Root cause analysis/Pareto analysis
Root cause analysis (RCA) is a class of problem-solving methods aimed at identifying the root causes of problems or events. It is based on the Ishikawa diagram (also fishbone diagram, or cause and effect diagram) named after its founder Kaoru Ishikawa (Figure 48.1a). The Ishikawa diagram shows the causes of a certain event. It was first used in the 1960s and is considered one of the seven basic tools of quality management, along with the histogram, Pareto chart, check sheet, control chart, flowchart and scatter diagram (Figure 48.1b). This principle is used in a root cause analysis and tries to explain the variations in a particular process. The analysis is generally used in both financial analysis and the analysis of operations, such as in business process redesign (BPR) projects (see Chapter 43).
The RCA is used to explain the variation in any process (or outcome of a process). A certain amount of variability is normal and does not necessarily cause significant disturbance. However, unwanted variation can cause serious losses or damage, delays and reduced productivity, especially if it occurs in critical processes. The first essential step is to find the causes of variation and to quantify the effect. The main causes, which are generally easy to solve, should be taken care of first. The technique is particularly valuable for the analysis of critical processes that show undesirable variance.
Figure 48.1 (a) Ishikawa diagram, or cause and effect diagram. (b) Pareto diagram
Root cause analysis usually starts with the formation of a project team, including managers, suppliers, customers and employees. Next, the team defines the problem and decides which variation causes the most critical disturbance in the system under study. Then the team maps out the process and identifies the issues that can cause variance in the data/evidence-gathering phase. Following this, issues that contributed to the problem are identified and their root causes found. However, the root causes might not be immediately evident, in which case brainstorming techniques are required. Subsequently, the root causes identified (usually large in number) are illustrated on a whiteboard in order to discuss and sharpen the findings. Recommendations for solutions now have to be developed and actually implemented.
The root causes can be organised by categorising them and by distinguishing between main root causes and smaller effects. This provides the input needed to draw a ‘cause and effect’ diagram. The diagram provides an overview of the possible causes of variation. It is essential to study the possible root cause in the diagram in detail, to see the extent of the cause of variation. The Pareto diagram is often used to present the findings. Analysing root causes generally shows that 80 per cent of the variation is caused by 20 per cent of the causes.
Root cause analysis is not a single, sharply defined methodology; there are many different tools, processes and philosophies regarding RCA. To maximise the effect of the use of RCA, it is advisable to start with the most critical processes and/or the most disturbing variances. This ensures that success will propagate the broader use of the model. However, try to avoid finding causes of variation that have only a small effect on the lead time, productivity or costs.
Blanchard, K.H., Schewe, C., Nelson, R. and Hiam, A. (1996) Exploring the World of Business, New York: WH Freeman.