CN107292007A - A kind of product Design Decision Making based on effectiveness supports system and method - Google Patents
A kind of product Design Decision Making based on effectiveness supports system and method Download PDFInfo
- Publication number
- CN107292007A CN107292007A CN201710429349.0A CN201710429349A CN107292007A CN 107292007 A CN107292007 A CN 107292007A CN 201710429349 A CN201710429349 A CN 201710429349A CN 107292007 A CN107292007 A CN 107292007A
- Authority
- CN
- China
- Prior art keywords
- alternative
- evaluation index
- module
- action
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
System and method is supported the invention provides a kind of product Design Decision Making based on effectiveness, DSS includes base module, single-action function module, multi-purpose function module and overall expected utility module, and each module possesses independent input/output interface.The evaluation index of Alternative designs of the present invention is interval value, and objectivity is strong, can adapt to the uncertainty of desired value in product design process；The dynamic change of product design demand is can adapt to using the method for the DSS, when external environment design changes in demand, the parameter that can be influenceed according to change is handled by input/output interface in corresponding module, avoid and recalculate, the validity of design is improved, and evaluates efficiency high.
Description
Technical field
The present invention relates to a kind of schemes evaluation method of product design, and in particular to a kind of product design based on effectiveness is determined
Plan supports system and method.
Background technology
Intelligent design system provides possibility for the intellectually and automatically that product design is researched and developed, and product design scheme
Evaluate and preferably its important part.Designer is intelligent design system In-put design demand, and system is pushed away via it
Reason module can produce a variety of alternatives.But designer, which often only expects, unique a kind of to be integrated optimal scheme and is used for subsequently
Simulation analysis and actual production.Accordingly, it would be desirable to which a kind of schemes evaluation method realizes the evaluation and sequence of multinomial alternative
To draw optimal case.
Schemes evaluation method is actually a kind of decision process of product design scheme.In the past popular design decision
Method, schemes evaluation method (FDM-GRA), many granularities category such as based on fuzzy decision figure (FDM) with grey correlation analysis (GRA)
Property index Fuzzy estimate and Choquet integral models method, Network Analysis Method, the design decision side of multi-level ATTRIBUTE INDEX can be added
Method, the design Optimal Decision-making method for considering evaluation criterion weight and grey correlation analysis.The studies above is attempted to be based on
Decision-making technique is applied on selection product design optimal case, lays particular emphasis on the situation of the weight with reference to known each design attributes
Under, design research method is given a mark by expertise and evaluated, and scheme validity is low, also have ignored and exists in design
Probabilistic influence.
Currently in order to firearm design process uncertainty present in is solved, it is main to use Monte Carlo simulation and interval
Number scheduling theory, the Robust Optimal Design based on asymptotic global agent model, the fuzzy classification comprehensive evaluation model of products scheme.
Existing achievement in research, realizes decision support to product design process from different perspectives and design parameter is not true
Qualitatively express, but still lack one kind and take into full account that design parameter is uncertain among product design process and design requirement is dynamic
Change this two class and design probabilistic decision support method, have impact on the validity of product design scheme.
The content of the invention
In view of this, system and method, Neng Goushi are supported the invention provides a kind of product Design Decision Making based on effectiveness
The uncertainty designed in product design process is answered, the validity of design is improved, and evaluate efficiency high.
A kind of product Design Decision Making based on effectiveness supports system, and the DSS includes base module, list
Utility function module, multi-purpose function module and overall expected utility module, each module possess independent input and output
Interface；
The base module is used to store conventional design and evaluation index, according to the n related to product demand
Individual evaluation index selects alternative；
The single-action function module by input/output interface receive designer based on utility theory input it is individual to n
The attributes preferred value of evaluation index, fits n single utility function according to attributes preferred value, phase is exported by input/output interface
The function coefficients value answered extracts alternative, with the evaluation index area of alternative on control panel from base module
Between as integrating range, and calculate using single utility function n single-action functional value of each alternative, export to the overall phase
Effectiveness module is hoped, n is positive integer；
The multi-purpose function module, the property value of the n-1 evaluation index inputted using input/output interface builds nothing
Attributes preferred composite equation group, exports the weight k of each evaluation indexiOnto control panel, i=1~n；
The overall expected utility module is based on weight kiThe overall phase of each alternative is exported with functional value with single-action
Hope value of utility and ranking to control panel；
The DSS further comprises reprocessing analysis module, for testing before overall expected utility ranking
The robustness of the alternative of two, Δ k and Δ x, the Δ L for changing the alternative by input/output interface is asked again
The overall expected utility of the Alternative designs of the front two is solved, to judge whether ranking changes, Δ k refers to the change of weight
Change, Δ k is added to the k in multi-purpose function module respectivelyi, the new weight of output is calculated again；It is bent that Δ x refers to single utility function
The change of rate, changes single-action with the function coefficients value being fitted in function module, calculates single-action functional value again accordingly；ΔL
Refer to the interval length change of the evaluation index of alternative, change the evaluation index area of alternative in single-action function module
Between, calculate single-action functional value again accordingly；
Alternative in base module is increased by input/output interface, changes the ranking of all alternatives；
By input/output interface increase base module in evaluation index, single-action with function module increase a single-action letter
Number, changes the weight k of each evaluation indexiAnd the ranking of all alternatives；Change alternative by input/output interface
The evaluation index of scheme is interval, changes the alternative in single-action with the single-action functional value in function module.
