Quality Improvement and Excellence Models cheat sheet

Key terms

Improvement versus control

Control detects and corrects deviation from an established standard, returning the process to its existing capability. Improvement changes the capability itself, so the process performs at a level it never previously achieved.

Common and special cause variation

Deming's distinction, and the foundation of the whole subject. Common cause variation is inherent in the process and affects every outcome; special cause variation arises from something identifiable and unusual. Special causes are removed by investigation; common cause variation is reduced only by changing the process.

Continuous and breakthrough improvement

Continuous improvement makes small incremental gains constantly, from within, at low cost and low risk. Breakthrough improvement makes large step changes through redesign or investment, at higher cost and risk. Mature organisations run both, and the two are complements rather than alternatives.

Prevention over detection

Detecting a defect at inspection costs more than preventing it at source, and the cost of correction rises by an order of magnitude at each stage it escapes. This is the economic case for building quality in rather than inspecting it in.

Improvement requires data

Opinion identifies symptoms; data identifies causes. The habitual failure is to act on the most recently complained-about problem rather than the most consequential one.

Improvement must be held

An improvement that is not standardised decays as staff change and volumes vary. Standardisation is part of improvement and not an administrative afterthought.

Check sheet

A structured form for collecting data at the point it arises, designed so that recording is quick and classification is consistent. It is the first tool in most investigations because it converts impression into evidence.

Histogram

Displays the distribution of a variable, revealing centre, spread and shape. It shows immediately whether a process is centred on target and whether its spread fits the specification.

Pareto chart

A bar chart of causes ordered by frequency or cost, with a cumulative line. It operationalises the observation that a small number of causes account for most of the effect, and it is how improvement effort is prioritised.

Cause and effect diagram

Also called fishbone or Ishikawa diagram. Structures the search for causes under headings, commonly the six Ms: machine, method, material, manpower, measurement and environment. Its value is that it forces consideration of categories nobody has raised.

Scatter diagram

Plots two variables against each other to reveal whether they are related. Its standing warning is that correlation is not causation, and a plausible mechanism must be established before acting.

Control chart

Plots a process characteristic over time against statistically derived limits, distinguishing common from special cause variation. It is the tool that tells an operator whether to act.

Stratification or flow chart

Sources differ on the seventh tool. Stratification separates data by source, shift, machine or operator to reveal patterns concealed in the aggregate. The process flow chart documents the sequence of steps and is where most improvement work begins.