Friday, September 11, 2009

Capability Analysis in JMP

I received a note asking how to do a capability analysis in JMP 8. Actually nothing much have changed in how to do it. There are additional features which are really helpful but the basic steps on computing Cp, Cpk, Pp, and Ppk are basically shared by versions JMP 5, JMP 6, JMP 7, and JMP 8. Here are the detailed steps. I will be using JMP 8's sample data 'Semiconductor Capability.jmp'.

Capability Analysis in JMP

Method1
Step1: Data Requirements
Data should be in one column and data type set as continuous.
Step2: Go to Univariate Platform
Create a histogram of your measurement data as shown below.
Step3: Click on the second hot button
It is the red triangle as shown below. On the drop down menu, select 'Capability Analysis'
Step4: Specification Limits
On the resulting dialog box, fill up the necessary information regarding the specification limits and target values. In this case they are 104.41295,118.15322, and 131.89349. For Pp and Ppk, check on the tick box 'Long Term Sigma'. For Cp and Cpk, select either on the tick box 'Moving Range...' and specify the range span, or on the tick box 'Short Term Sigma...'. Note that whichever you choose JMP would use the terms Cp and Cpk in its output. It would specify in the title box of the output though the Sigma your analysis is referring to.
Step5: Capability Analysis Output
After pressing 'OK', JMP will give you this output.

Method2:
Step1: Data Requirements
Same as above
Step2: Set up the column that contains your data
Right click on the column header to see the following menu. Select 'Column Info'.
Step3: Column Properties
On the drop down menu 'Column Properties', select 'Spec Limits'.
Step4:Specification Limits
Fill up the necessary information regarding the specification limits and target values. In this case they are 104.41295,118.15322, and 131.89349, and then click 'OK'. Notice on the left side of your data table, and asterisk (*) sign is placed beside the column name where you just set up the specification limits.
Step5: Go to Univariate Platform
Create a histogram of your measurement data as shown below. Capability Analysis using Long Term Sigma is included by default.

Additional Features:
One of the reasons why I favor JMP over Minitab is its animated graphics. In JMP 8, Capability Analysis includes a feature called 'Capability Animation. To use this, click on the hot button as shown below.
The output window is like the one shown below.You can change the sigma to see its impact on the capability measures. For a JMP script like this where you can change the mean as well, refer to this entry or this entry.

Thursday, September 10, 2009

Analysis of Means (ANOM) in JMP

This is in response to the query sent to me regarding on how to do an Analysis of Means in JMP. For those who are not familiar with this method, you can refer to this SAS link regarding the definition and history of the Analysis of Means, and to this presentation for the technical details. With regards to how to do Analysis of Means in Minitab, refer to this video.

How to do Analysis of Means (ANOM) in JMP

From the 2005 Winter issue of JMPer cable (issue 16) it says that "ANOM is not a standard procedure in JMP. However, scripts generate ANOM decision charts for comparing means with either equal or unequal sample sizes.You can download these scripts at http://www.jmp.com/news/jmpercable. The script for equal sample sizes is named ANOMB.jsl and the script for unequal sample sizes is named ANOMUB.jsl."

The JMP JSL Scripts

Here are the JMP scripts that are downloadable from SAS JMP's website.
JMP's Analysis of Means for Equal Sample Sizes
Clear Globals();
R = Column Dialog(
 Resp = Col List( "Y, Response",
  Data Type( Numeric ),
  Min Col( 1 ),
  Max Col( 1 )
 ),
 Treat = Col List( "X, Treatment",
  Min Col( 1 ),
  Max Col( 1 )
 ),
 V List(
  "Alpha",
  SL = Radio Buttons( "0.1", "0.05", "0.01", "0.001" )
 )
);

If( r["Button"] == -1, Throw( "User cancelled" ) );

TN = Column( R["Treat"] );
RN = Column( R["Resp"] );
SigLevel = R["SL"];

alpha = {0.1, 0.05, 0.01, 0.001};

Summarize(
 a = By( TN ),
 c = Count,
 mean Response = Mean( RN ),
 stdev Response = Std Dev( RN )
);
Ymin = Min( mean Response );
Ymax = Max( mean Response );

k = N Items( a );
K1 = Log( k );
K2 = Log( k-2 );
a1 = a[1];

tc = Sum(c);
MSEnum=summation(i=1, k, (stdev Response[i]^2));
nu = tc - k;
MSE = MSEnum / k;

