• SPC is a collection of statistical techniques useful in

achieving process stability and improving capability through

What is SPC? “Statistical Process Control(SPC)” http://www.linkedin.com/in/balaji-l-r

the reduction of variation. Cheat Sheet Attribute Data Control Chart- Formulas
• SPC was born in the 1920’s when Walter Shewhart developed the first Control Chart is a graphical tool used to Control Chart
Control Charts which is used to determine process performance with determine process performance with time.These
time. charts help determine whether the process is stable or not.
• Data collection are mostly attribute and variable Variable Control Chart explains the process data in terms of its
• Data are used for controlling and monitoring the process process variation, piece to piece variation, and its process average
• It’s a tool to improve the quality and reduce variation in the process Types:X bar and Range Chart, X bar and S Chart, I (Individual) and
(DMAIC approach) MR (Moving Range) Chart.
• IATF Requirement – Clause 9.1.1.2 and AIAG Manual – 2nd Edition Attribute Control Chart, they have discrete values and can be
counted for recording and analysis.
Data Types
Types: p, np, c, u Chart
Variable data comes from measurements on a continuous scale,
such as: temperature, time, distance, weight, diameter etc.,
Attribute data is based on upon discrete distinctions such as good/bad,
go/no-go etc.,

“SPC is based on the analysis of data, so the first step is to decide

what data to collect”

Common (Chance) Cause Variations that are consistently acting on Variation Variable Data Control Chart- Formulas
the process. Produce a stable and repeatable distribution over time (in a
state of statistical control)
Special (Assignable) Cause Variations that affect only some part of the
process output. Often intermittent and unpredictable.

“In most cases, some level of variance to a process is inevitable”

Control Limits (Voice of Process) are calculated as three Control Limits

standard deviations away from the mean on either side of the • Pre-Control Chart is a tool used to quickly Pre-Control Chart
mean. Also known as process limits, these are derived basis the process data check the status of a process.
and represents the area within which natural variation can be expected • It is typically used by shop floor workers / machinists to
(common cause variation). Any observation/data point outside the control determine if some action is required on the process or not.
limits is a reason for investigation and will always have • If the product is out of specification, the process is stopped
an assignable cause (special cause) and adjusted. If the product is near the target, the process
Upper Control Limit is 3 standard deviations above mean continues, if the process is nearly out of specification, the
Lower Control Limit is 3 standard deviations below mean process is adjusted toward the center.
p = Fraction of defective units; np = Number of defective
Specification Limits (Voice of Customer) are the Specification Limits Defect and Defective
units; c = Number of defect’s; u = Number of defects per
boundaries within which the products or services should A “defective” is an entire unit that fails to meet specifications. units; np = Subgroup size; k = Number of subgroups;
operate in order to be acceptable by the customer. A “defect” is an undesirable condition within a unit. X = Observation value; R = Range of subgroup observations;
Any product / service which falls outside the specification limits is considered A defective unit can have multiple defects associated with it. A2, D3, D4, d2, and E2 are all Constants ~Balaji L R
as a defect.
 Different Zones of a Control Chart
8 Rules for Control Chart “Statistical Process Control(SPC)” http://www.linkedin.com/in/balaji-l-r

Zone A is the area between the 2nd standard deviation and 3rd Cheat Sheet Cp and Cpk
standard deviation
Process Stability is a process where the Process Stability and Capability
Zone B is the area between the 1st standard deviation and 2nd
standard deviation observed variation is largely within the control limits or in other words
Zone C is the area between the centerline (process average) and almost all of the observed variation is the Common Cause variation. There
the 1st standard deviation are minimal assignable causes of variation in a stable process. A stable
process will rarely have outliers or unusual output values

Process Capability is a process where the observed variation is largely

within the specification limits or in other words has minimal rejections. Such Any specification limit has an upper and lower Why 3σ in Cpk
a process almost always produces as per customer requirements. Capability limit. Because the 6 sigma or 6 standard deviations account for
analysis uses measures like Cp, Cpk, Pp, Ppk to determine the process nearly all eventualities on a process (assuming normal distribution)
capability. shouldn’t be surprised to see the “/ 3” because we are looking at
“Process capability assessment should only be performed after first only one side of the distribution.
Rule 1 – Any single data point outside of either the upper or demonstrating process stability”
Cp & Cpk Interpretation
lower control limits (>3σ) – outside Zone A.
Capability Analysis is the process of Capability Analysis and Index
Rule 2 – Two out of three consecutive data points in Zone A
assessing if a process and/or system is capable to meet the specified
(same side) – greater than 2σ, but less than 3σ.
customer requirements or in other words how capable is the process and/or
Rule 3 – Four out of Five consecutive data points in Zone B
system to produce defect free items or services
(same side) – greater than 1σ, but less than 2σ.
Capability Index is a metric used to evaluate process capability and show how
Rule 4 – Eight or more consecutive points on either side of the
capable process is in terms of meeting customer requirements.
Centerline –a run
“Cp,Pp and Ppk,Cpk are Capability indices”
Rule 5 – Six points in a row, all increasing or decreasing – a trend
Rule 6 – 14 points in a row, alternating up and down – systematic Cp and Pp are a monitoring indices for Process Capability & Performance • Atleast one specification limit must be stated to compute Cpk
variation the spread of your process compared to • Smaller Cpk – More defects and Larger Cpk – Better Process
Rule 7 – 15 points in a row within Zone C (<1σ) – to little variation the specification spread. • Generally Cpk values is 1.33 or 1.67 or as per customer
Rule 8 – 8 points in a row outside of Zone C (>1σ) on either side Cpk and Ppk are ways to measure the capability of a process because they specific requirements
are monitoring indices for process centering. • Cp always greater than Cpk
Rational Subgrouping is defined as a Rational Sub-grouping
Difference between Cp and Pp is how the Standard Deviation is • Cpk can only have positive values
collection of units that are all produced under the same conditions determined. Pp we use sampling(s) and have to calculate an estimated • If mean of the process is outside the specification limits
in order to measure the variation between the subgroups instead of standard deviation of the sample. In Cp we are assuming a stable process and means Cpk is negative
between individual data points. Since the items in a subgroup are will likely have enough data to calculate a population standard deviation(σ). • Cpk = 0,means actual process average matches or fall outside
relatively homogeneous, dividing the data set into rational one of the specification Limits
subgroups helps in analyzing the difference between subgroups and “Use Pp & Ppk (Long Term Performance) for New process & • Cpk = 1 - Process is Ok ; Cpk=2 – Good Process ;
the underlying assignable reasons (special causes) Cp & Cpk (Short Term Performance) for Process under statistical control” Cpk=3 – Excellent Process ~Balaji L R