Further, system, the decision-making are supported using the product Design Decision Making based on effectiveness as claimed in claim 1
Support method is：
Step one, according to the n evaluation index selection alternative related to product demand in base module；
Step 2, designer is based on utility theory and inputted in single-action function module by input/output interface to n
The attributes preferred value of individual evaluation index, fits n single utility function according to attributes preferred value, exports corresponding function coefficients value
Onto control panel；
Step 3, extracts alternative from base module, and integrated area is used as using the evaluation index interval of alternative
Between, n single-action functional value of each alternative is calculated and exported using single utility function, and n is positive integer；And by defeated
The property value for entering output interface n-1 evaluation index of input is built without preference combinations of attributes equation group, is exported each evaluation and is referred to
Target weight kiOnto control panel, i=1~n；
Step 4, based on weight kiThe overall expected utility and ranking of each alternative are exported with functional value with single-action
Onto control panel；
Step 5, changes Δ k, Δ x or the Δ L of ranking front two alternative, the total of the alternative is solved again
Body expected utility and ranking, to judge whether the ranking of the front two alternative changes.
Further, it is necessary to further be included in the knowledge described in step one when introducing new alternative, after step 4
New alternative and corresponding evaluation index is added in library module, DSS calculates the totality of new alternative
Expected utility and by all alternatives again ranking.
Further, it is necessary to further be included in the knowledge described in step one when introducing new evaluation index, after step 4
Add new evaluation index in library module, designer is based on utility theory in single-action with being inputted in function module to new evaluation
The attribute preference of index, corresponding single utility function is fitted according to attribute preference, exports corresponding function coefficients value；Point
The interval up-and-down boundary of the new evaluation index of all alternatives Shu Ru not be integrated, each alternative side in base module is utilized
The interval and single utility function of the evaluation index of case, exports all single-action functional values；Recalculate weight and based on weight and
Single-action exports the overall expected utility and ranking of each alternative with functional value.
Further, when the evaluation index interval of alternative changes, change in the multi-purpose function module of step 3
Become the interval of evaluation index and recalculate single-action functional value.
Further, the attribute bias data obedience of n evaluation index is uniformly distributed in the step 2.
Beneficial effect：
1st, DSS and method of the invention, obtain the overall of each design based on utility theory and expect effect
With value, the validity height of design；The evaluation index of Alternative designs is interval value, and objectivity is strong, can adapt to product
The uncertainty of desired value in design process；The present invention will solve working flow control processing, when external environment design changes in demand
When, it can be handled in corresponding module, it is to avoid recalculate, carried by input/output interface according to the parameter of change influence
High evaluation efficiency.
2nd, reprocessing analysis module of the invention, is solved by changing design requirement and tests the robust of design again
Property, further increase the validity of design.
3rd, to possess independent input on the control panel defeated for each module that DSS of the present invention includes
Outgoing interface, can quickly be reused, adaptive capacity is strong, improves design efficiency in similar designs or new design.
Brief description of the drawings
Fig. 1 is the flow chart of decision support of the present invention；
Fig. 2 is Modular Flexible stencil design block diagram of the invention；
Fig. 3 is the solution of the present invention knowledge the Library Panel figure；
Fig. 4 is attribute knowledge base and single-action function module panel figure of the invention；
Fig. 5 is overall expected utility module faceplate figure of the invention；
Fig. 6 is reprocessing analysis module map of the present invention.
Embodiment
The present invention will now be described in detail with reference to the accompanying drawings and examples.
System is supported the invention provides a kind of product Design Decision Making based on effectiveness, DSS includes knowledge base
Module, single-action function module, multi-purpose function module and overall expected utility module, each module possess independent defeated
Enter output interface, correspondingly presented on the control panel with modularization.As shown in Fig. 2 the design process of Modular Flexible template
Mainly it is made up of five steps, including：Template establishment, model customization, the design of individual module, integral mold plate design, example are tested
Card.First the design requirement and function of whole design decision process various pieces are analyzed, decision process split into some
The individual complete independent module for possessing standardization input/output interface, entirely the trade-off decision process based on utility theory split into
Scheme knowledge base, attribute knowledge base, single-action function module, multi-purpose function module, overall expected utility module, post processing point
Module is analysed, it is necessary to be extracted to parameter among module design process, and data processing is carried out, finally realizes the reality of module
Exampleization.
Base module includes scheme knowledge base and attribute knowledge base, and scheme knowledge base is used for the design side for storing product
Case, the content that attribute knowledge base is used to store in corresponding evaluation index, scheme knowledge base and attribute knowledge base is interrelated.Root
Alternative is selected according to the n evaluation index related to product demand.
