Available online at www.sciencedirect.

com

ScienceDirect
Procedia CIRP 60 (2017) 332 – 337

27th CIRP Design 2017

Product Family Flexible Design Method based on Dynamic Requirements

Uncertainty Analysis
Wei Weia,*, Jun Jib, Thorsten WuestcˈFei Taoa
a
Advanced Manufacturing Technology and Systems Research Center, Beihang University, Beijing, China, 100191.
b
Information Center of China North Industries Group Corporation, Beijing 100089, China.
c
Industrial and Management Systems Engineering Department, Benjamin M. Statler College of Engineering and Mineral Resources, West Virginia University,
Morgantown, WV 26506, United States.
* Corresponding author. E-mail address: weiwei@buaa.edu.cn

Abstract

Developing product families has been recognized as an efficient and effective means to realize sufficient product variety to satisfy a range of
customer and support mass customization manufacturing. This paper presents a product family flexible design method based on dynamic
requirements uncertainty analysis. The product family dynamic uncertain requirements analysis and forecasting techniques is researched in this
paper, aims to improve the dynamic response ability of the product family to the change of the market demand in the future. Firstly, the
multi-domain transmission mode of dynamic requirements was discussed and the product family flexible design model was proposed. Then the
sensibility of design parameter to dynamic requirements was analyzed and the variation index of design parameter was calculated. As a result,
the product platform of product family flexible design was constructed. A product family flexible design prototype system was also developed,
and the application verification for flexible design of forging press product family was carried out to demonstrate the validity of the proposed
method.
©©2017
2017TheTheAuthors. Published
Authors. by Elsevier
Published B.V. This
by Elsevier B.V.is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the scientific committee of the 27th CIRP Design Conference 2017.
Peer-review under responsibility of the scientific committee of the 27th CIRP Design Conference
Keywords: Product Family; Flexible Design; Dynamic Requirements; Uncertainty analysis; Immune clonal algorithm;

A. INTRODUCTION In the few years, scholars have done a lot of researches

As the more effective method to meet the growing
individual request, the technology of platform-based product
development was the hot topic in the past few years, the
product platform offer plenty of benefit: different products
which come from the same platform can share a core set of
common platform elements. As a result, company can reduce
development time and costs, simplify system complexity and
improve the ability of upgrade products. But the weakness of
platform-based product development exposed gradually. Due
to the product family members share too many common
elements to highlight individuality. For that reason, the
research on flexible product family design is worth. Compare
with product platform, the flexible product platform consider
the uncertainty factors includes the product development in
the future. The flexible platform based on dynamic
requirements uncertainty is the more effective method to meet
problem of mass customization.
on the question of flexible platform and dynamic
requirements uncertainty. Suh E S [1] proposed a flexible
platform strategy by incorporating flexibility into product
platforms, the design process generate multiple design
alternatives by analysis demand uncertainty [2], and then
filtered the profitable flexible component design with
minimum cost and economic profitability, the proposed
process was demonstrated in automotive application case. Q
Ma [3] integrated the method of parametric design into
flexible product platform, and presented a rapid design
method base on flexible product platform by adjusting key
parametric, the application on belt conveyor demonstrated
this method fit the problem of product family design
efficiently. Kangyun Shi [4] combined the advantages of the
modular-based and the scale-base product platform design,
analyzed the unknown uncertainties related to customer needs,
find out the key design parameters, through mapping between

2212-8271 © 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the scientific committee of the 27th CIRP Design Conference
doi:10.1016/j.procir.2017.01.037
 Wei Wei et al. / Procedia CIRP 60 (2017) 332 – 337 333

