Scribd
Upload
12 views

Lecture 21

The document discusses quality management and variability in production processes. It defines key terms like quality characteristics, sources of variation, and types of variation. It then explains the six sigma approach to quality improvement, including concepts like process capability, process centering, and controlling variation to reduce defects per million. The goal of six sigma is to improve processes so they are capable and centered to minimize non-conforming output.

Uploaded by

Sejal Gupta
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
12 views

Lecture 21

The document discusses quality management and variability in production processes. It defines key terms like quality characteristics, sources of variation, and types of variation. It then explains the six sigma approach to quality improvement, including concepts like process capability, process centering, and controlling variation to reduce defects per million. The goal of six sigma is to improve processes so they are capable and centered to minimize non-conforming output.

Uploaded by

Sejal Gupta
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 16

Operations Management

Dr. Jinil Persis


Assistant Professor
IIM Kozhikode

Lectures on “Quality Management”


Variability
High

Low
Variability
• Undesirable
• Causes product variation (w.r.t. quality characteristics)
• Reduces capacity utilization (designed vs. actual capacity)
• May lead to defects/product failure; scrap or rework
• Wastage of time and resources
• Production loss
• Variability is inevitable
• Variability should be maintained within acceptable range
Quality characteristics
• Quality characteristic is a product characteristic demanding
conformance for acceptance
• Within specification limit – accepted (conforming)
• Outside specification limit – rejected (defective/non-conforming)
• Measurable characteristics of products
• Eg. Product dimensions, weight, mechanical properties like tensile strength
• Qualitative characteristics of products
• Eg. Color, appearance (glossy, etc.), fragrance

1
𝑣𝑎𝑟𝑖𝑎𝑡𝑖𝑜𝑛 ∝
𝑄𝑢𝑎𝑙𝑖𝑡𝑦
• Consider a chain of cake shop. They provide 1kg, 1.5 kg, 2 kg cakes.

• Quality characteristics: Weight of the cake, Diameter and other dimensions, color, taste

• Identify the causes of variation in every quality characteristic.

• Many cases, variability can be eliminated by standardization.


• Diameter variability can be eliminated by having standard pan size. Say they have a 8 inch
pan for making 1.5 kg.
• Weight/color variability can be eliminated by having proper measuring scales (digital scales)
• Taste variability can be eliminated by standardizing process and material
• Process standardization can be done by controlling the process variables, like,
temperature, time, mixing, etc.

• Some causes of variability are unknown/unexplainable/random


Sources of Variation in Production Processes

Measurement
Operators Methods
Materials Instruments

INPUTS PROCESS OUTPUTS

Tools Human
Machines Environment Inspection
Performance

6
Types of Variation
• Uncontrollable variation (common/random causes) are a
natural part of a process
• Variation inherent in the process by design
• Variation inherent in the process unknown
• Special (assignable) causes of variation can be recognized
and controlled
Failure to understand these differences increases variation in a
system
Problem solving approach - DMAIC

Solve root causes of


quality problems

Establish best
practices

Ensure that the


solutions are
permanent

Replicate solutions
in other relevant
business operations

What
What is
is six
six sigma?
sigma? Process
Process capability?
capability?
Quality improvement
Quality improvement is linked with
production rate, capacity of the process,
productivity.

10 15 10 Mins/unit 32 units/shift

4 mins rework

10 11 10 Mins/unit 43 units/shift

2 min waste

10 9 10 Mins/unit 48 units/shift

Process improvement is continuous … standardized process is the outcome


Six sigma
• Consider a variable quality characteristic, say, weight of a part.
• It follows normal distribution 𝑋~𝑁 𝜇, 𝜎
• Under normal variations, the sample means should fall within ±3𝜎
• If not, the process is not under control
• The control limits are driven by the natural variability of the process (measured by the
process standard deviations), that is, by the natural tolerance limits of the process.
• It is customary to define the upper and lower natural tolerance limits, say UNTL and LNTL, as
3𝜎 above and below the process mean.
• If all the units made has quality characteristic within these control limits, then the process is
said to be under control and the variation existing is due to natural causes.
• The specification of X = 𝑇 ± 𝜏 where T – target mean and 𝜏 – tolerance
• If any unit falling outside the specification limits is categorized as defective.
• The process under control must be capable (producing conforming units)
• If not, the process has to be redesigned or thoroughly checked (inspected)
• Specification limits and control limits are statistically different.
Six sigma (Contd.)
• During design, the process is centred (Process mean = Target mean).
During production, the process tend to get off centred.
• Chance of defective production is induced with off centred process
• Have tighter tolerance and good process design such that the natural
tolerance limits are within the specification limits.

