EVOLVING BUILDING BLOCKS FOR DESIGN USING GENETIC ENGINEERING: A FORMAL APPROACH.
JOHN S.GERO AND VLADIMIR A.KAZAKOV
Key Centre of Design Computing,
Department of Architectural and Design Science,
The University of Sydney,NSW2006Australia.
e-mail:john,kaz@https://www.doczj.com/doc/491107746.html,.au
Abstract.This paper presents a formal approach to the evolution of a representation for
use in a design process.The approach adopted is based on concepts associated with genetic engineering.An initial set of genes representing elementary building blocks is evolved
into a set of complex genes representing targeted building blocks.These targeted build-
ing blocks have been evolved because they are more likely to produce designs which ex-
hibit desired characteristics than the commencing elementary building blocks.The tar-
geted building blocks can then be used in a design process.The paper presents a formal evolutionary model of design representations based on genetic algorithms and uses pattern recognition techniques to execute aspects of the genetic engineering.The paper describes
how the state space of possible designs changes over time and illustrates the model with
an example from the domain of two-dimensional layouts.It concludes with a discussion
of style in design.
1.Introduction
There is an increasing understanding of the role that a design language and its rep-resentation play in the ef?ciency and ef?cacy of any design process which uses
that language(Coyne et al.,1990;Gero et al.,1994).A recurring issue is what
is the appropriate granularity of a language.If building blocks which constitute
the elements of a design map onto a design language then the question becomes
what is an appropriate scale for those building blocks in terms of the?nal design.
At one extreme we have parameterised representations where the structure of a
design is?xed,all the variables which go to de?ne a design are prede?ned and
what is left is to determine the values of those variables.This de?nes a very small
design space,small in terms of all the possible designs which might be able to be produced for that design situation.At the other extreme we have elementary build-
ing blocks which can be combined in a very large variety of ways and which,as a
2John S.Gero AND Vladimir A.
Kazakov
S e
Figure 1.
is the design space produced by all the possible combinations of the elementary build-ing blocks,
is the design space produced by all the combinations of the values of the parameterised variables,is the design space of interesting designs for the design situation.
consequence de?ne a very large design space,the vast part of which covers designs which are likely to be uninteresting in terms of the current design situation.The designs produced by the parameterised design representations are a subset of those capable of being produced by the elementary building block representation,Fig-ure 1.Examples of building block representations include constructive systems such as design grammars as exempli?ed by shape grammars (Stiny,1980b).Ex-amples of parameterised variable representations include a wide variety of design optimization formulations (Gero,1985).
The advantage of the use of the elementary building blocks representation is the coverage of the entire design space they provide,whereas the advantage of the parameterised variable representation is the ef?ciency with which a solution can be reached.
We present here a formal approach which generates a targeted representation of a design problem.A targeted representation is the one which closely maps on to the problem at hand.As an example consider a layout planning problem in archi-tectural design.One representation may be at the material molecular level,where molecules can be combined to make a variety of materials and particular combina-tions in space produce physical objects;here the potential solution space includes designs which bear no relations to architecture.A targeted representations may be to represent rooms such that the potential solution space primarily includes designs which are all recognizably architectural layouts.
In order to simplify our analysis we consider designs which are assembled from
Evolving Building Blocks for Design Using Genetic Engineering3
Figure2.The set of building blocks for Froebel’s kindergarten gifts(Stiny,1980b). some?nite collection of spatial elements(we call them building blocks or compon-ents)along with assembly rules.It is assumed that the assembly rules do not affect the components-the design object is a union of non-overlapping building blocks. We start with some set of building blocks which we call elementary components.It is assumed that they cannot be decomposed into any smaller ones.We call a set of building components and assembly rules a representation of the design problem and the set of elementary components and corresponding rules the basic represent-ation.We call it a representation because it implicitly de?nes the set of all designs (design state space)which can be produced using this set of building blocks and assembly rules.