Single-action function module is received designer by input/output interface and n is evaluated based on what utility theory was inputted
The attribute preference of index, fits n single utility function according to attribute preference, exports corresponding by input/output interface
Function coefficients value extracts alternative on control panel from base module, is made with the evaluation index interval of alternative
For integrating range, and n single-action functional value of each alternative is calculated using single utility function, export and expect to imitate to overall
With module, n is positive integer.
Multi-purpose function module, the property value of the n-1 evaluation index inputted using input/output interface is built without preference
Composite equation group, exports the weight k of each evaluation indexiOnto control panel, i=1~n.
Overall expected utility module is based on weight kiThe overall of each alternative is exported with single-action with functional value to expect to imitate
With value and ranking on control panel.
DSS further comprises reprocessing analysis module, for testing overall expected utility ranking front two
Alternative robustness, Δ k and Δ x, the Δ L for changing the alternative by input/output interface solve institute again
The overall expected utility of the Alternative designs of front two is stated, to judge whether ranking changes, Δ k refers to the change of weight, will
Δ k is added to the k in multi-purpose function module respectivelyi, the new weight of output is calculated again；Δ x refers to single utility function curvature
Change, changes single-action with the function coefficients value being fitted in function module, calculates single-action functional value again accordingly；Δ L refers to standby
The length change for selecting the evaluation index of scheme interval, the evaluation index for changing alternative in single-action function module is interval, phase
The single-action functional value of calculating again answered.
Alternative in base module is increased by input/output interface, changes the ranking of all alternatives；
By input/output interface increase base module in evaluation index, single-action with function module increase a single-action letter
Number, changes the weight k of each evaluation indexiAnd the ranking of all alternatives；Change alternative by input/output interface
The evaluation index of scheme is interval, changes the alternative in single-action with the single-action functional value in function module.
Specific method is as shown in Figure 1：
Step A, STEP A1：Build in advance in the scheme knowledge base of product design, scheme knowledge base and store conventional set
Meter scheme；
STEP A2：Attribute knowledge base is built, evaluation index is stored, knowledge support is provided for design design decision, its
Source includes the attributive character embodied among the data such as experimental data, design manual and actual design process.
STEP A3:Alternative is selected according to the n evaluation index related to product demand.
Step B,
STEP B1:Designer is by answering lottery industry problem (lottery questions) in single-action function module
Input determines the attributes preferred value of 5 different levels to alternative, when being effectiveness U=(0,0.25,0.5,0.75,1) respectively
Property value.
STEP B2:N single utility function is fitted according to attributes preferred value, corresponding function coefficients value is exported；By product
The attribute of alternative is divided into the small attribute of prestige, hopes large attribute and hopes mesh attribute, and the fit procedure of single utility function is provided respectively：
1) large attribute is hoped.The effectiveness that large attribute characterizes the attribute of certain in design is hoped to be passed with the increase dullness of property value
The attribute of increasing.For hoping large attribute, it is first determined the property value of 5 different levels preferences
And x1, by curve matching exponentially type utility function, the utility function type given tacit consent to herein is u (x)：
In formula：aL, bL, cLAnd dLValue correspond to the coefficient of the exponential type effectiveness after curve matching respectively.2) small category is hoped
Property.Hope the effectiveness of certain attribute in small attribute characterization firearm design scheme with the attribute of the increase monotone decreasing of property value.It is right
For small attribute is hoped, it is first determined the property value of 5 different levels preferences And x1, such as
Shown in Fig. 3, pass through curve matching exponentially type utility function u (x)：
In formula：aR, bR, cRAnd dRValue correspond to the coefficient of the exponential type effectiveness after curve matching respectively.
3) mesh attribute is hoped.Mesh attribute is hoped to refer to that the effectiveness of certain attribute reaches or close to the better category of Target Attribute values
Property.For the prestige mesh attribute based on specific objective value, the input of curve matching includes the property value of the right and left.For with
Attribute maximum is turned to for the attribute of target, it is necessary to the attribute on the left of desired value is as input, and for being minimised as with attribute
For the attribute of target, then the attribute on right side is needed as input.The property value of 9 different levels preferences is determined firstAnd x1, pass through curve matching exponentially type effectiveness letter
Number u (x)：
In formula：A, b, c and d value correspond to the exponential type utility function u (x) after curve matching=a+bx+ce respectivelydx
In coefficient, subscript represents left side or right side in desired value, and L represents left, and R represents right.
STEP B3:The same evaluation index for inputting comprehensive all alternatives respectively by input/output interface is interval
Up-and-down boundary, using the interval and single utility function of the evaluation index of each alternative in base module, output is each standby
N single-action functional value of scheme is selected, n is positive integer.
Single attribute utility function of " prestige mesh " property is used, and assumes that specific option priorities value is obeyed and is uniformly distributed, probability is close
Spending function is：
F (x)=1/ (xu-xl) (4)
In formula：xl, xuIt is the interval up-and-down boundary of the same evaluation index of comprehensive all alternatives.
