Quality 101 - SPC For a Healthy Manufacturing Process

Statistical process control can help identifylight, less than 4 ounces, the electrical properties
problems, suggest corrective actions and reduceof the substrate could adversely affect
overall costs while preventing potentially seriouscomponent performance, leading to failures in the
issues.field-a very expensive, time-consuming problem. If
To assess pulmonary health, physicians routinelythe substrate is too heavy, the material
order electrocardiograms (EKG) for their patients."giveaway" will cost the company more money
Electrodes are attached to the patient's chest andthan necessary. The financial risk of a substrate
signals from the heart are "listened to" by abeing too light is greater than that of it being too
machine. A paper printout plots heart rhythmheavy, but both pose financial risks to the
over time, allowing doctors to identify peaks,manufacturer.
valleys and patterns in the heart's electricalEvery 15 minutes, a single substrate is weighed
impulses. Using the printout, physicians can assessby operators.
heart health and uncover previously unknownControl charts typically comprise two different
problems. After issues are uncovered, doctorscharts. The top chart shows the individual weights
can prescribe medications, suggest correctivethrough time. The bottom chart is called a moving
actions, and prevent or eliminate potentiallyrange chart. This chart plots the raw deviations
serious health issues.between any two weights, thereby revealing
Now imagine applying an EKG to a company'sweight variation.
most critical manufacturing processes. What ifThe initial control chart made during study one
electrodes could be attached to an assembly line,exposes useful information about weights:
or to the fill heads of a bottling process? Doing so1. The moving range chart shows much less
would allow manufacturers to assess machinevariation in weights between subgroups 26 and
health, know when it is "sick" and prevent58.
potentially expensive problems. This technology is2. Weights naturally cycle from high to low several
available for manufacturers today, and it is calledtimes during the day.
statistical process control (SPC).3. There is a slow, gradual trend in weights
Statistical Process Controlbetween subgroups 25 and 60.
SPC is a data-driven methodology for analysis and4. The overall average (mean) weight is 4.31
improvement that quality professionals have beenounces.
using since the early 1900s. This scientific method5. Because the expected weight is 4 ounces, on
of analysis involves graphing importantaverage, nearly one-third of an ounce per
characteristics to determine trends that indicatesubstrate is "given away" in the study.
potential quality issues. In the past 20 years,The control chart patterns-cycles, trends, for
manufacturers have implemented automated SPCexample-helped the implementation team better
systems so they can quickly and easily controlunderstand start up, warm up and other issues
product and process quality on the shop floor.affecting the manufacturing process. After
SPC's prime objective is to provide informationspeaking with operators, engineers and hardware
about manufacturing processes in real time.vendors, the team made machinery modifications
Quality features and data values are plotted inand integrated procedural changes aimed at
time series on a chart. Like an EKG's printout, theminimizing weight fluctuation and variation.
result is a simple, visual graph that provides crucialAfter making the changes, more data was
information to an operator. With a bettergathered during a second study. The
understanding of their production processes,post-improvement results reveal this information:
operators are able to make fact-based decisions1. No clear patterns or trends are present.
to control their machines. Ultimately, this powerful,2. No weights fall outside control limits.
real-time information allows better control of3. The lack of patterns, trends and points outside
processes, prevents off-quality product fromcontrol limits indicates that no statistically
being produced and reduces overall costs.significant changes in weights have occurred in the
Control Chartssecond study.
At the heart of SPC technology is a graph called4. The overall mean has been reduced to 4.16, a
a control chart. Control charts are used bysavings of 0.15 ounce per substrate.
operators on the shop floor to help identify when5. The variability in weights has been greatly
corrective actions are required. Control limits, orreduced compared to the first study.
decision limits, clearly indicate when correctiveThe second control chart made during study two
action is needed and, conversely, when it is not.indicates dramatically improved process control. In
Control limits should not be confused withaddition, because thousands of substrates are
specification limits. Specification limits define amade each week, overall manufacturing and
window of acceptability based on customer needsmaterial costs are significantly reduced.
or engineering requirements and are an indicatorControlling manufacturing processes is tough
of whether a product is good or bad.work, particularly when data analysis tools are
Case Exampleunavailable. However, SPC was specifically
A high-tech electronics manufacturer makesdesigned to help companies control processes
substrates for use in military applications. A newwith simple, graphical methods of analysis. Like an
production line has just been installed forEKG for manufacturing, a robust SPC software
manufacturing 4-ounce substrates. Because of thecan help identify problems, suggest corrective
extremely high cost of the materials used in theactions and reduce overall costs while preventing
substrate, the company wishes to carefullypotentially serious issues.
control substrate weight. If the product is too