Lab3 TMHP03: System Design Analytics

2019

1. Requirement specification using the QFD matrix:

1. a) Try to understand the given model setup: Have a detailed look on all worksheets. Identify
the model System Characteristics (SC), the System Parameters (both, DP & UP) and the
-already for you defined- customer Requirements. These "input" sheets are marked with green
background in the sheetname. You can see the mathmeatic description by clicking on the
respective cells in the "model" sheet / Column F.

1.b) In the Add-ins menu, choose the DAOtools (macros), go into the QFD section and eexecute
once the "CR comparison" to make a pairwise comparison of the different requirements.
Define the comparison on the "CR priorities" worksheet:
. - 0 if the line CR is less important than the column CR
. - 1 if the CRs are equal important
.- and 2 if the line CR is more important than the column CR.
After you have done this, run "CR priorities" in order to calculate the CR priorities.
!! Do NOT run again a "CR comparison, otherwise your define CR pairwise comparison
will be reseted/cleared!!

1.c) In theDAOtools choose the "New QFD matrix" command. This will generate the Quality
Function Deployment (QFD) matrix.
Fill the table with reasonable influence foctors connecting the system characteristics with the
customer requirements. Use the values blank, 1,3,9 to indicate strength of relationship:
.- 0 = no relationship/influence
.- 1 = low relationship/influence
.- 3 = medium relationship/influence
.- 9 = large/high relationship/influence

1.d) Use the "Design Priorities" command to calculate the priorities for the system
characteristics. Try to interprete this values.
2.a) Analyse the motorcycle model by using the "Design sensitivity" command in the "Analysis"
menu in the DAOtool. Check out the Normalized sensitivity matrix, (NSM). This command will
calculate the sensitivities by pertubating the "systemParameters" values of the Design
Parameter (type "DP"), by the value in "Variability". The values in "systemParameters" are used
in the model pointed to by the cells under "Value" in "systemCharacteristics".
2.b) Change the model "SystemParameters" to try to achieve the target values. The NSM matrix
can be used to give guidance for which parameters that could be changed in order to get the
desired system characteristics.

2.c) Study the correlation between the System Characteristics using the Adjusted System
Characteristics Correlation (ASCC) command. What are the principle conflicts in the design?

2.d) Study the influence of uncertainty in the "RM (Robustness matrix)" and the "EIM (Effective
influence matrix)" respectivelly the "TSI (Total Sensitivity Index)" sheets. Decide what uncertainty
that dominates for each system characteristics. Use "conditional formatting" in the Format menu
to highlight the "hot spots" in the design, What are the principle uncertainties?

2.e) Run the Establish object function command note the objective function value you obtained
when you tuned the parameter. Then run the "Optimize using COMPLEX-RF". This will optimize
the design parameters. Is the found solution improved? (There might be a glitch here causing an
error here, but it should be sufficient to redo the command). (Note that the target values can be
adjusted to compensate for uncertainty in the system characteristics.)
2.f) Set "Cost" as a demand ("D") and run again the Establish object function and run the
optimizer (Optimize using COMPLEX-RF) once more. What happens?

3. a) Make your own example by inserting the following model. Start by creating a total new
workbook and rename it. Begin by running the "Design sensitivity" command, this will create new
sheets to enter the formulas and parameters. Write the new formulas in the cells under "Value" in
the "System characteristics" worksheet, and write new variable names. "Variability" must also be
adjusted to be small values compared to the actual values (i.e 0.01). Then write the parameters
x1, x2, x3 in "System parameters" worksheet. Then do the sensitivity analysis.
Note that the "system group" columns must not be empty, but they can be deleted. "Type" must
be "DP" in the "systemParameters".

y 1  x1  0 .5 x 2
y 2  2 x1 x2
y 3  x1 x 2 x3

The point of analysis should be.

x 10=1
x 20=1
x 30=1
3 b) Create the system characteristics correlation (ASCC) and try to interprete the 3x3 ASCC
matrix values.

4) Take the car (SUV) model from the previous lab (Statistical model Analyisi; lab 1) and make a
sensitivity analysis on that. Use acceleration (est_acc2)and fuel consumption (est_fuel) as
"systemCharacteristics" and length, frontarea and power as "systemParameters". Start by
making a new copy of the workbook and proceed as in the previous example (except optimizing).
Note that It is possible to connect the cells cross workbooks. So you do not have to rewrite the
model.
4 b) Analyse the system characterstic correlations (ASCC) of your SUV.
Are the ASCC values reasonable and can you explain/motivate the values?

