/06/5xwhy/), when used correctly, is very helpful in uncovering these causing events until the initial event is found. Another very helpful technique is to reconstruct the exact timeline of events related to all objects that played a role in the story of the problem, something I sometimes refer to as reconstructing “the film”. Key in these techniques, however, is to rigorously apply the introduced logic and to verify the actual occurrence of the events.
The Initial Active Cause
But how do you know you arrived at the initial causing event? Simply put, the initial causing event is not preceded by any other immediately preceding causing event related to either agent or target. Let’s look at our example again to illustrate the logic:
In the earliest causing event now under investigation, the team did not find anything inherently undesirable with the act of loading the pipe (the active agent) onto the stack of pipes already on the vehicle (the passive target). The act of loading the pipe was scheduled and executed properly and as such, it was not an undesirable situation.
There was, however, something wrong with the stack of pipes on the vehicle (the passive target). The pipes were not secured with stakes that standardly are on the vehicle that should have been used in loading pipes.
Then again, starting to use the wrong vehicle (without the stakes) did not immediately precede the causing event. Several pipes had already been loaded onto the vehicle without causing the pipes to start rolling. Therefore, the wrong vehicle was a necessary condition.
Consequently, the pipe loaded onto the stack of pipes triggering all pipes to start rolling cannot be considered a cause, as this pipe being loaded did not represent an undesirable situation. And the stack of pipes on the wrong vehicle, albeit out-of-standard, was a necessary condition as it did not immediately precede the rolling pipes. So, the stack of pipes on the wrong vehicle can also not be considered the cause.
The example shows that the initial causing event, although being preceded by another event, neither agent nor target can be considered a cause. As a reminder: a cause was defined as the active agent or target in the (prime) causing event, whereby the active agent or target in itself also represents a problem.
The causal diagram of the event chain has been depicted in the following figure, using an extended version of the TriPod Beta diagramming technique.
Figure: Tracing back the causal event chain up to the initial causing event.
So, would you now consider the rolling pipes the root cause? Probably you wouldn’t, and rightfully so. We could, however, define the rolling pipes as the initial active cause. An initial active cause thereby can be defined as the problem that was brought into existence by the initial causing event.
So how to proceed?
Next Post: Continuing the Logic
In the first post, I set the stage by defining the starting points and some basic initial concepts (like problem, cause, agent, target, event and Tripod Beta’s causal diagramming technique). I also introduced an example that I will use throughout the series. In the second post, I added some more concepts (necessary condition, defensive and control barriers). This third post traced back the causal event chain while introducing and applying concepts like causing events, the initial causing event, and the initial active cause.
My next post will take us to the systemic level of our problem and I will further explore the concept of barriers and make the link to standards in Lean thinking. At this point in the logic, I will also introduce the problem of occurrence and the problem of non-detection and how this can help us in finding the root of the problem.