"There are a thousand hacking at the branches of evil for one who is striking at the root, and it may be that he who bestows the largest amount of time and money on the needy is doing the most by his mode of life to produce that misery which he strives in vain to relieve."
― Famous quote by Henry David Thoreau from Walden, 1854
― Famous quote by Henry David Thoreau from Walden, 1854
To find out the solution to a problem, we need to go to its root. This is one of the fundamental principles of management. There are a myriad techniques that have been invented to find the causes of an event. To be more precise, industries go through highs and lows on an everyday basis. To combat a certain effect that has taken place, one needs to go to the base of the situation. Thus, by means of this step-by-step method, one will finally unearth why the effect was so. This process is called root cause analysis. It is basically a procedure to find out the root cause of a problem and then make arrangements to solve it. It is one of the most widely used methods in almost all known applications. The write-up below will give you a list of some of the best root cause analysis tools and their comparison.
Comparison of the Tools
RCA tools have major similarities and minute differences. They are used according to the situation and the application. Certain criteria based on which they can be compared include:
Clarity of the Problem Statement
- The tool must be able to define the problem statement accurately, and most of the root cause analysis tools do score in this department.
- The prominent ones that define the problem statement clearly are fishbone diagram, fault tree analysis, tree diagram, FMEA, 5-whys, change analysis, Pareto analysis, events and causal factors chart, flowcharts, and barrier analysis.
- Tools, like the bar chart, scatter diagram, and storytelling have a slightly limited scope of defining the problem.
Establishment of Causal Relationships
- A causal relationship can be established between two variables when one of them changes due to the other.
- This kind of relationship thus, establishes a connection between the variables and the events to follow.
- Surprisingly, many of the tools are not effective at defining causal relationships, that is to say, the causes can be listed accurately, but the subtle relationship between them may not be properly defined, and they instantly focus on the reasons.
- The tools that fall under this category are fishbone diagram, Pareto analysis, change analysis, barrier analysis (because the reason for the barrier failure is not clearly defined), 5-why chart and the tree diagrams.
- The ones that are effective at defining all causal relationships are the fault tree analysis, event and causal factors chart, histogram, and the RealityCharting tool.
Support for Identified Causes
- Support for identified causes means that the tool should provide a causal path to the root causes.
- This criterion is almost similar to the previous one, which indicates that the tools would practically be the same, with a few exceptions.
- Ironically, the 5-why chart, which does not define a causal relationship between its variables, very well takes you down to the root causes.
- Again though, this approach is traditional and works only in some applications.
- Another tool that provides an effective path to the root causes is the fault tree analysis chart.
- The others, like the fishbone diagram, tree diagram, Pareto analysis, barrier analysis, and change analysis are definitely not accurate in reaching the root causes.
Solutions to Avoid the Problem in Future
- Most of the tools do not have symbols for countermeasures either.
- The tools that provide solutions in the diagram itself include the fault tree analysis chart, tree diagram, and flowcharts.
Basic Tools and Techniques for Root Cause Analysis
Fishbone diagram (cause and effect)
- It is also known as the cause and effect diagram or Ishikawa diagram.
- It is named so because the way the causes and effects are enlisted resembles the internal structure of a fish.
- The effect is placed at the head, while the causes are indicated as branches.
- These are the major causes. Furthermore, each cause is analyzed and sub-causes are identified.
- The sub-causes may be further divided into secondary causes, and on it goes.
- The root cause can be traced to any of these secondary or sub causes.
- To know more about the fishbone diagram, click here.
- It is one of the oldest and best tools. It helps clearly outline the problems and their causes.
- It begins with the main problem on top.
- This diversifies into two branches, based on the causes.
- This spreads out into further causes, and keeps expanding.
- As the hierarchy increases, the root cause will be embroiled in one of the lower ranks.
- There are many forms of tree diagrams that are used according to the situation and application.
- It is also used in quality control and depth analysis.
- Though quite an efficient tool, it is not used aptly for every situation.
- It requires one to ask the question 'Why' at every step, until the base reason is found.
