Quality control - Concrete Mix Design, Quality Control and Specification - page 231

Quality control
The purpose of quality control (QC) is to ensure the continuous production
of an item of the required quality at minimum cost. Many countries still
attempt to do this for concrete by testing specimens at an age of 28 days
and provisionally rejecting concrete that does not comply. It is rarely feasi-
ble to actually discard concrete that has been in place for 28 days so further
testing by drilling cores or by ultrasonic testing ensues. Also the concrete in
trucks not sampled for testing remains under suspicion.
What is needed, after setting up appropriate production facilities and
selecting suitable materials and mix designs, is to detect and rectify any
departure from the intended quality at the earliest possible stage.
Although the most significant requirement of concrete may not be
strength, strength is the best means of detecting change from an initially
satisfactory mix. However variations in strength itself can themselves often
be predicted from earlier data including slump, temperature, and especially
test specimen density (which should always be determined on arrival at the
laboratory rather than at test, preferably within 24 hours).
MMCQC (multigrade, multivariable, cusum QC) remains the best way
of detecting change in the concrete being produced and it remains more
effective to aim QC at the detection and cause of change rather than at
checking conformance to a specified limit. Putting it simply, if a change is
detected in the quality of the concrete being produced and supplied and the
cause of that change can be established, then the change is genuine with-
out waiting for statistical confirmation. This is the purpose of multigrade
cusum graphs. For example, if strength shows a reduction from its previ-
ous average value and density also shows a reduction, while either slump
or temperature (of the concrete, at the time of the slump test) shows an
increase, then it is clear that the strength reduction is due to extra water. In
fact it may not even be necessary to await the early strength result as it can
clearly be predicted.
Of course typical concrete will no longer be composed solely of aggre-
gates, ordinary Portland concrete (OPC), and water, so that many other
causes of strength reduction are possible. Any variable (such as batch