The expected utility of single attribute is solved according to the single utility function (1) (2) (3) and formula (4) of fitting, single attribute is expected
Value of utility calculating formula is as follows：
E (u)=∫ u (x) f (x) dx (5)
In formula：U (x) represents single-action functional expression, and f (x) represents property value probability density function.Substitute into each alternative
Evaluation index interval export n single-action functional value respectively.
Step C：Referred to by inputting the property value of n-1 evaluation index and building without the output evaluation of preference combinations of attributes equation group
Target weight is on control panel；
STEP C1:Weight kiIt is determined by solving the equation group being made up of n linear equation.Effect based on superposition
With functional form and without preference combinations of attributes equation group, multiple linear equations for having identical effectiveness, these equations composition are set up
Equation group be used for solve weight ki。
STEP C2:With multi-purpose combination without attributes preferred composition equation group.It is by designing to have n-1 equation in equation group
Personnel are set up by being combined without preference property value, and remaining one represents that proportionality constant sum is 1,
Shown in specific multi-purpose function building process such as formula (6).
In formula：AndRepresent the value of attribute i different levels.Represent worst level
(its effectiveness is 0)；Representative is removedTwo different levels in addition, they meet(for exampleIt is right
The effectiveness answered is 0.55Effectiveness for 0.45)；It is then the property value specified by policymaker, the property value must make two
The effectiveness of individual different many property values combination is equal so that policymaker to the two combinations without preference, i.e.,
STEP C3：Input n-1 attributeExport weight ki。
Step D：Based on weight kiThe overall expected utility and ranking of each alternative are exported with functional value with single-action
And be stored in multi-purpose function module.
STEP D1：Utilize single attribute expected utility formula (5), calculate overall expected utility so that designer to
The overall expected utility of option can be rapidly converted into after fixed some utility function parameters and option priorities parameter.
In formula：kiRepresent attribute i proportionality constants, E (ui(Ai)) represent attribute i expected utility.
The overall expected utility of each scheme is calculated using formula (8) and the overall expected utility of each scheme is entered
The design decision that row is ordered as scheme provides foundation.Overall expected utility highest scheme is referred to as " most worthy " scheme, always
The scheme that body expected utility is number two is referred to as " second " scheme.
DSS further comprises reprocessing analysis module, for testing overall expected utility ranking front two
Design robustness.Choose the option of ranking front two：" most worthy " scheme and " second " scheme, then test this
The overall expected utility of two options for some design requirements change or scheme innovation caused by scheme, attribute and design
Whether the response that dynamic change occurs for parameter etc. is enough to influence its original ranking.It can be seen that from expected utility calculating process, may
Parameters variation can only occur in two formula, i.e. formula (5) and formula (8).Parameters variation in the two formula is used respectively
Δ k and Δ x, Δ L are represented.Wherein Δ k refers to the change of weight, and Δ k is added to the k in multi-purpose function module respectivelyi, then
It is secondary to calculate the new weight of output；Δ x refers to the change of single utility function curvature, changes single-action with the function system being fitted in function module
Numerical value, calculates single-action functional value again accordingly；Δ L refers to the length change in the evaluation index interval of alternative, changes single
The evaluation index of alternative is interval in utility function module, calculates single-action functional value again accordingly.
The specific implementation method of reprocessing analysis is to increase or decrease Δ k, Δ x and Δ L.Δ k, Δ x and Δ L are given tacit consent to herein
Changing value be 5%, then recalculate the E (u) and E (U) of ranking front two option, and by the two options E (U) change
Situation is shown in the control panel of reprocessing analysis module.
Further, can be according to the parameter of change influence in corresponding module when external environment design changes in demand
Middle processing,
(1) change of amount of projects
Further it is included in after step D and new alternative is added in step A base module and corresponding evaluate refers to
Mark, DSS calculates the overall expected utility of new alternative and by all alternatives again ranking.
(2) change of evaluating indexesto scheme
Further it is included in step A base module after step D and adds new evaluation index, designer is based on effect
With theory in single-action with the attribute bias data inputted in function module to new evaluation index, be fitted according to attribute bias data
Go out corresponding single utility function, export corresponding function coefficients value；The new evaluation for inputting comprehensive all alternatives respectively refers to
Interval up-and-down boundary is marked, the interval and single utility function of the evaluation index of each alternative in base module, output is utilized
All single-action functional values；Recalculate weight and the totality of each alternative is exported based on weight and single-action functional value
Expected utility and ranking.
(3) change of preset parameter
The change of preset parameter refers to that among new decision-making (these parameters belong to fixed to some parameters in former decision-making
Parameter) value changed.For example in new decision-making, because further experiment causes some scheme Attribute Value (such as
The intensity of material) bound changed, become more accurate；Or the importance of a certain attribute (such as wearability) there occurs
Change is, it is necessary to pay the utmost attention to.Now change the interval of evaluation index in step C multi-purpose function module and recalculate list
Utility function value.