The kindergarten gifts of Froebel(Stiny,1980b)is a typical example of such types of design problem.One of many possible elementary representations and as-sembly rules for it is shown in Figures2and3.One can easily extend it by adding further elementary building blocks and/or further assembly rules.
Targeted representations
A great variety of designs can be produced within a basic https://www.doczj.com/doc/491107746.html,ually the designer is interested in some particular class of designs.Assume we have some additional set of composite building blocks and an additional set of assembly rules to handle them.We can calculate the number of these composite building blocks which can be found in all possible designs in that particular class and the number of elementary building blocks used to build the rest of these designs(each elementary building block should be counted only once as a member of composite building or elementary building block,the largest composite blocks are counted?rst and the elementary blocks are counted last).Then we can calculate the frequency of usage of these composite building blocks and elementary building blocks in the entire design space.The same values can be calculated for all designs which have the property or properties we are interested in.If the frequency of the usage of the composite building blocks is much higher for the designs of interest than for all designs built from the elementary building block and the frequency of elementary components usage is much lower than that of the composite building blocks for the design space of interest then we can use the composite building blocks instead
4John S.Gero AND Vladimir A.Kazakov
Figure3.The set of six assembly rules for Froebel’s kindergarten gifts.
of elementary one to produce designs of interest with much higher probability.In other words a representation exists which maps into the area of interest of the entire design space.Let us call it the targeted representation for the particular class of designs.Obviously different targeted representations can be produced which cor-respond to different sets of composite building blocks.We characterize these rep-resentations by their“complexity”which is de?ned recursively as:0-complexity for the basic representation,1-complexity for the representation whose building blocks are assembled from elementary building blocks,2-complexity for the rep-resentation whose building blocks are assembled from the building blocks of0-complexity and1-complexity,etc.Assume an evolution occurs in an abstract space of complex representations:initially only elementary building blocks exist then a cycle proceeds when a new set of composite building blocks is produced from the ones which are currently available.Then a representation of i-complexity(and building blocks of i-complexity)simply means that composite building blocks of this representation have been produced during-th step of this evolution.
Different composite building blocks of the same-complexity may contain dif-ferent numbers of elementary building blocks:for example,assume some build-ing block of3-complexity contains3elementary building blocks and one of the composite building blocks of4-complexity is assembled from2building blocks of 3-complexity and thus contains6elementary components and another one is as-sembled from one block of3-complexity and one block of0-complexity and thus contains4elementary components.It is also clear that because there are different ways to assemble the same composite building block it may be produced multiple
Evolving Building Blocks for Design Using Genetic Engineering5
(a)
(b)(c)
Figure4.The set of composite building blocks of different complexity for building a staircase;(a) 1-complexity,(b)and(c)2-complexity.
times in representations of different complexity level during the evolution.
The search for a reasonably good design using the basic representation is very dif?cult because signi?cant part of the search effort is wasted in the search of un-useful parts of the design space.If the targeted representation is used instead of ele-mentary one the probability of producing designs of interest becomes much higher, the design space becomes smaller and the design problem less complicated and easier to deal with.The approach presented in this article automatically generates the hierarchy of more and more complex building blocks(in general);ones which are more and more close to the targeted representations which are capable of pro-ducing better and better designs.
Assume we work with the representation of the kindergarten blocks shown in Figures2and3and are trying to design a two-level building with walking ac-cess from one?oor to the next.The search for a design with this property is quite dif?cult because only a very small fraction of all feasible objects exhibits it and the probability of discovering the combination of building blocks which makes a staircase during the search is low.However,if we add a composite object of1-complexity(Figure4)and corresponding assembly rules Figure5to the repres-entation we increase this probability,and if we add a composite building block with 2-complexity(Figure4)then this probability increases further.