the parameters and the physical structure, the critical flexible ­&1  ½ ­ )5  ½
(1)
° ° ° °
elements were determined, and the flexible product platform °  ° °  °
$® ¾
® ¾
was constructed by extracting the common and the flexible °  ° °  °
°¯&1 V °¿ °¯)5 Q °¿
elements. Uncertainty was a concept appeared in the field of
philosophy, statistics, economics, psychology and ª U U  UN  UQ º
engineering science. Aim at the optimization problem of « » (2)
« U U  UN  UQ »
uncertain structures design, J Cheng [5]~[6] proposed a «       »
$ « »
constrained interval optimization model firstly, Mechanical «UP  UP   UPN  UPQ »
performance indices was described as the objective and «       »
« »
constraint functions of the design vector and interval ¬«UV  UV   UVN  UVQ ¼»
uncertain parameters in this model. An algorithm integrating Where, ULM means the impact of the j -th function to
radial basis function, interval analysis, and non-dominated i -th request. The bigger the ULM is, the more important M -th
sorting genetic algorithm (NSGA-II) was put forward to solve function is to the i -th request: 0  rij  1 .
the optimization problem of uncertain structures design.
2) Analysis and Forecast of dynamic requirements
Paper [7] believed uncertainty was composed by
inner-uncertainty and outer-uncertainty. A large number of Uncertainty
researches about dynamic uncertain requirements forecasting Customer requirements always is fuzzy and uncertain, the
method were done in Paper [8] ~ [11]. fuzzy mathematical theory [12] is applied in this paper to
Concept of flexible product platform was introduced to convert the uncertain requirements into numerical model. The
improve the ability of product family dynamic response to demands intensity can be divided into six levels, and defined
market changes. After discussed the multi-domain by the values in Table 1:
transmission mode of dynamic requirements. A two-stage
multi-objective optimization based platform design Table 1 Demand intensity and the corresponding value evaluation index
methodology (MOPDM) is proposed. Finally, universal Intensity strongest stronger common weak weaker Irrelevant
motor platform is designed by the process proposed in this Value 9 7 5 3 1 0
paper. The result shows that the method this paper proposed
is better than one-stage MOPDM. A matrix about the requirement important is established:
9 >9 9  9 1  9 1 @ (3)
B. The Mathematic Model of Dynamic Requirements
Where V means the demand intensity matrix, where V1ˈ
Uncertainty
V2,VN can only value from the table1.
1) the multi-domain transmission mode of dynamic Due to requirements can be mapped to function, the
requirements delphi technique [13] is applied to complete the requirement
Product development process is described by four
– function matrix˖
design-domains according to axiomatic design theory. There
is different design variables in each domain, it can also be ª $ %  0 680 º
« »
described custom needs, function requirements, design « , [,$ [,%  [,0 ,V » (4)
« ,, [,,$ [,,%  [,,0 ,, V »
parameters and process variables. The neighboring domain % 9« »
can impact each by variable mapping. It can be seen from «       »
« 1 [1$ [1%  [10 16 »
Fig.1, changes in customer’s requirements can be mapped to « »
the change of the function and structure of product. The ¬«680 $N %N  0N  ¼»
dynamic needs lead to various products. It is a complex Where, [ 10 means degree the function 0 -th contributes
problem to balance the ability of evolution and development to the requirement 1 -th, 1 6 means the importance of
time. In conclusion, market uncertainty cause the structure’s the 1 -th requirement among all the requirements, the
uncertainty. higher 1 6 is, the more important this requirement is, 0 N means
Design product product the contribution to all the requirements. The higher 0 N is, the
Task
function
design
physical
manufacture
process more importance function is.
Consumer domain domain
Domain domain
(FRn)
(DPm) (PVp) The importance higher than liminal value I is regarded
(CNs) DP1 ⁞
CN1
CN2
FR1
FR2
⁞
DPh
PVe
⁞ as important requirements, and contribution higher than
FR3 PVg
⁞
CNn
⁞
FRn
DPl
⁞ ⁞
PVi
liminal value M is regarded as key functions. As result:
>&1  &1   &1 [ @
DPm ⁞
&1 (5)
)5 >)5  )5   )5 = @
Fig 1. The multi-domain transmission mode

The mapping-relationship between consumer domain and

function domain can described by the formula (1), A is
named as request-function matrix.
334 Wei Wei et al. / Procedia CIRP 60 (2017) 332 – 337