Illustrated with reference to example, it is theoretical according to the above-mentioned trade-off decision based on effectiveness, to above-mentioned by taking gun barrel as an example
Method is verified, is devised the probabilistic firearm design program decisions of consideration and is supported system.
Gun barrel is one of most basic part of Rifle blackening, and its main function is to confer to the certain direction of bullet and initial velocity
Degree.Gun barrel design phase of the design in body of a gun scheme, out-of-core techniques design and bullet design process before this is true
Most design parameters are determined.
Table 1 show the Tactical Technical Requirements index of input.Because gun barrel is in shooting, due to gunpowder high temperature, high-pressure fire
The effect of medicine gas, and occur mechanical friction with bullet, in order to realize above target, therefore require that barrel materials have higher resist
Tensile strength (is not less than 50 kilograms/millimeter2), enough impact flexibility (not less than 5 kg-ms/centimetre2), higher yield point
(it is not less than 50 kilograms/millimeter2), good machinability and enough wear-resistant, ablation resistance.At present from heavy machine gun to pistol
Gun barrel typically commonly use 50BA or 50AE steel, antiaircraft machine gun and the heavy machine gun having then use 30SiMnMoVA or 30CrNi2MoVA
Steel, also useful other materials makees gun barrel, but selects most Experience Design for depending on designer on barrel materials.
According to these design requirement targets, six evaluation indexes are filtered out in dependence knowledge base, and know from scheme according to evaluation index
Four alternatives are filtered out in knowledge storehouse and are available for consideration, as shown in table 2.It is quantitative category to have three in the attribute of six evaluation indexes
Property, i.e. tensile strength, yield point and impact value, the other three is quantitative attributes, i.e. processing characteristics, wearability and ablation resistance.
Expectation of the designer to these attributes had both contained " prestige mesh " property such as tensile strength and has expected to reach that desired value is 95 kilograms/milli
Rice2, non-" prestige mesh " property is also contains, such as processing characteristics wishes value, and the higher the better.The all properties value of each option is equal
It is uncertain, is a scope determined by bound, such as option 50BA tensile strength span is 44-69 public
Jin/millimeter2.Assuming that all attribute values are obeyed and are uniformly distributed.Designer needs to select from alternative most expire
The option of sufficient preceding aim.
The Tactical Technical Requirements index that table 1 is inputted
The barrel materials alternative of table 2
As shown in figure 3, illustrating the scheme knowledge base in computer support system.When having new material or a new processing work
Skill needs to be compared when being selected with aid decision person, can quickly be configured on the control panel.
Control panel shown in Fig. 4 is the attribute knowledge base and single-action function computation module figure of each alternative, is had
Support the solution of each attribute list utility function on the panel in effect ground.Control panel shown in Fig. 5 is calculated by multi-attribute-utility value
A series of module, scheme totality block combiners such as expected utility computing module and sensitivity analysis module are formed.
Overall firearm design program decisions supports that system operation procedures and flow are as follows：
Step 1：The design requirement quickly designed according to gun barrel selects evaluation index and filters out alternative.It is in table 2
The scheme selected based on above-mentioned Tactical Technical Requirements index, candidate scheme quantity is n=4.
Step 3：Determine decisionmaker's preference.It is policymaker firearm design side obtained from by answering lottery industry problem in table 3
The attributes preferred value of the different levels of case.
Step 4：The single utility function of fitting.Table 4 is the exponential type utility function that each property value based on table 3 is fitted to attribute
Obtained a, b, c, d coefficient values are exported afterwards.
Step 5：Solve the expected utility of single utility function.Table 5 is to combine the obtained exponential type utility function of table 4 and standby
Select each single-action functional value for the firearm design alternative that each evaluation index interval computation of scheme obtains.
Step 6：Determine weight.Table 6 is the single utility function tried to achieve by the utility function of superposition and without preference combined type
The weight of value is the weight of evaluation index.
Step 7：Solve overall expected utility.Table 7 is the weight of the single attribute expected utility and table 6 that combine table 5, foundation
FormulaThe overall expected utility for each firearm design scheme tried to achieve.So as to obtain
The firearm design scheme of " most worthy " is 30SiMnMoVA, and its Technology for Heating Processing is：870 DEG C of high temperature quench/650 DEG C of high temperature and returned
Fire.
The property value of the different levels effectiveness of table 3
Each attribute single-action function coefficients table of table 4
Each evaluation index expected utility of table 5
The weight of table 6
Each scheme of table 7 totality expected utility
Further, it is necessary to when introducing new alternative, as this is a kind of to be fabricated by using new composite
High heat conduction lightweight gun barrel, wherein for steel inner tube, the epoxy resin of fibre reinforced, this carbon fiber are wound outside it
Enhanced epoxy resin surface plates layer of metal nickel；And steel outer tube is used with high specific strength, the carbon of high ratio modulus mechanical property
Fibre-reinforced composite, it is ensured that light-weighted while effectively improve gun barrel fire accuracy, but its processing characteristics
It is relatively poor.The details of this new six attribute of alternative are as shown in table 8.Due to the introducing of new departure, originally
Four alternative select permeabilities are transformed to five alternative select permeabilities, and designer needs to re-start problem modeling
And make a choice.Due to attribute description, the most decision-making of single attribute utility function and multiattribute utility function etc.