Genetic engineering
Genetic engineering,as used in this paper,is derived from genetic engineering no-tions related to human intervention in the genetics of natural organisms.In the ge-netics of natural organisms we distinguish three classes:the genes which go to make the genotype,the phenotype which is the organic expression of genotype, and the?tness of the phenotype in its environment.When there is a unique identi-?able?tness which is performing particularly well or particularly badly amongst all the?tness of interest we can hypothesize that there is a unique cause for it and
6John S.Gero AND Vladimir A.Kazakov
Figure5.The set of additional assembly rules for handling composite building blocks. that this unique cause can be directly related to the organism’s genes which ap-pear in a structured form in its genotype.Genetic engineering in concerned with locating those genes which produce the?tness under consideration and in modify-ing those genes in some appropriate manner.This is normally done in a stochastic process where we concentrate on populations rather than on individuals.
Organisms which perform well(or badly)in the?tness of interest are segreg-ated from these organisms which do not exhibit that?tness or do so only in a min-imal sense.This bifurcates the population into two groups.The genotypes of the former organisms are analysed to determine whether they exhibit common char-acteristics which are not exhibited by the organisms in the latter group(Figure6). If they are disjunctive,these genes are isolated on the basis that they are respons-ible for the performance of the?tness of interest.In natural genetic engineering these isolated genes are either the putative cause of positive or negative?tness.If negative then they are substituted for by“good”genes which do not generate the negative?tness.If they are associated with positive?tness they are reused in other organisms.It is this later purpose which maps on to our area of interest.
One can interpret the problem of?nding the targeted set of building blocks as an analog of the genetic engineering problem:?nding the particular combin-ations of genes(representing elementary building blocks)in genotypes which are responsible for the properties of interest of the designs and regular usage of these gene clusters to produce designs with desired features.
7
Evolving Building Blocks for Design Using Genetic Engineering
Figure6.The genotypes of the“good”members of population all exhibit gene combinations,X, which are not exhibited by the genotypes of the“bad”members.These gene combinations are the ones of interest in genetic engineering.
2.Building blocks
Thus,we establish that different building blocks de?ne different design state spaces (which are,in their turn,the subsets of the entire basic design space).More form-ally we assume that for the design space of interest a set of composite building blocks exists which is suf?cient to build any design of interest from it(or which are suf?cient to build a signi?cant part of any of these designs of interest).
We search for these building blocks using the consequence of the assumption made in the introduction about frequencies of composite components usage:on av-erage the sampling set of designs with the desired characteristics(the“good”ones) contains more of such composite building blocks than the sampling set of designs that do not have these characteristics(the“bad”ones).In some cases it is even true in a deterministic sense-that only the designs which can be built completely from some set of composite building blocks possess the objective characteristics,all the rest of the entire basic state space does not have them.One can easily come up with corresponding examples.
In the next section we describe an evolutionary algorithm which generates“good”and“bad”sampling sets using the current set of building block(set of elementary block at the beginning)and use genetic engineering concepts to determine new composite blocks which are closer to the“targeted”ones than the current set of building blocks.These two steps proceed in cycle while the“good”sampling set converges to the sampling set from the desired design state space and the set of
8John S.Gero AND Vladimir A.
Kazakov
a
b
rule 8
Figure7.The assembly(transformation)rules used in the example.
complex building blocks comes closer and closer to the targeted set.
If the basic assumption about more frequent use of some composite building blocks to generate the particular class of designs is not true for some problem then
the targeted representation for this problem does not exist and the algorithm which
is proposed below will not generate an improved representation but will be equi-valent to the algorithm for solving the routine design problem(Gero and Kazakov, 1995)and will simply generate the improved designs.
If the sequence of assembly actions is coded as a real vector then the problem
of?nding the complex building blocks becomes the problem of?nding the key patterns in the coding vector-the combinations of codes within it which are likely
to be associated with the property of interest in the designs.The vast arsenal of pattern recognition methods can be used to solve this problem.Essentially they are just search methods for subsets in a coding sequence which on average are more frequently observed in objects with desired characteristics than in the rest of the population.