C. Flexible Platform Design based on Dynamic should be gave subjectively, and The parameters whose
sensitivity higher than O and variation index higher than E
Requirements Uncertainty Analysis
is regarded as design variable parameters, other parameters
3) The sensitivity of design parameters on the analysis of can be regarded as design constant parameters.
demand uncertainty 4) The process of two-stage multi-objective optimization
There is multiply design-parameters in product family, based platform design methodology
before establishing the product platform, the important thing Many researches used one-stage optimization procedure
is divided those design-parameters into the common platform which optimized the platform settings and corresponding
parameters and the non-platform variables according to their members of family in one stage. When the number of design
properties. The result of division will affect the commonality variables increase, the dimensionality of the optimization
of product platform. In order to divide those parameter problems become too complex and expensive to be dealt only
objectively, the concept of sensitivity [14] and variation index by one optimization algorithm. A two-stage optimization
[15] are introduced to assist the parameter divide-process. approaches is proposed, which divides the task into two
Sensitivity shows the influence degree of design parameter stages: first, decide which variables are shared and their
on the product performance, the small sensitivity parameters settings for platform configuration, second, generate all
can be regarded as constant parameters in platform, which product’s optimal values for variables. As it shows in Fig 2.
can be optimized on the platform, the large sensitivity
parameters can be regarded as variable parameters, which can Platform Design Indicidual Design
Perform
be optimized on the product. The variation index represent DOE
Combine CI and PI
Design
the variation degree of design parameter. For the platform, Calculate
Model
the bigger the sensitivity, the smaller the index, the better the 0LQ  0D[)˄[˅ EM
mean and
EM
Best
standard compromise
platform. I[  IQ[  OA deviation of OA solution
First partial derivative method is utilized to calculate the design
variables
sensitivity of each parameter. Supposing the target function
of product performance is )[  ^I[ I[  I[ `( P
Fig. 2. Optimization framework of MOPDM
is the number of performance), when
[ 3 N ^[[   [ Q `( Q is the number of parameter)
make )[  get the best result, the product 3 N ’s sensitivity In Fig2, the multi-objective optimization evolutionary
N
of M -th to L -th performance in the is 3 -th product can be algorithm (EMOA) is applied twice. CI denotes the similarity
described as: factor of design variables among instance products and PI
expresses the general performances of product family. They
'IL[  (6) are two competing objectives during the design of scale-based
0 LMN
'[ M
product platform.
Where: '[ M means minor changes of parameter [ ; The particular optimization steps of the MOPDM are given
'IL[  is fluctuation of L -th product performance cause next:
by the minor changes of parameter [ .
Formula (7) is the sensitivity matrix about the 3 -th
N Step1: Identify the design variables through product
product, which presents all the parameter’s fluctuation to all analysis.
the product performance. Step2: Perform DOE to check for possible reduction of
design variables.
ª 'I[  'I[  'I P[ º
« '[  (7) Step3: Identify the constraints and multi-objectives for
« '[  '[  »»


« 'I[  'I[  'I P[ » optimization


0N « '[ '[  '[  » Step4: Make sample runs to determine best MOEA
« 
»
«     » parameters for the problem and optimize every instance
'I
«  [ 'I[  'I P[ »
«¬ '[ Q
 independently base on MOEA
'[ Q '[ Q »¼
Step5: Calculate mean (mi) and standard deviation (δi) of
The global sensitivity of M -th design parameter is: design variables, then identify platform common parameter
+

¦ 0 LMN and scaling variable sets by δi /mi

0 *LM N  (8) Step6: Set platform common parameters as mi., identify
+
the constraints and design variables.
Where: 0 LMN is the local sensitivity about N -th product
Step7: Optimization to derivative products based on
of M -th design parameter to L -th performance; + is the
platform common parameters using MOEA.
number of product which come from product family.
The variation index of product can be measured by mean D. The Overview of Product Family Flexible Design
and variance:
Method Based on Dynamic Requirements Uncertainty
GM G M  PM (9)
Analysis
Where, G M is variation index of the design parameter,
P M is mean of [ M , G M is variance of [ M . The process model of flexible platform based on dynamic
After get the sensitivity and variation index, the liminal requirements uncertainty is illustrated in Fig.3. Firstly, get the
value O of sensitivity, liminal value O of variation index requirements form custom, through fuzzy mathematic theory
 Wei Wei et al. / Procedia CIRP 60 (2017) 332 – 337 335

and delphi technique, converting the uncertain requirements variables are scaled up or down to form a series of derivative
to important requirements CN and key function FR. Second, products and optimization every derivative products. For a
divide design parameters into common parameters P and the single derivative product i, platform common parameters and
scaling variables V based on the sensitivity and variation scaling variables constitute all design variables. Furthermore,
index. Third, after the design parameter is reduced, according the product family is composed of the scale-based product
to the design function and constraints of product family, platform and its derivative instances.
while keeping the common parameters constant, the scaling