Submission: Like in the first laboratory, upload the worksheet(s) AND your report including LiU
ID (e.g. "abcxy123") on the LISAM course submission

e.g. "Lab2_abcde123_zyxwv987"
If you are in a group of two, please add the non-submitting partner to the group when submitting!
Answer:

Answer:

Note that you can create variable names in Excel

from "create from selection" under the "Formulas" tab
the text i.e. to the left of the value will then be the
variable name. In this way you do not need to use cell
labels. There must be atleast on column "Sub groups"
before the "Name" column in the worksheet.
 Design Analysis
Adjusted System Characteristics Correlation
###
###
C
ha
ra
ct
er
is
tic

y1

y2

y3
s

94.18 7.39 112.16

y1 94.18 1.00 -0.02 -0.02
y2 7.39 -0.02 1.00 1.00
y3 112.16 -0.02 1.00 1.00
Motorcycle
Relative
Target model Type GT(1)
System characteristics values Formula Unit uncertainty LT(-1)
y1 1 1.50 0.3 1
y2 1 2.00 0.2 -1
y3 1 1.00 0.1 1
0.1 -1

y1  x1  0 .5 x 2
y 2  2 x1 x2
y 3  x1 x 2 x3
Demand (D)
or whish (W) System characteristics priorities
W 1.00
W 1.00
W 1.00
W 1.00
Note that you can create variable names in Excel from "create from selection" under the
"Formulas" tab the text i.e. to the left of the value will then be the variable name. In this way
you do not need to use cell labels. There must be atleast on column "Sub groups" before the
"Name" column in the worksheet.
Motorcycle
System group System group System parameter Value Variabili Unit Type
x1 1.00E+00 0.01
x2 1.00E+00 0.01
x3 1.00E+00 0.01
Lower limit Upper limit
Customer requirements
Fun 0.00
Practical 0.50
Environmental friendly 1.00
Safe 1.50
Economical 2.00
SubsystemSubsystemSystem parx System chary Adydx0 dydx0 dydx
Motorcycle Motor Motor powe 17063.3 Range 94.17857 0.031196 -0.031196 -0.000172
Motorcycle Chassis Chassi cos 1427.508 Range 94.17857 0.322598 0.322598 0.021273
Motorcycle SafetySys Safety wei 16.17096 Range 94.17857 0.068991 -0.068991 -0.376886
Motorcycle Battery Battery wei 25.40543 Range 94.17857 0.897923 0.897923 3.315304
Motorcycle Motor Motor powe 17063.3 Acceleratio 7.390939 0.976307 -0.976307 -0.000423
Motorcycle Chassis Chassi cos 1427.508 Acceleratio 7.390939 0.24563 -0.24563 -0.001271
Motorcycle SafetySys Safety wei 16.17096 Acceleratio 7.390939 0.052416 0.052416 0.022631
Motorcycle Battery Battery wei 25.40543 Acceleratio 7.390939 0.080605 0.080605 0.022631
Motorcycle Motor Motor powe 17063.3 Top speed 112.1587 0.322941 0.322941 0.002123
Motorcycle Chassis Chassi cos 1427.508 Top speed 112.1587 0.107541 0.107541 0.008444
Motorcycle SafetySys Safety wei 16.17096 Top speed 112.1587 0.022966 -0.022966 -0.149813
Motorcycle Battery Battery wei 25.40543 Top speed 112.1587 0.035318 -0.035318 -0.149813
Motorcycle Motor Motor powe 17063.3 Recharge t 1 0 0 0
Motorcycle Chassis Chassi cos 1427.508 Recharge t 1 0 0 0
Motorcycle SafetySys Safety wei 16.17096 Recharge t 1 0 0 0
Motorcycle Battery Battery wei 25.40543 Recharge t 1 0 0 0
Motorcycle Motor Motor powe 17063.3 Handling 1.560729 0.046561 -0.046561 -4.26E-06
Motorcycle Chassis Chassi cos 1427.508 Handling 1.560729 0.481463 0.481463 0.000526
Motorcycle SafetySys Safety wei 16.17096 Handling 1.560729 0.103074 -0.103074 -0.009313
Motorcycle Battery Battery wei 25.40543 Handling 1.560729 0.158507 -0.158507 -0.009313
Motorcycle Motor Motor powe 17063.3 Safety leve 1.617096 0 0 0
Motorcycle Chassis Chassi cos 1427.508 Safety leve 1.617096 0 0 0
Motorcycle SafetySys Safety wei 16.17096 Safety leve 1.617096 1 1 0.1
Motorcycle Battery Battery wei 25.40543 Safety leve 1.617096 0 0 0
Motorcycle Motor Motor powe 17063.3 Running co 0.041516 0.031196 0.031196 7.59E-08
Motorcycle Chassis Chassi cos 1427.508 Running co 0.041516 0.32267 -0.32267 -9.38E-06
Motorcycle SafetySys Safety wei 16.17096 Running co 0.041516 0.068715 0.068715 0.000167
Motorcycle Battery Battery wei 25.40543 Running co 0.041516 0.10567 0.10567 0.000167
Motorcycle Motor Motor powe 17063.3 Emissions 0 0 0 0
Motorcycle Chassis Chassi cos 1427.508 Emissions 0 0 0 0
Motorcycle SafetySys Safety wei 16.17096 Emissions 0 0 0 0
Motorcycle Battery Battery wei 25.40543 Emissions 0 0 0 0
Motorcycle Motor Motor powe 17063.3 Cost 5100.39 0.327871 0.327871 0.098
Motorcycle Chassis Chassi cos 1427.508 Cost 5100.39 0.280023 0.280023 1
Motorcycle SafetySys Safety wei 16.17096 Cost 5100.39 0 0 0
Motorcycle Battery Battery wei 25.40543 Cost 5100.39 0.401501 0.401501 78.75
Design Analysis
Normalized Sensitivity Matrix
11/25/2019
19:08:34