- For example, if the problem is something, like "You keep getting cold and cough throughout the year". Apart from seasonal change, the internal reason for this could be a weak immune system. There is the first why and its answer.
- Ask the question again. Why is the immune system weak? It is weak because of poor diet and nutrition. (The second why)
- Why is the diet poor? It is poor because you consume junk food and processed meals, not home-cooked food. (The third why).
- Why don't you have homemade meals? There is no time to prepare and eat hot and healthy meals. (The fourth why).
- Why is there no time for the same? This is due to a stressful lifestyle. (The fifth why).
- Now we have the root cause. Due to a stressful lifestyle, you have no time to cook good food, so you have junk food, due to which you do not get enough nutrition and this leads to low immunity, and therefore you are prone to getting cold and cough often.
- This was a very simple and random example, but it gets more complicated in business.
- Also, this does not explore all the causes. I mean, the whys could have more than one answer as well.
- Nevertheless, this is an easy technique used for troubleshooting and quality management, and though some problems might require a few more whys to arrive at a concrete base, generally, five iterations are sufficient.
Fault Tree Analysis
- As the name suggests, it is used to identify the faults in a system.
- It uses AND, NOR, OR, and NOT gates in its diagrammatic representation. This is done to represent relationships.
- It works very well as a causal factor identification tool.
- It uses data that are specific to the failure of the system.
- It is used extensively in the engineering systems, especially in the design and analysis phase, to arrive at the root cause.
- Alternatively called flowcharts, they are one of the most widely used and easy tools.
- They use a variety of symbols to analyze and structure the problem.
- Some of these symbols include the decision box, statement box, arrows, and connectors (like A and B, as shown above).
- The problem goes through a number of stages and iterations to arrive at a concrete step.
- A number of conditions are encountered and solved.
- As this progresses step-by-step, eventually, one arrives at the root cause.
Some Other Tools
Some other tools and methods for root cause analysis are summarized below.
- It is a kind of mathematical tool that displays variables on the X-Y graph as Cartesian co-ordinates.
- It mostly consists of a linear curve or line with dots scattered around it.
- These dots are the points that are plotted on the graph for suitable X and Y values.
- Cause and effect relationships are tested through this diagram.
- If the data sets are identical, the scattering will be minimum, and if they differ, the scattering will increase accordingly.
- This visual depiction helps reach the root cause through effective comparison of the data sets in question.
- Also known as Shewhart chart, it is used in business phases to help determine if the business is under control in terms of its statistics.
- It is a graph that is traced over a timeline.
- It has a central, upper, and lower line. The upper and lower lines indicate the higher and lower limits of data respectively.
- Using existing data, points are plotted and joined to determine the level of control.
- Once the chart is complete, analyze the data.
- Determine which point is outside control and which is not.
- Based on the results, you can trace the root of the problem.
- Also called impact analysis, it focuses on the things that have changed, and helps identify differences.
- It is usually a six-step process. It focuses on the event, to begin with.
- After that, it concentrates on the conditions that caused the events.
- The situations that gave rise to those conditions are outlined individually.
- The situations are compared and the differences are identified.
- Then, the differences are analyzed and the consequences are written down.
- Finally, the consequences are compared within a timeline and the root cause is found out within these results.
- The chart can be drawn using regular work flow diagram symbols.
- It is also called the 80-20 rule, and it gets its base from the Pareto principle.
- It states that just 20% of your causes result in 80% of your problems.
- It focuses on the most common causes, and helps direct the analysis.
- It involves the usage of a barrier to identify the cause of the problem.
- It means that a barrier is used to find the cause. If the hazard manages to overcome the barrier and harm the target, it has certain loopholes.
- Barrier analysis, thus helps find out why the barrier failed. This leads us to the root cause.
- In case of spam identification, the firewall protects the data and hence, we know the data is protected and the algorithm is right.
- If this does not take place, the firewall is mistaken, and thus, we can identify that it is the root cause of the problem.
Event and Causal Factors Analysis
- It is more of a technique than a tool.