Uncertain requirement acquisition The first-stage(determine Second stage(optimization

and determination common parameters) every case)
ª&1 º Product 1
« » Constant Design
Requiremens get &1 «&1 » Common Parameters 3 ‰9
« » Parameter
9 >9 9  91  91 @ « »
¬&1V ¼
GM GM  PM
 3L Ÿ
Product 2
3 ‰9
Customer Key function 3LL ˈ
ˈˈP MOPDM
Uncertain
ª $ %  0 680º
and MOPDM
Requirements « » important Product i
«, [,$ [,%  [,0 ,V » ª'I[ 'I[ 'I [º
requirements « '[  P »
3 ‰9L
« ,, [,,$ [,,%  [,,0 ,,V » «  '[ '[ » Scaled Design
% 9« » «'I[ 'I[  'IP[»
ª&1 º
«       » 0N « '[ '[ '[ »
Parameter
« » «1 [1$ [1%  [10 16 » «  »
«&1» « » «     »   9M ž
&1 ª)5 º Product k
« »
¬«680 $N %N  0 N  ¼» « » «'I[ 'I[  'IP[» 9MM ˈ
ˈˈQ
« » )5
¬&1Q ¼ )5 « » «¬ '[Q '[Q '[Q »¼ 3 ‰9N
Important requirement and « »
« » Scaled Parameters
key function select ¬)5N ¼

Fig 3. The development process of flexible product family

E. CASE ILLUSTRATION: SERIES OF TRACTION ­0 0  0  0

°
MACHINE PLATFORMS °K 3RXW  3LQ 3LQ  3ORVV   3LQ
(11)
°
5) The uncertain requirement acquired and analyzed °
°0  >
S 5   5  K  K UV @
°
Motor is a kind of very common power plant, the example °0  S 5  K  K K UV
 
°
of one motor manufacturing company’s portable motor °
°0  ^>K   5  K  K @= F 6 ZD  >K  5  K @= V 6 ZI `UF
product platform is taken to show the whole process of design ®
°3 , 5 D  5 V   ,
product platform and product family, including how to ° ORVV
°
transform uncertain demands to the certain demands, and then ° U >K  5  K  K @= F
°5 D
mapped demands to the design of design parameters. Finally, ° 6 ZLUH
°
the proposed method is proved available. ° U >K  5  K @= V
°5 V
According to market survey, we get some requirements °
¯ 6 ZLUH
about the motor, CN= {easy to carry, save electricity, high
precision}, motor function FR= {little-weight, high-efficiency, The torque of motor:
lower-price, large-torque}. 1F (12)
Through the analysis and evaluation of experts, we get the 7 ),
S
requirement-function matrix bellow:
The range of design variable and the constraint conditions
ª )5  )5  )5  )5  680 º
« » (10) of each design variable are shown in Table 2, the bold font
« &1       »
% « &1       » section is the design parameters.
« »
« &1      » In order to convert two optimizing problem to the
«¬680     »¼
minimum problem, the maximal efficiency (η=Pout/Pin) is
From the matrix, FR1 ǃ FR2 is the primary function, transferred into the minimal efficiency loss
corresponding: little-weight and high-efficiency. CN1ǃCN2 (ηloss=1-η=Ploss/Pin).
is the main requirements, corresponding: easy to carry and
save electricity. Therefore, weight and efficiency is the main Table 2 Main design parameters information
design target about this platform. The optimization model of Parameter Value Ranges/ Constraint
motor platforms is illustrated as: Parameter Meaning
Name Condition
F = (max (torque), min (weight)).
6) Mathematical model of motor M weight of motor M İ 2.5Kg
Motor platform can generate derivative motor just change M1 weight of stator -
the scale of the number of stack, keep the constants design M2 weight of motor armature -
variable same. The motor platform is a typical scale-based
product platform. Design a family of same power motors that M3 weight of motor coil -
satisfies a variety of torque by scaling a common motor ¨ efficiency of motor ¨≥ 25%
platform around some scaling variables of the motor [16]. Pout output power P = 300
336 Wei Wei et al. / Procedia CIRP 60 (2017) 332 – 337

