14.4 The Seven Basic and Seven New Quality Tools

ECM Section IX.A.1 names two quality toolkits, and the exam tests whether you can put a tool in the right kit and pick the right one for a scenario. The basic seven tools of quality (B7) work on numerical data about what already happened. The seven new tools of quality (N7) organize verbal, qualitative information so a team can plan its way through a complex or unfamiliar problem. Hold that split and half of this question type answers itself.

The Basic Seven Tools (B7)

Check sheet — a pre-formatted tally form completed as events occur, so collection needs no memory and no re-keying. It collects; it never analyses. Planner use: for two weeks, every line stoppage gets a tick under a cause column and a shift row, producing data instead of anecdotes.

Histogram — a bar chart of the frequency distribution of one measured variable, showing centre, spread, and shape (skewed, bimodal, truncated) while deliberately discarding time order. Planner use: plot 200 actual purchase lead times against the lead time in the item master; two humps usually mean two suppliers, two routings, or two order sizes hiding inside one part number.

Pareto chart — bars of counts sorted largest to smallest with a cumulative percentage line, expressing the vital few versus the trivial many. It decides where the improvement budget goes first.

Cause-and-effect diagram — the Ishikawa or fishbone diagram, branching candidate causes off an effect, usually by the 6Ms (methods, machines, materials, manpower, measurement, mother nature). Section 15.3 develops it as a root-cause tool with the full 6M table; here, fix it as a member of the B7 and as the tool you reach for after a Pareto names the bar to attack.

Scatter diagram — plots paired values of two variables to test whether they move together. Planner use: order quantity against supplier days-late; a rising cloud suggests large orders arrive late, but a scatter shows correlation, never causation.

Control chart — plots a process statistic in time order against a centre line and control limits, separating common-cause noise from special-cause signals. Section 14.5 takes it apart chart by chart.

Flowchart (process flow diagram) — maps the real sequence of steps, decisions, handoffs, and rework loops. It comes first in practice, because you cannot improve a process you cannot draw. Exam note: some references put stratification or a run chart in the seventh slot instead of the flowchart, so the set may be worded either way; the other six are stable across sources.

Worked Pareto: a chronic shortage problem

One quarter of line-side stockout incidents, by cause:

The arithmetic: 142 + 96 + 48 + 22 + 14 + 8 = 330. Then 142 ÷ 330 = 43.0%; 96 ÷ 330 = 29.1%, so the cumulative line reaches 43.0 + 29.1 = 72.1%; 48 ÷ 330 = 14.5%, cumulative 86.7%. The 80% threshold is crossed by the third bar, so the vital few are supplier lateness, record error, and forecast miss — three of six causes carrying 286 of 330 incidents. Attacking transit damage first spends the budget on 2.4% of the problem.

A refinement the exam likes: Pareto by count and Pareto by cost rank differently. Twenty-two unphased engineering changes stranding $18,000 of obsolete stock each is $396,000, while 96 record errors at $400 to correct is $38,400 — the fourth bar on count outranks the second bar on cost by more than tenfold. Pick the axis that matches the objective before you plot.

The Seven New Tools (N7)

The N7 answer a different question: not "what does the data say?" but "how do we organize what people know and turn it into a plan?" Reach for them when the problem is unfamiliar, the causes are contested, or the deliverable is a project rather than a measurement.

• Affinity diagram — sorts a large pile of unstructured verbal input into natural groupings that the team names afterwards. The first move after a broad brainstorm.
• Interrelationship digraph — draws arrows between issues to show which drives which. Count them: most outgoing arrows marks the driver worth attacking, most incoming marks the symptom worth measuring.
• Tree diagram — decomposes a goal into means, then sub-means, until it reaches assignable tasks. Its value is completeness: it exposes the branch nobody staffed.
• Matrix diagram — puts two or more lists on the axes of a grid and marks the strength of each relationship. The house of quality in quality function deployment (QFD) is a matrix diagram.
• Matrix data analysis (prioritization matrix) — the numerical member of the set: scores options against weighted criteria and ranks them.
• Arrow diagram (activity network diagram) — sequences implementation tasks with dependencies and durations so the critical path is visible.
• Process decision program chart (PDPC) — walks down a plan asking "what could go wrong at this step?" and attaches a countermeasure to every credible failure.

Which Tool, In Which Order

Handed a chronic shortage on a purchased component, run the B7 in sequence, then hand off to the N7:

1. Flowchart the real requisition-to-receipt path, including the loops nobody documented.
2. Check sheet every shortage, tagging cause, part, supplier, and shift.
3. Pareto the tally to isolate the vital few.
4. Cause-and-effect diagram the top bar, then drill each candidate with 5 Whys (section 15.3).
5. Scatter diagram to test the leading hypothesis — order quantity versus days late.
6. Histogram the supplier's actual lead times; a planning lead time set at the mean makes half of all orders late by construction.
7. Control chart the metric afterwards to prove the gain held instead of drifting back.

Then the N7 build the countermeasure plan: affinity to cluster ideas, tree to break the countermeasure into tasks, arrow diagram to schedule them, PDPC to pre-plan the rollout's failure modes.

Exam traps

• A histogram has no time axis; if the question cares when the change happened, it wants a run or control chart.
• A Pareto chart is not a plain bar chart — sorted descending plus a cumulative line is what makes it one.
• A check sheet collects; it never analyses.
• A scatter diagram shows correlation, not cause.
• The N7 need no numeric data at all — that is their purpose, and why they live in planning rather than inspection.