Insert Into( a, "Total" );
c|/= tc;
mean Response |/= (tmeanY = Col Mean( RN ));
stdev Response |/= Col Std Dev( RN );
 
Include( "Exact Factors for ANOM.JSL" );

Show(SigLevel);
Show(alpha);
PMP = h(SigLevel, k, nu) * Sqrt( MSE ) * Sqrt( (k-1) / tc );

UDL = tmeanY + PMP;
LDL = tmeanY - PMP;

cc = Control Chart(
 Sample Size( TN ),
 KSigma(3),
 Chart Col( RN,
  XBar(
   Needle(1),
   Connect Points(0),
   Show Control Limits(0)
  )
 )
);

ccr = cc << Report;

ccr[TextBox(3)] <<Delete;

ccr[AxisBox(2)] << Max( k+1 );

ccr[FrameBox(1)] << Add Graphics Script(
 Pen Color( 3 );
 H Line( LDL );
 H Line( UDL );
);

ccr[OutlineBox(2)] << Prepend(
 tb = Text Box( "alpha = " || Char( alpha[SigLevel] ) )
);

ccr[OutlineBox(2)] << Prepend(
 tb2 = Text Box( "LDL = " || Char( Round(LDL, 2) )  )
);


ccr[OutlineBox(2)] << Prepend(
 tb3 = Text Box( "UDL = " || Char( Round(UDL, 2) )  )
);

updateChart = Expr(
 PMP = h(SigLevel, k, nu) * Sqrt( MSE ) * Sqrt( (k-1) / tc );
 UDL = tmeanY + PMP;
 LDL = tmeanY - PMP;
 ccr[PictureBox(1)] << Reshow;
 tb << Set Text( "alpha = " || Char( alpha[SigLevel] ) );
 tb2 << Set Text( "LDL = " || Char( Round(LDL, 2) )  );
 tb3 << Set TExt( "UDL = " || Char( Round(UDL, 2) )  );
);

ccr << Append(
 OutlineBox( "Analysis of Means",
  H List Box(
   Button Box( "alpha = 0.1",
    SigLevel = 1;
    updateChart;
   ),
   Button Box( "alpha = 0.05",
    SigLevel = 2;
    updateChart;
   ),
   Button Box( "alpha = 0.01",
    SigLevel = 3;
    updateChart;
   ),
   Button Box( "alpha = 0.001",
    SigLevel = 4;
    updateChart;
   )
  ),
  Outline Box( "Means/Standard Deviations",
   Table Box(
    String Col Box( "Treatment", a ),
    Number Col Box( "Count", c ),
    Number Col Box( "Mean", mean Response ),
    Number Col Box( "Std Dev", stdev Response )
   )
  )
 )
);

JMP's Analysis of Means for Un-Equal Sample Sizes
Clear Globals();

R = Column Dialog(
 Resp = Col List( "Y, Response",
  Data Type( Numeric ),
  Min Col( 1 ),
  Max Col( 1 )
 ),
 Treat = Col List( "X, Treatment",
  Min Col( 1 ),
  Max Col( 1 )
 ),
 V List(
  "Alpha",
  SL = Radio Buttons( "0.1", "0.05", "0.01", "0.001" )
 )
);

If( r["Button"] == -1, Throw( "User cancelled" ) );

TN = Column( R["Treat"] );
RN = Column( R["Resp"] );
SigLevel = R["SL"];

alpha = {0.1, 0.05, 0.01, 0.001};

Summarize(
 a = By( TN ),
 c = Count,
 mean Response = Mean( RN ),
 stdev Response = Std Dev( RN )
);
Ymin = Min( mean Response );
Ymax = Max( mean Response );

k = N Items( a );
K1 = Log( k );
K2 = Log( k-2 );
a1 = a[1];

tc = Sum(c);
MSEnum = Summation( i=1, k, ((c[i]-1)*stdev Response[i]^2) );
nu = tc - k;
MSE = MSEnum / nu;

Insert Into( a, "Total" );
c |/= tc;
mean Response |/= (tmeanY = Col Mean( RN ));
stdev Response |/= Col Std Dev( RN );
 
Include( "Exact Factors for ANOMUB.JSL" );

Show( SigLevel );
Show( alpha );

PMP = {};

UDL = {};
LDL = {};