Pin input power -

Ploss loss power -
T={0.05, 0.1, 0.15, 0.2,
T torque of motor
0.25, 0.3, 0.4, 0.5}N·m
Rs resistance of wire -
Ra resistance of armature -
R > h1
R outer diameter of stator
1 ≤ R ≤ 10mm (c)T=0.15N·m (d) T=0.2N·m
h1 thickness of stator 1 ≤ h1 ≤ 15mm
h2 stack thickness of stator 1 ≤h2 ≤ 7mm
²s density of steel 7800Kg/m3
h3 gap of air 0.5mm
Zc number of rotor turns 100 ≤ Zc ≤ 1500
Swa cross-section of rotor coil 0.2 ≤ Swa ≤ 2.0 m2
Zs number of stator turns 100 ≤ Zs ≤ 600
Swf cross-section of stator coil 0.2 ≤ Swf ≤ 2.0m2 (e)T=0.25N·m (f) T=0.3N·m
²c density of wire 8900 Kg/m3
Swire cross-section of wire 19.625h10-6m3
ρ electrical resistivity of wire 0.0172μ¡gm
I Current size 0 ≤ I ≤ 5A
7) Motor platform of product family design optimization
The multi-objective optimization model about the motor
platform is established; in this stage, artificial immune
algorithm is used to calculate the sensitivity and the variation
(g)T=0.25N·m (h) T=0.3N·m
index of design variables. Take the different torque
requirement as constraint condition, take the design parameter
as the optimization object, take the weight and the efficiency Fig. 4. The process of improved immune clonal algorithm
as the optimization objective. Set the scale of antibodys group In the first stage, the comprehensive optimal result is
Pop=500, maximum times of iterations G=1000, the regarded as the best result, the sensitivity and of every design
probability of crossover immune operation p=0.6. The result
is shown in fig4 (a)-(h). With the increase of torque, the parameter to the optimization target from the comprehensive
efficiency loss and weight also increase, but the overall trend optimal result. When solving the sensitivity of design
is identical. Choose the comprehensive optimal result (the parameters to design targets, the design parameters take
pareto result which get the best result as the formula
different values, save those values and calculate the average,
U PK  P0 ( P P  )) from those results.
variance and variation index through the equation (9). the
threshold of sensitivity is set¬1= 0.20 for weight target, the
threshold of sensitivity is set¬2= 0.15 for efficiency target;
the threshold of variation index is set£= 10%. The design
variables {Swf, Swa, R, I} are the common platform parameters;
{h2, Zc, Zs, h1} are the non-platform variables average as the
value common platform parameter, {I = 4.1, Swf = 0. 35, t
=5.6, Swa = 0.22}.
(a)T=0.05N·m (b) T=0.1N·m
8) Examples of motor design optimization
After ensure the common parameter, just adjust the design
parameter to make the weight and efficiency get the best both,
so, single product instance will optimized by improved
artificial immune algorithm. the value table 3 is the finally
result after optimization and selection.
 Wei Wei et al. / Procedia CIRP 60 (2017) 332 – 337 337

Table 3 Optimization result of electromotor design product instance G. Acknowledgment

