Root cause analysis
The aim of root cause analysis is to identify the factor(s) that resulted in one or more past events. Often the event is unwanted and by identifying the root cause it is possible to prevent the event from happening again.
Depending on the problem, applicable tools can be Fishbone / Cause and effect diagrams, 5 Why's, Graphical analysis (Pareto, Histogram, Box plot, Run Chart and Scatter plot including Regression and Correlation factor), hypothesis testing, Component of variation or design of experiment.
Please note that searching for a root cause is only relevant when the unwanted event is due to special cause variation. If the normal variation is causing unwanted result(s), you cannot look for a root cause. Overreacting to normal variation by adjusting, will likely worsen the situation. Statistical process control (SPC) is used to distinguish between normal and special variations to respond correctly to measurement results.