Let us illustrate the execution of the cycle just outlined using a simple2-dimensional graphical example.We will describe it in more detail later but for now on it is suf-
?cient to say that there is only one elementary block here-the square and that a design is assembled from cubes using the8rules shown in Figure7.Any design can be coded as a sequence of these rules used to assemble it.Assume we are trying
to produce a design which has the maximum number of holes in it and that each design contains not more than20squares.We start the cycle by generating a set of coding sequences and corresponding designs Figure8.Then we notice that a num-
ber(4)of the designs have the maximal number of holes(designs1,2,4,and7-the “good”sampling set)contain the composite building block and that for three of them their coding sequences contain the pattern.We also notice that only
a few(none in this case)of the designs without holes(designs3,5,8and10-the “bad”sampling set)contain this block and none contain this pattern in their coding sequence.Then we can generate the next population of coding sequences using the identi?ed sequence as a new rule which uses the composite building block
in the design.Assuming that we employ some optimization method to generate this new population we can expect that the“good”sampling set from the new pop-
Evolving Building Blocks for Design Using Genetic Engineering
9
design 2good design 3{3,2,2,6,5,8,2,1,4,4,3,1}{2,3,2,3,4,3,5,6,5,1,6,2}design 6
neutral {6,4,1,2,3,4,5,2,1,7,4}
design 9
Figure 8.The identi?cation of the pattern
and corresponding composite building block
in the genotypes of “good”designs.ulation is better than the previous one (that is,the designs which belong to it have on average more holes than the ones from the previous “good”sampling set).Then we again try to identify the patterns which are more likely to be found in designs from this “good”sampling set than from the “bad”one.This time these patterns may contain the previously identi?ed patters as a component.Then we generate a new population of designs using these additional pattern sequences of rules as an additional assembly rule and so on.
The sizes of the sampling sets in realistic systems is likely to be much larger
10John S.Gero AND Vladimir A.Kazakov
than the ones in this example and much more sophisticated techniques(Pearson and Miller,1992)should be employed to single out these key patterns.
3.Evolving building blocks
For a more formal analysis of the evolution of the building blocks we use the shape grammar formalism(Stiny,1980a).It consists of an ordered set of initial shapes and an ordered set of shape transformation rules which are applied recursively.A particular design within the given grammar is completely de?ned by a control vector which de?nes the initial shape and transformation rules applied at each
stage of recursive shape generation.According to the discussion in the Introduc-tion we consider a particular class of shape grammar similar to the kindergarten grammar(Stiny,1980b),where any shape is a non-overlapping union of building blocks and feasible shape transformations are addition,replacement or deletion of the building blocks.
Let be a set of currently available building blocks,and
be a set of assembly rules applicable to these blocks. Then the control vector,,,, de?nes the population of designs,.
The length of the control vector,is a variable.
If we add new complex building block
and new assembly rules
for its handling then we get a new extended set of rules
,,and. Now we can produce the design which corresponds to the vector whose components belong to the extended and.Note that the additional building blocks and assembly rules are generated recursively:they are completely de?ned in terms of the previous and.
We assume that the design problem has a quanti?able objective vector-function ,and can be formulated as optimization problem
(1)
The problem(1)over the representation with a?xed set of building compon-ents and assemble rules can be solved using any of optimization methods(Gero and Kazakov,1995)but the stochastic algorithms like genetic algorithms(Hol-land,1975)and simulated annealing(Kirkpatrick et al.,1983)look most prom-ising at the moment.We have chosen the genetic algorithm.
The evolutionary method has the following structure:
Algorithm
(0).Initialization.Set counter of iteration.Take the set of elementary build-ing blocks and corresponding assembly rules.Generate some
Evolving Building Blocks for Design Using Genetic Engineering11 random population of,calculate and.Set the relative thresholds for the design’s ranking;they are used during an evolution stage to divide the design into“good”,“bad”and“neutral”sampling sets,that is, the parts of population which exhibit()best,()worse and intermediate rel-ative?tness level.