common platform Optimization
parameters
non-platform variables
Results
This research is supported by the National Natural Science
ID
Foundation of China (Grant number 51675028, 51505437)
h2 Zc Zs h1 Swa R Swf I M ¨
and the Fundamental Research Funds for the Central
1 0.87 685 47 5.6 0.22 2.3 0.22 3.12 0.44 73.18 Universities.
2 1.15 710 71 5.6 0.22 2.3 0.22 3.37 0.48 71.93
H. Reference
3 1.52 748 62 5.6 0.22 2.3 0.22 3.56 0.53 70.07
[1] Suh E S, Weck O L D, Chang D. Flexible product platforms: framework
4 1.95 816 67 5.6 0.22 2.3 0.22 3.91 0.62 68.69
and case study[J]. Research in Engineering Design, 2007, 18(2):67-89.
5 2.31 882 85 5.6 0.22 2.3 0.22 4.29 0.69 64.84 [2] Suh E S, Weck O L D, Chang D. Flexible product platforms: framework
and case study. Research in Engineering Design, 2007, 18(2):67-89.
6 2.67 965 75 5.6 0.22 2.3 0.22 4.52 0.75 60.33
[3] Ma Q, Tan R, Jiang P, et al. Flexible Product Platform Based on Design
7 3.02 1068 79 5.6 0.22 2.3 0.22 4.85 0.87 57.46 Parameters. Building Innovation Pipelines through Computer-Aided
8 3.45 1120 65 5.6 0.22 2.3 0.22 5.15 0.96 53.62 Innovation, 2011:7-15.
[4] Kang Yunshi, Based on the flexible product platform of product family
9) Comparison of results between different optimization development. Computer Integrated Manufacturing System, 2009,
15(10):1880-1889.
algorithms [5] Cheng J, Liu Z, Wu Z, et al. Direct optimization of uncertain structures
based on degree of interval constraint violation[J]. Computers &
Table 4 Comparison of results obtained by different optimization algorithms Structures, 2016, 164:83-94.
[6] Cheng J, Duan G F, Liu Z Y. Interval multiobjective optimization of
Optimization algorithm Diversity Convergence Run time(s) structures based on radial basis function, interval analysis, and
One-stage MOPDM 0.281 7 0.3067 47.8 NSGA-II[J]. Journal of Zhejiang University-SCIENCE A, 2014,
15(10):774-788.
Two-stage MOPDM 0.2653 0.3562 27.5 [7] Miller R, Lessard D R. Evolving Strategy: Risk Management and the
PPCEM 0.3192 0.3289 37.9 Shaping of Large Engineering Projects. Ssm Electronic Journal, 2007.
[8] Halpern J Y. Reasoning about uncertainty. MIT press, 2005.
In table 4, the product platform concept exploration [9] De Weck O L, Neufville R D, Chaize M. Staged deployment of
method (PPCEM) was proposed in 2001 by Simpson. This communications satellite constellations in low earth orbit. Journal of
method is the very classic method in the field of product Aerospace Computing, Information, and Communication, 2004, 1(3):
119-136.
family design. In order to compare the efficiency and
[10] De Neufville R, de Weck O, Frey D, et al. Uncertainty management for
performance of PPCEM and the method this paper proposed, engineering systems planning and design. Engineering Systems
the concept of Frontier Approaching Criteria [17] and Set of Symposium, MIT, Cambridge, MA. 2004.
Decentralized Diversity Criteria [18] are introduced. As to the [11] Silver M, Deweck O. Time-expanded decision network methodology for
Frontier Approaching Criteria, bigger approaching the designing evaluable systems. Proceedings of the 11th AIAA/ISSMO
Multidisciplinary Analysis and Optimization Conference. Portsmouth,
leading edge value is, better the convergence is. As to the Set
USA: AIAA, 2006:6-8.
of Decentralized Diversity Criteria, smaller the solution sets [12] Peter Guttorp. Fuzzy Mathematical Models in Engineering and
disperse the various values is, better the diversity of pareto is. Management Science[M]. Sole distributors for the U.S.A. and Canada,
The result in table4 is average value after ten times calculate Elsevier Science Pub. Co, 1990.
in one computer. The result shows that the method this paper [13] Linstone H A, Turoff M. The Delphi method: Techniques and
applications[J]. Journal of Marketing Research, 1975, 18(3):363-364.
proposed is better in the pareto diversity, pareto convergence
[14] Scott M J, Antonsson E K. Using indifference points in engineering
and efficiency. decisions. Imprecision in Engineering Design, 2000: 225.
[15] Chen Y L, Chu W L, Xu Y S. Adaptability-oriented parametric product
F. Conclusion and Future Work platform design. Computer Integrated Manufacturing Systems, 2007,
13(5): 877-884.
In allusion to product family flexible design, first, this [16] Simpson T W, Maier J R, Mistree F. Product platform design: Method
paper discusses the multi-domain transmission mode of and application. Research in Engineering Design, 2001.13(1): 2-22.
dynamic requirements, researches the requirements analysis [17] Zitzler E, Deb K, Thiele LˊComparison of multi-objective evolutionary
algorithms: Empirical results. IEEE Transactions on Evolutionary
and forecasting techniques, and proposes a two-stage product Computation, 2000, 18(2) : 173-195ˊ
family flexible design method, introduce the sensitivity and [18] Deb K, Pratap A, Agarwal S. A fast and elitist multi-objective genetic
variation index to filtrate the common platform parameters in algorithm: NSGA-IIˊIEEE Transactions on Evolutionary Computation,
the first stage. Then, this paper proposes an improved 2002, 6(2): 182-197ˊ
artificial immune algorithm, because the fitness function is
evolutionary change, that made more accurate in filtrating
common platform parameters; this algorithm has stronger
global searching ability and local search ability, and the stress
in global search at the same time takes into account the local
search. At last, the developed method is better than one-stage
multi-objective optimization based platform design
methodology.