For( i=1, i<k+1, i++, Insert Into( PMP, m(SigLevel, k, nu) * Sqrt(MSE)*Sqrt((tc - c[i])/(tc*c[i]))));
For( i=1, i<k+1, i++, Insert Into( UDL, tmeanY + PMP[i]));
For( i=1, i<k+1, i++, Insert Into( LDL, tmeanY - PMP[i]));

cc = Control Chart(
 Sample Size( TN ),
 KSigma(3),
 Chart Col( RN,
  XBar(
   Needle(1),
   Connect Points(0),
   Show Control Limits(0)
  )
 )
);

ccr = cc << Report;

ccr[TextBox(3)] << Delete;

ccr[AxisBox(2)] << Max( k+1 );

For( i=1, i<k+1, i++,
 Eval(
  Substitute(
   Expr(
    ccr[FrameBox(1)] << Add Graphics Script(
     Pen Color( 3 );
     H Line( sss, eee, LDL[iii] );
     H Line( sss, eee, UDL[iii] );
    );
   ),
   Expr( sss ),
   i - 0.5,
   Expr( eee ),
   i + 0.5,
   Expr( iii ),
   i
  )
 );
);

ccr[OutlineBox(2)] << Prepend(
 tb = Text Box( "alpha = " || Char( alpha[SigLevel] ) )
);

ccr[OutlineBox(2)] << Prepend(
 tb2 = Text Box( "LDL = " || Char( Round(LDL, 2) )  )
);


ccr[OutlineBox(2)] << Prepend(
 tb3 = Text Box( "UDL = " || Char( Round(UDL, 2) )  )
);

updateChart = Expr(
 For(i=1, i<k+1, i++, PMP[i]=m(SigLevel, k, nu)* Sqrt(MSE)*Sqrt((tc - c[i])/(tc*c[i])));
 For(i=1, i<k+1, i++, UDL[i]= tmeanY + PMP[i]);
 For(i=1, i<k+1, i++, LDL[i] = tmeanY - PMP[i]);

 ccr[PictureBox(1)] <!<! Reshow;
 tb << Set Text( "alpha = " || Char( alpha[SigLevel] ) );
 tb2 << Set Text( "LDL = " || Char( Round(LDL, 2) )  );
 tb3 << Set TExt( "UDL = " || Char( Round(UDL, 2) )  );
);

ccr << Append(
 OutlineBox( "Analysis of Means",
  H List Box(
   Button Box( "alpha = 0.1",
    SigLevel = 1;
    updateChart;
   ),
   Button Box( "alpha = 0.05",
    SigLevel = 2;
    updateChart;
   ),
   Button Box( "alpha = 0.01",
    SigLevel = 3;
    updateChart;
   ),
   Button Box( "alpha = 0.001",
    SigLevel = 4;
    updateChart;
   )
  ),
  Outline Box( "Means/Standard Deviations",
   Table Box(
    String Col Box( "Treatment", a ),
    Number Col Box( "Count", c ),
    Number Col Box( "Mean", mean Response ),
    Number Col Box( "Std Dev", stdev Response )
   )
  )
 )
);

Wednesday, September 9, 2009

How to do Bartlett's Test and Levene's Test in JMP

Lately I have been receiving many queries regarding this topic. So here you go, the detailed steps on how to do variation comparison in JMP using Bartlett's Test and/or Levene's Test. Before you proceed though make sure to read my previous article regarding my experience with JMP's Levene's Testhere.


Step by Step of How to do Bartlett's Test and Levene's Test in JMP
Step1: Prepare your dataset like this
Note: Data is borrowed from JMP's sample dataset Baseball.jmp. Make sure your grouping is in one column (in this case the "Player") and your dataset is stacked (in this case the "Batting") like shown below.Notice that the grouping column is either an ordinal or nominal data while the measurement data is set as continous.
Step2: Go to the Two-way flatform
You will see the interface as shown below. Put your data into the "Y, Response",and the grouping variable into the "X, Factor".
Step3: Click on OK and you will see the following output
Step4: Test for Un-equal Variance
Conduct a Test for Un-equal varaince by selecting the option from the hot button. This is the red triangle as shown below:
Step5: You will see this output
Scroll down and you will see both Bartlett's Test and Levene's Test.
Other results like O'Brien and Brown-Forsythe are included as well.