Cheng Jun in original configurable module template instances to achieve, and new decision-making only needs to do one on the basis of original instance
A little necessary modifications.Need the new alternative described in table 8 being added in primary template example, then update scheme
Ranking, specific step is as follows：
1) essential information of specified scheme：Typing information is to instantiate a new alternative in base module；
2) range of attributes of specified scheme：N evaluation index area of newly-increased alternative is inputted by input/output interface
Between and obey distribution.
3) update scheme ranking：After all necessary input information are all set, the ranking of all schemes will automatically more
Newly.The expected utility of new ranking displaying scheme " advanced composite material (ACM) " is 1.635387, is ranked the first.
The attribute information of the advanced composite material (ACM) of table 8
Further, it is necessary to when introducing new evaluation index, it is assumed that a new attribute-elongation is, it is necessary to foregoing five
Accounted in Scheme Choice decision problem.The reason for elongation is introduced into is to be fractureed which characterizes firearm material by pulling force
When, tensile elongation is with the ratio of original length, and the quality for barrel materials is particularly significant.The information related to the attribute configuration point
For four aspects, as shown in table 9：
(1) new evaluation index is added in base module；
(2) attribute preference of the different levels of elongation utility function needs, including on the left of desired value and right side are set up
Preference；
(3) multiattribute utility function weight k is determinediThe input information needed, i.e., the n-1 in a lot of property value combinations of unbiased
The value of individual evaluation index；
The span of (4) five alternative elongations.Introduce after new attribute, the transformation of six Attributions selection decision problems
Into seven Attributions selection decision problems, designer needs in consideration table 9 information of elongation and problem is modeled again
To make new decision-making.Most of knowledge can be reused in conventional example, make some necessary modifications.Need the modification made
It is one new attribute of instantiation and ranking again is carried out to five alternatives, comprises the following steps that：
1) specified attribute essential information：The relevant information of typing elongation is with reality in the attribute knowledge base of base module
One new attribute of exampleization；
2) attribute preference of the input to the different levels of elongation；
3) multiattribute utility function right assessment relevant information is configured：Specify without each attribute in preference combinations of attributes equation group
Property value.
4) span of each scheme elongation is configured：The bound of comprehensive all alternative elongations is general as solving
The border of rate density function.
5) recalculate weight and the overall expected utility of each alternative is exported based on weight and single-action functional value
Value and ranking：New ranking shows that the scheme for coming front two is respectively " advanced composite material (ACM) " and 30SiMnMoVA, corresponding
Expected utility is respectively 1.87532 and 1.78562.
The configuration information of the attribute elongation of table 9
DSS further comprises reprocessing analysis module, it is considered to which some possible Parameters variations simultaneously test ranking
Robustness of first scheme " advanced composite material (ACM) " in the case of Parameters variation, selects it to carry out firearms to strengthen designer
The confidence of design.The scheme " advanced composite material (ACM) " and " second " scheme 30SiMnMoVA expected utility ranked the first connects very much
Closely, respectively 1.87532 and 1.78562, both only poor 0.0897.It is by both rankings are tested in reprocessing analysis module
No to change because of the change of some parameters, i.e., whether ranking is sensitive to Parameters variation.
The present invention will illustrate the process of reprocessing analysis by taking Δ L as an example.The tensile strength of scheme " advanced composite material (ACM) " is former
The span of beginning is 90-100 kilograms/millimeter2, by progressively reduce this scope (progressively reducing design parameter ambiguity) come
Response and the program and the scheme 30SiMnMoVA ranking contrast situation of the overall expected utility of the program are tested, such as Fig. 6 institutes
Show.Intensity of variation is arranged to 5% in figure, and change frequency is arranged to 7 times.It was found from the image automatically generated：When tension is strong
When the span of degree narrows down to original 80%, scheme 30SiMnMoVA expected utility is risen by original 1.78562
To 1.82782；Even if hereafter the span of tensile strength is continued to zoom out, scheme 30SiMnMoVA expected utility is still tieed up
Hold 1.82782 level it is as shown in Figure 6.It follows that：Scheme ranking reduces simultaneously to tensile strength span
Insensitive, it is safe to select the scheme " advanced composite material (ACM) " ranked the first.
In summary, it can obtain, material proposition is wanted from Fig. 6 in the case of based on this Tactical Technical Requirements
Ask, the scheme of " most worthy " is " advanced composite material (ACM) ".Based on knowable to the fact that above-mentioned, designer is selecting the material
On the basis of carry out follow-up design work, it is possible to reduce with reference to the time of a large amount of knowledge, save substantial amounts of design iteration process, and
Strengthen the validity of decision-making.All in all, it is considered to which probabilistic firearm design program decisions supports system to provide designers with
A kind of relatively flexible Design Method.Designer can select according to Tactical Technical Requirements and designer itself preference
Select the design of " most worthy ".The characteristics of for firearm design process, introduce configurable module template and cause module more
It is easy to modification, while also reducing data redundancy.