Motorcycle
System
characteri
System Target Actual stics
characteristics Unit value value priorities
Range km 100.00 94.18 1.32 1.00
Acceleration (0-70) s 8.00 7.39 1.35 1.00
Top speed km/h 100.00 112.16 0.49 1.00
Recharge time hour 1.00 1.00 0.00 1.00
Handling 1.00 1.56 0.79 1.00
Safety level 1.00 1.62 1.00 1.00
Running cost/km EUR/km 0.20 0.04 0.53 1.00
Emissions 1.00 0.00 0.00 1.00
Cost EUR 4000.00 5100.39 1.01 1.00

System
parameter
s priorities
Design Analysis
Normalized Sensitivity Matrix
11/25/2019
19:08:34

Battery weight
Safety weight
Motor power

Chassi cost
System Target Actual
characteristics Units value value 17063.30 1427.51 16.17 25.41 System characteristics pr
Range km 100.00 94.18 -0.03 0.32 -0.07 0.90 1.00
Acceleration (0-70) s 8.00 7.39 -0.98 -0.25 0.05 0.08 1.00
Top speed km/h 100.00 112.16 0.32 0.11 -0.02 -0.04 1.00
Recharge time hour 1.00 1.00 0.00 0.00 0.00 0.00 1.00
Handling 1.00 1.56 -0.05 0.48 -0.10 -0.16 1.00
Safety level 1.00 1.62 0.00 0.00 1.00 0.00 1.00
Running cost/km EUR/km 0.20 0.04 0.03 -0.32 0.07 0.11 1.00
Emissions 1.00 0.00 0.00 0.00 0.00 0.00 1.00
Cost EUR 4000.00 5100.39 0.33 0.28 0.00 0.40 1.00
System para 1.74 1.76 1.32 1.68
System characteristics priorities
Motorcycle
System Characteristics Target value Formula Unit Intermediate results
Range 100.00 #DIV/0! km Motor weight
Acceleration (0-70) 8.00 #DIV/0! s Total weight
Top speed 120.00 #DIV/0! km/h Total weight incl dirver
Recharge time 1.00 1.000 hour Cruise speed(m/s)
Handling 0.80 #DIV/0! Chassi weight
Safety level 1.00 #DIV/0! Motor cost
Running cost/km 0.02 #DIV/0! EUR/km CdA
Emissions 1.00 0.000 Power reqiured cruise
Cost 4000.00 0.098 EUR Battery capacity
Battery power
Available power
Motor efficiency
Battery efficiency
Battery cost
Energy running cost
Battery running cost
Running cost
Formula Unit
1.00 kg
#DIV/0! kg
#DIV/0! kg
0.00 m/s
#DIV/0! kg
0.10 EUR
#DIV/0!
#DIV/0! W
0 Wh
0 W
0.00E+00 W
6.32E-01
0.00E+00
0 EUR
#DIV/0! EUR/km
#DIV/0! EUR/km
#DIV/0! EUR/km
Parameters for COMPLEX-RF
Parameter Value Comment
alpha 1.3 Reflection factor, >1, Box recommended 1.3
betha 0 Randomizaton factor >=0, 1 is a large factor
gamma 0 Forgetting factor >=0, 1 is a large factor
tolx, 0.01 Relative tolerans in x
Noevalfact 199.3157 Max number of evaluations = Noevalfactor*k
Write optim 1 Should be TRUE or FALSE
adjustTarg 1 Should be TRUE for robust optimization by moving target values