(1)Evolution of complex building blocks.For every component of the objective function divide the population into3groups:
“good”(,“bad”(, and“neutral”(the rest of population).
Determine combinations,of the current building blocks which distinguish the“good”sampling set from the“bad”one statistically signi?cantly using any one of the pattern recognition algorithms. Add it to the current set of building blocks.Add corres-ponding new assembly rules to.
(2)Generation of new https://www.doczj.com/doc/491107746.html,pute new population using available in-formation about current population.The com-ponents of belong to the new extended and.The depends on the op-timization method employed.If the genetic algorithm has been chosen then
is to be calculated using standard crossover and mutation operations.Because the updated grammar includes the grammar from the previous generation the search method guarantees that the new population is better than the previous one(at least no worse)and the new“good”sampling set is closer to sampling set of the design state space of interest.
(3)Repeat steps(1)and(2)until the stop conditions are met.
The stop conditions usually are the termination or slowing down of the im-provement in and/or the end of new building blocks generation.
4.Example
Evolving the targeted representation
As an example we take the problem of the generation of a2-dimensional block design on a uniform planar grid(derived from(Gero and Kazakov,1995)).There is just one elementary component here-a square and the eight assembly rules(trans-formation rules in terms of a shape grammar)which are shown in Figure7.If the position where the current assembly rule tries to place the next square is already taken then all the squares along this direction are shifted to allow the placement of new square.It is assumed that the transformation rule at the-th assembling stage is applied to the elementary block added during the-th stage.The characterist-ics of interest are geometric properties of the generated design.In order to demon-strate the idea,assume that the generated design can not consist of more than32
12
John S.Gero AND Vladimir A.Kazakov 0.10.2
0.3
0.40.50.60.70.8
0.9020406080100120140160180
T O T A L F R A C T I O N O F C O M P L E X G E N E S GENERATIONS Figure 9.The fraction of composite building blocks in the total pool of building blocks used to assemble the population vs.generation number.The objective function has two components:the area of closed holes and the number of connections between holes and the outside space.The initial set of building blocks contains only elementary building blocks.Evolution proceeds until it naturally dies off.
elementary components.We generate a new population during the stage (2)of the Algorithm using the modi?cation of the simple genetic algorithm tailored to handle multidimensional objective functions (Gero and Kazakov,1995).We implement a very simple pattern recognition algorithm based on the statistical frequency ana-lyses of double and triple element building blocks with a high cut-off threshold for the acceptance of the patterns.For more complex systems more sophisticated technique is needed.
During the ?rst iteration we begin with the set of building blocks which con-tains only the elementary ones and search for the designs with maximal area of en-closed holes and maximal number of connections between the holes and outside space.The evolution was allowed to proceed until a stable condition was reached.The result are shown in Figures 9and 10.By plotting the fraction of the complex building blocks in the total pool of building blocks used to assemble the popula-tion at different generations Figure 9,one can see how complex building blocks become dominant and how its fraction reaches a stable level after 110-120itera-tions.The fractions of building blocks of different complexity in the total pool at different generation are shown in Figure 10.One can see that during the ?rst 40generations the total fraction of composite building blocks arises monotonically.For the ?rst 10generations this rise is completely provided by the increasing num-ber of 1-complexity composite building blocks in the population.Then (from 15to 30generations)the fraction of 1-complexity blocks remains stable but the num-ber of 2-complexity building blocks increases and provides the continuing increase
Evolving Building Blocks for Design Using Genetic Engineering
1300.1
0.2
0.30.40.50.60.7
0.8
020406080100120140160180
F R A C T I O N O F C O M P L E X
G E N E S GENERATIONS 1-COMPLEXITY 2-COMPLEXITY 3-COMPLEXITY 4-COMPLEXITY 5-COMPLEXITY 6-COMPLEXITY 7-COMPLEXITY 8-COMPLEXITY Figure 10.The fraction of composite building blocks with different complexities in the total pool of the building blocks used to assemble the population vs.generation number.This ?gure shows the building blocks of different complexities which are summed to produce the total fraction shown in Figure 9.
