Seven Basic Quality Tools

The seven basic quality tools are the standard toolkit for shop-floor quality work. Kaoru Ishikawa, the Japanese engineer who championed them in the 1960s, argued that the vast majority of quality problems could be solved with these seven tools by ordinary workers, without statisticians or specialized training. Decades of practice in factories around the world has proven him right. The tools are simple enough to teach a small team in a day, and powerful enough to solve most of the quality problems a small shop will face in a year.

"Seven small tools. Most quality problems do not need more. Most teams do not use even three of them well."

How the seven basic quality tools work

The standard seven, with one variant on the seventh slot, are:

1. Cause-and-effect diagram, also called a fishbone diagram or Ishikawa diagram. Maps possible causes across categories like Man, Machine, Method, Material, Measurement, and Environment.
2. Check sheet. A simple paper form for collecting data as it occurs, usually with tally marks against predefined categories.
3. Control chart. A time-series plot with statistical control limits, used to distinguish common-cause from special-cause variation.
4. Histogram. A bar chart of the distribution of a dataset, used to see the shape and spread of a process.
5. Pareto chart. A sorted bar chart ranking causes from largest to smallest, with a cumulative percentage line.
6. Scatter diagram. A plot of paired observations to test whether two variables move together.
7. Stratification (in some lists) or flow chart (in others). Stratification breaks data into subgroups to surface hidden patterns. Flow chart maps the sequence of a process.

The tools are designed to work together. A typical quality investigation starts with a check sheet to capture data, runs a Pareto chart or histogram to find the biggest issue, uses a fishbone to map possible causes, tests the most likely cause with a scatter diagram, and installs a control chart to monitor the fix. Five tools, one investigation, no statistical software, no consultants.

The reason the seven endure is that they are right-sized for shop-floor problems. They are powerful enough to handle most variation and pattern questions, simple enough to teach in a morning, and cheap enough to deploy on every workstation that needs them.

Where the seven basic quality tools fit on the shop floor of a small manufacturer

Imagine a 30-person sheet metal fab shop with a creeping rework rate. Over the last six months it has climbed from 1.5 percent to about 4 percent of shipped value, and the team cannot identify a single cause. The owner is considering hiring a quality engineer. Before that, the shift lead runs a quality investigation using the seven tools.

A check sheet goes to the inspection bench for three weeks with eight defect categories. The first cut of the data plots into a pareto-chart showing that two defect categories, weld porosity and a specific dimensional drift, account for 68 percent of rework. The team runs a fishbone diagram session for each of the two top defects. For the dimensional drift, they suspect a tool wear correlation; a scatter diagram of tool age against measured dimension on 50 paired observations confirms the relationship. A countermeasure is installed (a tool-life standard with a visual signal), and a control chart goes up at the machine to confirm the variation stays inside limits over the following months.

Five of the seven tools used in one investigation. Rework drops to 1.1 percent within a quarter. No consultant, no software, no new hire. That is what the toolkit is for.

Common mistakes with the seven basic quality tools

• Using them as a checklist. Trying to use all seven on every problem produces theatre. Use the tool that fits the question.
• Skipping the data discipline. A Pareto chart with sloppy categories is misleading. A histogram with too few data points is noise. The tools depend on honest data.
• Treating the seven as the whole field. They cover most shop-floor problems but not all. For complex variation problems involving many interacting variables, the seven may need to be supplemented with designed experiments and regression.
• Knowing three tools well. Most teams default to the two or three tools they have used before. Make a habit of using the one that actually fits the question, even if it requires a refresher.
• Stopping after the diagnosis. The tools surface causes and patterns. Without a countermeasure and follow-up, the investigation was theatre.

Seven basic quality tools and related Lean tools

The seven tools are deeply connected. A check sheet feeds the data that a pareto-chart ranks. A fishbone diagram maps possible causes when a target is picked. A control chart and a histogram cover the time and shape views of a process, respectively, and complement each other in most ongoing monitoring work.