In summary, presently preferred embodiments of the present invention is these are only, is not intended to limit the scope of the present invention.
Within the spirit and principles of the invention, any modification, equivalent substitution and improvements made etc., should be included in the present invention's
Within protection domain.
Claims (6)
1. a kind of product Design Decision Making based on effectiveness supports system, it is characterised in that the DSS includes knowledge
Library module, single-action function module, multi-purpose function module and overall expected utility module, each module possess independent
Input/output interface；
The base module is used to store conventional design and evaluation index, is commented according to the n related to product demand
Valency index selects alternative；
The single-action function module is received designer by input/output interface and n is evaluated based on what utility theory was inputted
The attributes preferred value of index, fits n single utility function according to attributes preferred value, exports corresponding by input/output interface
Function coefficients value extracts alternative on control panel from base module, is made with the evaluation index interval of alternative
For integrating range, and n single-action functional value of each alternative is calculated using single utility function, export and expect to imitate to overall
With module, n is positive integer；
The multi-purpose function module, the property value of the n-1 evaluation index inputted using input/output interface is built without preference
Combinations of attributes equation group, exports the weight k of each evaluation indexiOnto control panel, i=1~n；
The overall expected utility module is based on weight kiThe overall expected utility of each alternative is exported with functional value with single-action
Value and ranking are on control panel；
The DSS further comprises reprocessing analysis module, for testing overall expected utility ranking front two
Alternative robustness, Δ k and Δ x, the Δ L for changing the alternative by input/output interface solve institute again
The overall expected utility of the Alternative designs of front two is stated, to judge whether ranking changes, Δ k refers to the change of weight, will
Δ k is added to the k in multi-purpose function module respectivelyi, the new weight of output is calculated again；Δ x refers to single utility function curvature
Change, changes single-action with the function coefficients value being fitted in function module, calculates single-action functional value again accordingly；Δ L refers to standby
The length change for selecting the evaluation index of scheme interval, the evaluation index for changing alternative in single-action function module is interval, phase
The single-action functional value of calculating again answered；
Alternative in base module is increased by input/output interface, changes the ranking of all alternatives；Pass through
Evaluation index in input/output interface increase base module, increases a single utility function in single-action function module,
Change the weight k of each evaluation indexiAnd the ranking of all alternatives；Alternative is changed by input/output interface
Evaluation index it is interval, change the alternative in single-action with the single-action functional value in function module.
2. a kind of product Design Decision Making based on effectiveness supports method, it is characterised in that uses and is based on as claimed in claim 1
The product Design Decision Making of effectiveness supports system, and the decision support method is：
Step one, according to the n evaluation index selection alternative related to product demand in base module；
Step 2, designer is based on utility theory and n is commented by input/output interface input in single-action function module
The attributes preferred value of valency index, fits n single utility function according to attributes preferred value, exports corresponding function coefficients value to control
On panel processed；
Step 3, extracts alternative from base module, and integrating range, profit are used as so that the evaluation index of alternative is interval
N single-action functional value of each alternative is calculated and exported with single utility function, and n is positive integer；And it is defeated by inputting
The property value that outgoing interface inputs n-1 evaluation index is built without preference combinations of attributes equation group, exports each evaluation index
Weight kiOnto control panel, i=1~n；
Step 4, based on weight kiThe overall expected utility and ranking of each alternative are exported to controlling with functional value with single-action
On panel processed；
Step 5, changes Δ k, Δ x or the Δ L of ranking front two alternative, the overall phase of the alternative is solved again
Value of utility and ranking are hoped, to judge whether the ranking of the front two alternative changes.
3. the product Design Decision Making based on effectiveness supports method as claimed in claim 2, it is characterised in that need to introduce newly
Further it is included in during alternative, after step 4 and new alternative is added in the base module described in step one and corresponding
Evaluation index, DSS calculates the overall expected utility of new alternative and by all alternative weights
New ranking.
4. the product Design Decision Making based on effectiveness supports method as claimed in claim 2, it is characterised in that need to introduce newly
Further it is included in the base module described in step one during evaluation index, after step 4 and adds new evaluation index, designs
Personnel based on utility theory in single-action with the attribute preference inputted in function module to new evaluation index, according to attribute bias
Value fits corresponding single utility function, exports corresponding function coefficients value；The new of comprehensive all alternatives is inputted respectively
The interval up-and-down boundary of evaluation index, evaluation index interval and single-action letter using each alternative in base module
Number, exports all single-action functional values；Recalculate weight and based on weight and each alternative side of single-action functional value output
The overall expected utility and ranking of case.
5. the product Design Decision Making based on effectiveness supports method as claimed in claim 2, it is characterised in that when alternative
When evaluation index interval changes, change the interval of evaluation index in the multi-purpose function module of step 3 and recalculate single-action
Use functional value.