in the total fraction of composite building blocks.From generations 40to 70this total fraction is stable with approximately half of building blocks of 1-complexity and half of 2,3and 4-complexities.Then the number of 1-complexity blocks and total number of complex blocks declines sharply and from 70until 110generation a transitional process occurs with a complex redistribution of populations between representations with different complexities.At the end of this period the building blocks of 8-complexity saturate the population when the fractions of the other com-plex building blocks are shifted towards a noise level only.One of the evolution paths in the space of complex building blocks is shown in Figure 11(a).Some of the designs produced are shown in Figure 11(b).Here arrows show which pre-viously evolved composite building blocks are used to assemble the new building block.The 0-complexity block and its contributions are omitted.As we already noted composite blocks of the same complexity level sometimes have different numbers of elementary components.Coincidently,the 5-complexity block is re-produced again in the representations of 6-,7-and 8-complexities and is one of the dominant blocks at the end of the evolutionary process.
Using targeted representation.
The set of targeted building blocks evolved during this process is then used as an initial set of building blocks during the second experiment when we produce the designs with maximal total area of holes inside and maximal number of connec-tions between these holes inside the structure.Here the ?tnesses are close to but
14John S.Gero AND Vladimir A.Kazakov
Figure11.(a)An example of the evolutionary paths in the evolution of a complex building block, (b)some of the designs produced using the set of evolved complex blocks.
Evolving Building Blocks for Design Using Genetic Engineering
1500.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
020406080100120140160180
T O T A L F R A C T I O N O F C O M P L E X G E N E S GENERATIONS Figure 12.The fraction of composite building block in the total pool of building block used to as-semble the population vs.generation number.In this experiment the objective function is the num-ber of closed holes and the number of connection between the closed holes inside the structure.The initial set of building blocks is inherited from the ?rst experiment and is the targeted representation.not the same as those used to evolve the targeted representation.This experiment is used to test whether the targeted representation is likely to be used more than the original,elementary building blocks.If the targeted representation is used rather than the elementary building blocks then we have achieved our goal of evolving a representation can be used to produce designs which exhibit desired characterist-ics more readily.The results are shown in Figures 12and 13.One can see that the fraction of the composite building blocks used to produce these designs reaches the saturation level during the ?rst few iterations.The visible redistributions of the population between the composite building blocks of 5,6and 7-complexities are purely super?cial -this redistribution occurs between the same composite building blocks which are present in all these representations.Evolution of the representa-tion does not occur during this experiment -no new complex building block were evolved.This can be interpreted as an indication of closeness of the targeted rep-resentations for both problems.So if the targeted representation is evolved for one set of objectives then it can be usefully applied to any of the objective sets which are only slightly different to it.
Effects of incomplete evolution
In this experiment we repeat the ?rst iteration but stop the evolution prematurely after only 60generations.After this we repeat the second iteration using the evolved incomplete set of composite building blocks.The results are shown in Figures 14
16
John S.Gero AND Vladimir A.Kazakov 00.1
0.2
0.3
0.40.50.60.7
0.8
0.9020406080100120140160180
F R A C T I O N O F C O M P L E X
G E N E S GENERATIONS 1-COMPLEXITY 2-COMPLEXITY 3-COMPLEXITY 4-COMPLEXITY 5-COMPLEXITY 6-COMPLEXITY 7-COMPLEXITY 8-COMPLEXITY Figure 13.The fraction of composite building blocks with different complexities in the total pool of the building block used to assemble the population vs.generation number in the experiment.