Tuesday, September 8, 2009

Reliability and Reliability Engineering (Supplemental)

This is a supplementary material for the entry Reliability Part1. A bulk of reliability concepts revolves around the discussion of the Weibull Distribution and its associated Hazard Function.
The Weibull Distribution Function is characterized by 3 parameters namely Beta (for the shape), Eta (for the scale), and Gamma (for the location). Here is a JMP JSL script that demonstrates how the Weibull distribution changes as each of these parameters change. The output window is shown below:



The JMP JSL Script is shown below::



Beta_value = 1.5;
Eta_value = 2.0;
Gamma_value = 0;


EXPO = Expr(
MESSAGEBOX = New Window( "NOTICE:",
Border Box(
top( 10 ),
Left( 10 ),
Panel Box( "Read:",
Text Box( "If the value of the shape parameter Beta = 1, the Weibull becomes an Exponential Distribution" )
)
)
);
Wait( 2 );
MESSAGEBOX << CLOSE WINDOW;
);

TEXT_MESSAGE = "Shape parameter Beta = " || char(Beta_value) ||"; Scale parameter Eta = " || char(Eta_value) || "; Location parameter Gamma = " || char(Gamma_value);

NON_EXPO1 = Expr(Caption( "If the value of the shape parameter Beta = 1, the Weibull becomes an Exponential Distribution. " ));
NON_EXPO2 = Expr(Caption(TEXT_MESSAGE));

Weibull_Window = New Window( "Properties of the Weibull Density and Hazard Functions",
V List Box(
GraphWindow = Graph(
FrameSize(500,500),
Double Buffer,
FrameSize( 500, 300 ),
X Scale( 0, 3 ),
Y Scale( 0, 4 ),
Double Buffer,
Y Function( (Beta_value / x) * (x / eta_value) ^ Beta_value * Exp( -(x / eta_value) ^ Beta_value ), x ); /* density */
Y Function( (Beta_value / x) * (x / eta_value) ^ Beta_value, x ); /* hazard */
Pen Size( 2 );
Pen Color( "red" );
Text Color ("red");
Text Size(12);
text({1,.7},"Weibull Probability Distribution Function",);
Y Function(
(Beta_value / Eta_value) * ((x - gamma_value) / eta_value) ^ (Beta_value - 1) *
Exp( -1 * ((x - gamma_value) / eta_value) ^ Beta_value ),
x
);
Pen Color( "Blue" );
Text Color ("Blue");
Text Size(12);
text({0.8,1.2},"Hazard Function",);
Y Function( (Beta_value / Eta_value) * ((x - gamma_value) / eta_value) ^ (Beta_value - 1), x );
),
H List Box(
V List Box(
H List Box( TB1=Text Box("Shape Parameter Beta"), Text Box(" ") ),
Slider Box(
0,
10,
Beta_value,
TEXT_MESSAGE = "Shape parameter Beta = " || char(Beta_value) ||"; Scale parameter Eta = " || char(Eta_value) || "; Location parameter Gamma = " || char(Gamma_value);
GraphWindow << reshow;
Weibull_Window<<reshow;
If( BETA_value == 1,
NON_EXPO1,
NON_EXPO2
);
)
),
V List Box(
H List Box( TB2=Text Box( " Scale Parameter Eta" ), Text Box(" ") ),
Slider Box(
0,
10,
Eta_value,
TEXT_MESSAGE = "Shape parameter Beta = " || char(Beta_value) ||"; Scale parameter Eta = " || char(Eta_value) || "; Location parameter Gamma = " || char(Gamma_value);
GraphWindow << reshow;NON_EXPO2;
)
),
V List Box(
H List Box( TB3=Text Box( " Location Parameter Gamma" ), Text Box(" ") ),
Slider Box(
0,
10,
Gamma_value,
TEXT_MESSAGE = "Shape parameter Beta = " || char(Beta_value) ||"; Scale parameter Eta = " || char(Eta_value) || "; Location parameter Gamma = " || char(Gamma_value);
GraphWindow << reshow;NON_EXPO2;
)
)
)

)
);



TB1 << Font color( "BLUE" );
TB1 << set Font size( 10 );
TB2 << Font color( "BLUE" );
TB2 << set Font size( 10 );
TB3 << Font color( "BLUE" );
TB3 << set Font size( 10 );








Friday, September 4, 2009

Reliability and Reliability Engineering (Part 1)

From my previous entry 8 Dimensions of Quality I noted the Dr. David A. Garvin identified Reliability as one of the keys strategic objectives that quality practitioners can focus on. In this series I would discuss more the details of Reliability Management and Reliability Engineering. I will depend heavily on two great books, the “Quality Engineering Handbook” by Thomas Pyzdek (second edition, 2003 by Marcel Dekker, Inc), and of course “Juran’s Quality Handbook” by Joseph Juran and Blanton Godfrey (fifth edition, 1999 by McGraw-Hill). I will incorporate as well some examples on how to execute Reliability Analysis in JMP. I am using JMP 7.0.1 for this and will use its documentation as a reference.