6. the product Design Decision Making based on effectiveness supports method as claimed in claim 2, it is characterised in that in the step 2
The attribute bias data of n evaluation index are obeyed and are uniformly distributed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710429349.0A CN107292007A (en) | 2017-06-08 | 2017-06-08 | A kind of product Design Decision Making based on effectiveness supports system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710429349.0A CN107292007A (en) | 2017-06-08 | 2017-06-08 | A kind of product Design Decision Making based on effectiveness supports system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107292007A true CN107292007A (en) | 2017-10-24 |
Family
ID=60096280
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710429349.0A Pending CN107292007A (en) | 2017-06-08 | 2017-06-08 | A kind of product Design Decision Making based on effectiveness supports system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107292007A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126761A (en) * | 2019-11-20 | 2020-05-08 | 中国辐射防护研究院 | Method and system for estimating capacity of regional radiation environment |
| CN111680337A (en) * | 2020-06-04 | 2020-09-18 | 宁波浙大联科科技有限公司 | PDM system product design requirement information acquisition method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5734890A (en) * | 1994-09-12 | 1998-03-31 | Gartner Group | System and method for analyzing procurement decisions and customer satisfaction |
| CN104112035A (en) * | 2014-05-08 | 2014-10-22 | 浙江理工大学 | Effectiveness and fuzzy theory based collaborative decision-making method for product design |
-
2017
- 2017-06-08 CN CN201710429349.0A patent/CN107292007A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5734890A (en) * | 1994-09-12 | 1998-03-31 | Gartner Group | System and method for analyzing procurement decisions and customer satisfaction |
| CN104112035A (en) * | 2014-05-08 | 2014-10-22 | 浙江理工大学 | Effectiveness and fuzzy theory based collaborative decision-making method for product design |
Non-Patent Citations (1)
| Title |
|---|
| 张芳燕 等: "基于效用理论的产品设计决策方法及实例研究", 《机械设计》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126761A (en) * | 2019-11-20 | 2020-05-08 | 中国辐射防护研究院 | Method and system for estimating capacity of regional radiation environment |
| CN111680337A (en) * | 2020-06-04 | 2020-09-18 | 宁波浙大联科科技有限公司 | PDM system product design requirement information acquisition method and system |
| CN111680337B (en) * | 2020-06-04 | 2021-07-06 | 宁波智讯联科科技有限公司 | PDM system product design requirement information acquisition method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | A comparative review between Genetic Algorithm use in composite optimisation and the state-of-the-art in evolutionary computation | |
| Mousavi-Nasab et al. | A comprehensive MCDM-based approach using TOPSIS, COPRAS and DEA as an auxiliary tool for material selection problems | |
| CN107844835A (en) | Multiple-objection optimization improved adaptive GA-IAGA based on changeable weight M TOPSIS multiple attribute decision making (MADM)s | |
| CN105740467B (en) | A kind of method for digging of C-Mn steel industry big data | |
| Qin et al. | Harmony search algorithm and related variants: A systematic review | |
| CN110008553B (en) | Product design scheme optimization method based on life cycle cost and environmental influence | |
| EP1406195A1 (en) | Method and apparatus for evaluating a proposed solution to a constraint problem | |
| CN107832983B (en) | Casting smelting batch planning and scheduling method based on dynamic programming and genetic algorithm | |
| JPH0652176A (en) | Optimizing method and process plan generation device using the method | |
| CN110751355A (en) | Scientific and technological achievement assessment method and device | |
| CN109657411A (en) | A kind of solvent deasphalting unit modeling and optimization method based on data-driven | |
| CN113330440B (en) | Design assistance method and design assistance device for metal material | |
| CN107194609A (en) | A kind of product design system and method | |
| CN114004065A (en) | Transformer substation engineering multi-objective optimization method based on intelligent algorithm and environmental constraints | |
| Protalinskiy et al. | Strategic decision support in the process of manufacturing systems management | |
| Jing et al. | An integrated product conceptual scheme decision approach based on Shapley value method and fuzzy logic for economic-technical objectives trade-off under uncertainty | |
| CN109491709B (en) | Software code controllability comprehensive evaluation method based on AHP and neural network | |
| CN107292007A (en) | A kind of product Design Decision Making based on effectiveness supports system and method | |
| CN106773674A (en) | A kind of delayed coking multiple target real-time optimization method and system | |
| Chen et al. | Combining fuzzy iteration model with dynamic programming to solve multiobjective multistage decision making problems | |
| Zhou et al. | Supplier’s goal setting considering sustainability: An uncertain dynamic Data Envelopment Analysis based benchmarking model | |
| AU2003206912A1 (en) | Method and system for the automatic planning of experiments | |
| CN110728031B (en) | Multi-objective optimization method for balancing complex petrochemical process production energy based on ANN modeling | |
| Kim et al. | Stepwise benchmarking based on production function: Selecting path towards closest target | |
| Chen et al. | Selecting a weapon system using zero-one goal programming and analytic network process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171024 |