𝑈𝑆𝐿−𝐿𝑆𝐿
Potential capability, 𝐶𝑝 =
6𝜎
> 1, 𝑝𝑟𝑜𝑐𝑒𝑠𝑠 𝑐𝑎𝑝𝑎𝑏𝑙𝑒
𝐶𝑝 ቐ= 1, 𝑝𝑟𝑜𝑐𝑒𝑠𝑠 𝑗𝑢𝑠𝑡 𝑐𝑎𝑝𝑎𝑏𝑙𝑒
< 1, 𝑝𝑟𝑜𝑐𝑒𝑠𝑠 𝑛𝑜𝑡 𝑐𝑎𝑝𝑎𝑏𝑙𝑒
A – Capable &
Process Mean
centered

B – Process mean
centered & Not
capable

C – Process mean off


centered & Not
capable

D – Process mean off


centered & capable

E – Process mean off


centered & Not
capable

Ref: Managing quality with process control, HBR, 2013


Six sigma (Contd.)
• Sometimes, when the natural tolerance limit is
closer to the specification limits, we
intentionally keep the process off centred to
avoid scrap.
• If process is off centered, Actual capability =
𝐶𝑝𝑘 = min(𝐶𝑝𝑘1 , 𝐶𝑝𝑘2 )
𝑈𝑆𝐿−𝑥0 𝑥0 −𝐿𝑆𝐿
Where 𝐶𝑝𝑘1 = ; 𝐶𝑝𝑘2 =
3𝜎 3𝜎

Scrap Rework

LSL Target USL


Six sigma (Contd.)

• When the process is just capable, small variation in mean, leads to


defectives
• Hence, process capability should be at least 1.33 (1.66 for new process)

What is six sigma process?


Six sigma (Contd.) Consider this scenario

LSL USL USL-LSL Process mean Process SD k


10 12 2 11 0.3333333 3 Process mean=Target mean, Target mean+ 𝑘𝜎 =12. At k=3, 𝜎 = 0.333
10 12 2 11 0.25 4
10 12 2 11 0.2 5
𝑠𝑝𝑟𝑒𝑎𝑑
𝑘𝜎 =
10 12 2 11 0.1666667 6 2

As k level increases, process standard deviation (variation) reduces, the process improves, chance of defectives reduces
LSL USL USL-LSL Process mean Process SD k Prob. conforming
10 12 2 11 0.3333333 3 0.99730020394 =NORM.DIST(USL,Process mean,Process SD,TRUE)-
10 12 2 11 0.25 4 0.99993665752 NORM.DIST(LSL,Process mean,Process SD,TRUE)
10 12 2 11 0.2 5 0.99999942670
10 12 2 11 0.1666667 6 0.99999999803
As k level increases, process becomes better capable

LSL USL USL-LSL Process mean Process SD k Prob. conforming Cp


10 12 2 11 0.3333333 3 0.99730020394 1.000 Just capable
10 12 2 11 0.25 4 0.99993665752 1.333 Refer to excel
10 12 2 11 0.2 5 0.99999942670 1.667 sheet uploaded
10 12 2 11 0.1666667 6 0.99999999803 2.000 in moodle
Six sigma (Contd.)
When the process becomes off centred, say, with 1.5𝜎 shift

LSL USL USL-LSL Process mean Process SD k Prob. conforming Cp


10 12 2 11.5 0.3333333 3 0.93318940106 1.000
10 12 2 11.375 0.25 4 0.99379031568 1.333
10 12 2 11.3 0.2 5 0.99976737088 1.667
10 12 2 11.25 0.1666667 6 0.99999660233 2.000

Before shift After shift (±𝟏. 𝟓𝝈)

k Prob. Conforming ppm defectives k Prob. conforming Cp ppm defectives


3 0.99730020394 2699.796 3 0.93318940106 1.000 66810.6
4 0.99993665752 63.34248 4 0.99379031568 1.333 6209.684
5 0.99999942670 0.573303 5 0.99976737088 1.667 232.6291
6 0.99999999803 0.001973 6 0.99999660233 2.000 3.397673
Refer to excel
sheet uploaded
in moodle

You might also like