- It helps enlist events and the causal factors and conditions.
- All these events and conditions are aligned in a timeline.
- Once the diagram is drawn, it is analyzed step by step, eventually figuring the real reason.
- Assessment is done by asking questions that represent change in the events and conditions.
Management Oversight and Risk Tree Analysis (MORT)
- As the name suggests, it is basically used in order to avoid oversight while identifying causal factors.
- Only the most specific factors are listed and the management conditions responsible for these factors are also listed.
- Unplanned events and the corresponding losses are the main focus here.
- The analysis concentrates on why the accidents occur and how the losses can be avoided.
- The MORT analysis requires you to act as the interpreter between the situation and the procedure itself.
- The questions asked concentrate on a particular incident at a time and clarify related facts.
- This is a slightly complicated tool than the others and is often used in health management systems.
- It is not exactly a part of the RCA process, yet it is used in plenty of situations.
- It is the most basic analysis tool, i.e., it identifies the problem statement, writes down the problem description, identifies reasons in the simplest way, finds the reason for faults, etc.
- It uses tree diagrams to arrive at the solution.
Failure Modes and Effect Analysis (FMEA)
- It can be called an extension of the fault tree analysis method.
- It is also mainly used in engineering and manufacturing systems.
- It identifies a component and lists down the related failures, which are represented as nodes.
- The result of every failure is carefully assessed and the criticality of the node is found out.
- It is a simple causal process.
- It enlists a problem, and expands the chart by listing down actions and the corresponding conditions.
- This continues until every cause and its reason are stated, until there are no more left.
- After the chart is complete, it is scrutinized to find out the basic reason.
- It is similar to the current reality tree (CRT), which focuses on identifying the core conflict and related symptoms, using the 'if-then-else' cycle.
- It was developed in the 1960s by Charles Kepner and Benjamin Tregoe.
- It is a structured problem-solving methodology that prioritizes and evaluates data.
- It basically concentrates on minimizing the negative results, and making a related choice.
- It follows a four-step process, namely, situation, problem, decision, and analysis.
- The situation in hand is clarified first, along with the related conditions.
- The problem statement is defined clearly.
- A decision analysis is done using the risk management method.
- The potential threat is identified and steps are undertaken to minimize the negative consequences.
RPR problem diagnosis
- Used extensively in the software industry, the RPR (rapid problem resolution) technique is a diagnosis method that deals with failures and incorrect performance issues.
- First, you need to discover the problem, collect related information, and define the problem statement.
- Then the investigation process begins. Here, you need to create a plan to capture the related causes. You might need to iterate the steps, as in all the other tools and finally, you will arrive at the root cause.
- Once the root cause is identified, the last step involves fixing it. This is done by determining the different solutions and translating the diagnostic chart for further evaluation.
- It is a typical bar chart. It uses the usual graph elements to indicate a particular result.
- It is used extensively in mathematics and statistics, and consequently in business and finance.
- Data is plotted in the bar graph, which is further analyzed to arrive at the base of a problem.
- It is a graph that shows the frequency of the occurrence of data.
- It can take the form of a bar chart, histogram, or a Pareto chart based on the data patterns.
- The focus of this plot is the continuity of the measured values.
- It can be plotted by using values of a continuous type, i.e., if the horizontal axis shows weight or height, the vertical axis displays the number of times that value was measured in the group.
- They are used to classify and organize data.
- The data are classified and organized into themes, and connections are established between them.
- These connections are the causal relationships that will further help you identify the root cause.
- The relationships are expressed in the form of incoming and outgoing arrows. Count them and draw up a bar chart in the decreasing order.
- Further analysis of the bar chart will help you identify the effects and the root causes.
Remember that the pros and cons of the tools must be carefully evaluated and then a decision must be taken regarding which tool is to be used. For clarity and convenience, many managers go in for a combination of tools. This is already evident from a few examples given above. This can be quite an effective strategy, considering that there are some similarities in the techniques. It will give a better picture regarding the basic reason for the event to have been caused. This in turn will help the management go through the next phase of solving the problem.