0.10.2
0.3
0.40.50.60.70.80.9
1
0102030405060
T O T A L F R A C T I O N O F C O M P L E X G E N E S GENERATIONS Figure 14.The fraction of composite building block in the total pool of building block used to as-semble the population vs.generation number.In this experiment the objective function is the number of closed holes and the number of connections between the closed holes inside the structure.The ini-tial set of building blocks is inherited from ?rst iteration which has been prematurely terminated at generation 60.
and 15.In this case the evolution of the representation continues for about a fur-ther 10generations and we end up with the same set of evolved composite building blocks.The saturation of the population with the composite building blocks is also completed after these 10generations.Thus,one can start to evolve a representa-
Evolving Building Blocks for Design Using Genetic Engineering
1700.1
0.2
0.3
0.40.50.60.7
0.8
0.90102030405060
F R A C T I O N O F C O M P L E X
G E N E S GENERATIONS 1-COMPLEXITY 2-COMPLEXITY 3-COMPLEXITY 4-COMPLEXITY 5-COMPLEXITY 6-COMPLEXITY 7-COMPLEXITY 8-COMPLEXITY Figure 15.The fraction of composite building blocks with different complexities in the total pool of the building block used to assemble the population vs.generation number in the third experiment.tion for one set of objectives and then continue it for another closely related set of objectives.
If we commence by treating the problem as one of ?nding improved designs then from a computational viewpoint this form of evolution speeds up the conver-
gence to improved designs by up to
(in terms of the number of generations required)when compared with standard genetic algorithms.It appears that the use of a targeted representation can lead to the production of designs which are locally optimal.
However,if we use the completion evolution approach presented in the second experiment we get further improvements in performance.We will leave to the Dis-cussion section further discussion of the other advantages of the approach presen-ted.
5.Discussion
The analysis just presented can be easily extended to include general object gram-mars of types different to the kindergarten grammar.The proposed approach can be considered as an implementation of the simplest version of the genetic engin-eering approach to the generic design problem.From the technical point of view the algorithm presented is a mixture of a stochastic search method (which may be a genetic algorithm)and a pattern recognition technique.
The genetic engineering approach can be applied in a similar fashion to the problem of the generation of a “suitable”shape grammar (Gero and Kazakov,1995)where the complex building blocks correspond to the evolved grammar rules.
18John S.Gero AND Vladimir A.Kazakov
As already mentioned in the analysis of the numerical experiment,the evolved representations are highly redundant-the same composite building blocks are evolved many times along the different branches of the evolutionary trees.The redundancy level of the current set of composite building blocks can be reduced in a number
of different ways.The simplest is just to delete all the redundant copies from the current set.In the general case,we have to?nd the minimal representation of the subspace which can be generated using the current set of complex building blocks.
The introduction of ideas and methods from genetic engineering into design systems based on genetic algorithms opens up a number of avenues for research into both evolutionary-based design synthesis and into modi?ed genetic algorithms. In design systems based on such modi?ed genetic algorithms it is possible to con-sider two directions.
The?rst is to treat the sequence of the genes which results in certain behaviours
or?tness performances as a form of‘emergence’,emergence of the schema repres-ented by that gene sequence.The use of the genetically engineered complex genes changes the properties over time of the state spaces which are being searched.This allows us to consider the process as being related to design exploration modelled
in a closed world.The precise manner in which the probabilities associated with states in the state space change is not yet known.Clearly,this is also a function
of whether a?xed length genotype encoding is used or not.If a variable length genotype encoding is used with the genetically engineered complex genes then the shape of the state space remains?xed but the probabilities associated with the states within it change.If a?xed length genotype encoding is used with the ge-netically engineered complex genes then the shape of the state space changes in addition to the probabilities associated with states in the state space.