Definition of Reliability


In the entry 8 Dimensions of Quality I differentiated quality and reliability by an analogy. If quality is a snapshot of the goodness of a product at a certain point in time, reliability is the consistency of that level of quality as time goes on. Thus, if quality is a photograph, reliability is a video. For a formal definition however, we will quote from Pyzdek’s book:
“…reliability is defined as the probability that a product or system will perform a specified function for a specified time without failure.”
In addition Pyzdek also cautions that for a reliability figure to be meaningful it has to be defined within the context of specific operating conditions. You do not expect a laptop to work in the same level of quality when it is submerged in water as when it is used in a room environment. For that exact reason, warranty is limited to a pre-defined correct way of using a product. In the same manner warranty claims are first examined if valid by comparing the condition in which it failed against the conditions stipulated in the warranty.


Key Measures of Reliability


The following is a list of the common measures of reliability.
  • Mean Time to Fail (MTTF) or Mean Time to First Failure (MTFF) – This applies to products (or systems in general) that can not be repaired once it fails or break-down. When buying a light bulb for example, we assess how long it will take before it finally breaks down. The longer the time before it finally fails, the better we say that light bulb is.
  • Mean Time between Failures (MTBF) – This applies to products (or systems in general) that breaks down but can be repaired and return to use. MTBF is defined as the average exposure a product will take until it will fail again. This exposure value may take a unit of time or count of usage. Within the context of the unit used, the higher the value of MTBF, the better the reliability is. Thus a machine that broke down 20 days after the repair is better than the one that took only half a day. In the same way, an oven that has to be re-calibrated by a thermocouple every 20 uses is much worse than an oven that needs only recalibration every after 500 uses. Both MTBF and MTTF can satisfactory be modelled by an Exponential distribution or by its generalized form, the Weibull distribution
  • Failure Rate – This is the value of 1/MTBF. It is defined as the number of failures per unit of exposure. From the example above an MTBF of 20 days means failure rate of 1 machine per 20 days or 0.05 failures per day. Failure rate is important as it is often can be modelled by a Poisson distribution which facilitates ease of analysis.
  • Mean Time to Repair (MTTR) – This measures the amount of time the product or tool or system in general is down. It is defined as the elapsed time between occurrence of failure and re-endorsement for use. Because of this definition MTTR is only applicable for repairable entities.
  • Availability – This is defined as the proportion of time a product, or tool, or system in general is in a usable or operable state. Thus it is the ratio of time it is not under repair, to the total available time. Thus, Availability = MTBF/ (MTBF+MTTR).


The Life Cycle Model


Most failures of a product occurs either at the early stage (in cases there is an issue in the manufacturing process), or at the late stage where wear, tear, and degradation begins to take effect. This concept is called the System Life Cycle usually modelled by what is known as the Bathtub Curve. A typical Bathtub curve is shown below.



The Bathtub Curve above is produced by a Beta Distribution. In practice though, the infant stage is modelled by a Weibull distribution. That is, we can see the Bathtub Curve as an overlap of Weibull Distributions one of which is for the Infant Stage as shown below:





..to be continued

Wednesday, September 2, 2009

BASIC SPC CONCEPTS: Definition of Quality Part 2

...continuation from previous post

----------------------------

The 8 Dimensions of Quality



In his Harvard Review article, Harvard professor David A. Garvin developed a framework in which he categorized different aspects of quality. He identified eight categories of which he called the 8 Dimensions of Quality. These are Performance, Features, Reliability, Conformance, Durability, Serviceability, Aesthetics, and Perceived Quality. By itself this framework is enough to serve as a guide in developing a quality-focused strategic objectives.

The 8 Dimensions of Quality are as follows:

Performance
Performance refers to the primary reasons why the product or serviced is bought or sought. It is the main measures of a product’s edge against its competitions. For your computer’s Hard Disk Drive for example, performance would be characterized by Memory Capacity, Data Seek Time, and Power consumption. This dimension of quality can be measured and compared from brand to brand. This allows companies to benchmark the performance of their products against competitors and the consumers to objectively rank brands based on individual aspects of performance. This can also be done in the service oriented industries. A heart surgeon’s success rate for example can be compared against other surgeons. Overall performance rankings, however, are more difficult to develop for some cases of services, especially when they involve customer’s judgement and preferences. Haircut’s goodness for example is difficult to compare from barber to barber since it depends on each individual customer’s point of view.

