To perform Pareto analysis, you should first define the quality problem you want to solve and the effect you want to measure, such as defect rate, cost, or customer satisfaction. Next, identify the possible causes of the problem and group them into categories. You can use brainstorming, fishbone diagrams, or other techniques to generate ideas. Furthermore, collect data on the frequency or magnitude of each cause over a period of time. You can use surveys, observations, checklists, or records to gather information. After that, calculate the percentage of the total effect for each cause and rank them from highest to lowest. Then draw a bar chart with the causes on the x-axis and the percentage of the total effect on the y-axis. The bars should be arranged in descending order from left to right. Additionally, draw a line graph that shows the cumulative percentage of the total effect from left to right. The line should start at 0% on the left and end at 100% on the right. Finally, analyze the chart and identify the vital few causes that contribute to most of the quality problem. You can use a vertical line or a threshold value to separate them from the trivial many. Typically, these vital few causes are those that lie on the left of the 80% cumulative line or above a certain percentage of the total effect.