Features
Features are the additional surprises or offers that a product or service has in addition to its performance characteristics. They are usually secondary aspects next to performance, and may not be necessary but provide convenience supplements to the basic functions or services. A regular cell phone for example provides features like clocks, stop watches, and reminder alarms. Cell phones are not mainly bought due to its clocks, but users tend to attracted to cell phones that provides more additional functions and offers more convenience to the users. It is not always easy to separate primary performance functions from secondary convenient features as it is often dependent on the consumer’s needs and wants. A rule of thumb however is that performance involve objective, measurable attributes and being a must, it can be compared from brand to brand or from provider to provider. Features on the other hand can be subjective and not necessarily present for all brands or service providers. Aside from this individual needs and wants can bias a perception on products’ features giving prejudices in how a product is viewed as something of quality.
A note is in order here. A product can derive considerable success in market share by driving its features as a norm and translating a wanted feature into a needed performance measure.

Reliability
Quality is a snapshot. It is the goodness or fitness of use at this present point of time. Reliability however is about maintaining this level of quality for a sustained period of time. Reliability is the assurance that quality will be the same throughout the promised span of time. This period is usually called the warranty period. The Reliability dimension reflects this chance of a product malfunctioning or failing within this warranty period. These are commonly measured in terms of Mean Time to Failure (MTTF), Mean Time between Failures (MTBF), and Failure Rate per Unit Time (lambda, or failure rate). Note that these measures imply that a product is to be in use for a span of time, making it undefined for services or products that are consumed instantaneously. For example, when buying a decaf from a coffee shop, coffee’s reliability may not make any sense. Insurance and other pre-need services do have a reliability dimension though, even if they are services and not products. The key factor is time. If a product or service is to be extended or to be used for a sustained period of time, reliability is a relevant dimension of quality.

Conformance
Conformance is how much the product or service agrees to the expectations. These expectations are usually laid-out through the specifications or a well-accepted standard. This dimension can be measured in either defect rates or customer complaints. It should be emphasized that conformance cut both ways. A product or service should not be below expectation or above expectations. It should be just within the standard. The reason for this is that the best is for sure to be made by the customers or consumers as the baseline for comparison. If one product is above expectation, it will not be viewed as such. Customers will see it as all others are below the norm.

Durability
If Reliability is the assurance that quality will be the same throughout the promised span of time, Durability is the measure of the length of that promised span of time. It is the measure of product life. It can be defined as how much or how long the product can be of use before it needs replacement. A battery’s durability for example can be measured by its service time when used continuously. A punching bag’s durability meanwhile can be measured by the amount of hits and kicks it can take before deteriorating.

Serviceability
Serviceability is about the repair after the product breaks down. It is usually labelled as Customer Support of Product Support. It is how easy or fast the repair can be done, or in the case of a service given it is the courtesy given to the customer. For products or services that re not used or consumed instantaneously, customers usually accepts that it is a given that product will break down or a service will fail. In such cases, concerns about the time before service is restored or how fast a product can be mend arises. As an example I do pay for an internet connection from an Internet Service Provider (ISP). Sometimes the service breaks down and in such cases how the service personnel deals with my calls, how fast can the service be restored, and how effective their corrective actions are do matter to me as a consumer. And as such it greatly affects my decision on either retain my provider’s services or evaluate the possibility of having another provider.

Aesthetics
Aesthetics is the looks of the product or the feel of the service. It is more than the physical look, feel, sounds, taste, or smell however but more on how it is perceived as good by the customer, how the people around the customer see it as good and desirable. Much like the features it depends heavily on personal biases and preferences. As such, this dimension of quality is very difficult to measure. It provides a guide however on which target market to make a niche. For example a gold watch may look good for a top executive, but is a no for an adventurous athlete.

Perceived Quality
Perceived Quality is psychological rather than physical. It is how potential customers or consumers rate the product as good even without experiencing it. It can be of course influenced by exposure or aggressive marketing strategies. The heaviest influence though is reputation based on historical performance. No advertising can amount to the same impact as one catastrophic field failure. In the same manner, no advertising is more credible than a friend’s personal testimony. In cases when potential customers have not experienced the product or the service yet, then connotations of the brand name, its reputations and images, can influence inferences about its perceived quality. This is important because it can be a driving factor in impulsive buying, and can draw the line between being a commodity and a branded product or service.
 
Custom Search