The second is to treat the genetically engineered complex genes as a means of developing a representation for potential designs.A fundamental part of design-ing is the determination of an appropriate representation of the components which are used in the structure(Gero,1990)of the design.This is part of that aspect
of designing called‘formulation’,ie the determination of the variables,their re-lationships and the criteria by which resulting designs will be evaluated.In most computer-aided design systems the components map directly on to variables.Fur-ther,in such systems the variables are speci?ed at the outset,as a consequence there
is an unspeci?ed mapping between the solutions capable of being produced and the variables chosen to represent the ideas which are to be contained in the resulting designs.The genetic engineering approach described provides a means of automat-ing the representation part of the formulation process.The level of granularity is determined by the stability condition of the evolutionary process or can be determ-ined by the user.The targeted building blocks provide a high-level starting point for all later designs which are to exhibit the required characteristics as evidenced in the earlier designs.It is this latter requirement which is met by this formal method.
The following simple picture can be used to summarize the model described in
Evolving Building Blocks for Design Using Genetic Engineering19 this paper.A group of children are playing with the“Lego”game using not more than50squares.They join them together and want to build the object with the largest number of closed spaces inside.After each child has built his or her ob-ject the supervisor tries to?nd a combination of squares which is present in many of the best designs but is present in none or only in a few of unsatisfactory designs. Then he makes this combination permanent by gluing its components together and adds a bunch of such permanent combinations to the pool of building elements available to the children.Then the children make another set of objects using these new building blocks as well as an old ones.The supervisor tries to?nd another “good”composite block and the process is repeated.Thus,two steps occur in each cycle:?rst children make a set of new designs from currently available blocks and combination of blocks and second the supervisor tries to single out the additional combination of blocks that should be employed.If there are no such combinations which distinguish“good”design from the“bad”ones then we will not get new combinations but only the improved designs.
Style
The choice of particular variables and con?gurations of variables is a determin-ant of the style of the design(Simon,1975).The label‘style’can be used in at least two ways:either to describe a particular process of designing or as a means of describing a recognizable set of characteristics of a design.Thus,it is possible to talk about the‘Gothic’style in buildings or the‘high tech”style of consumer goods.Precisely what goes to make up each of these styles is extremely dif?cult to articulate even though we able to recognize each of these styles with very little dif?culty.An appropriate question to pose is:how can we understand what pro-duces a style during the formulation stage of a designing process?This brings us back to the concepts described in this paper.
‘The history of taste and fashion is the history of preferences,of various acts of choice between different alternatives......[But]an act of choice is only of symptomatic signi?cance,is expressive of something only if we can really want to treat styles as symptomatic of something else,we cannot do without some theory of the alternatives’(Gombrich,quoted from(Simon,1975)).
If we use a particular style as the?tness of interest then it should be possible to utilise the genetic engineering approach described in this paper to determine if there is a unique set of genes or gene combinations which is capable of being the progenitors of that style.For this to occur satisfactorily a richer form of pattern recognition will be needed than that alluded to here.We will need to be able to determine a wider variety of gene schemas in the genotypes of those designs which exhibit the desired style.
The use of genetic engineering in evolving schemas of interest opens up a po-tential subsymbolic model of emergence including the emergence of domain se-
20John S.Gero AND Vladimir A.Kazakov
mantics(Gero and Jun,1995).Style can be considered as a form of domain se-mantics.This is of particular interest in design synthesis since,if domain semantics can be captured in a form such as described in this paper,then they can be readily used to synthesize designs which exhibit those semantics and even that style.This is analogous to the induction of a shape grammar which captures the characteristics of designer’s style.
6.Acknowledgments
This work is directly supported by a grant from Australian Research Council,com-puting resources are provided through the Key Centre of Design Computing. References
Coyne,R.,Rosenman,M.,Radford,A.,Balachandran,M.and Gero,J.S.:1990,Knowledge-Based Design Systems,Addison-Wesley,Reading.
Gero,J.S.:1990,Design prototypes:a knowledge representation schema for design,AI Magazine 11(4):26–36.
Gero,J.S.(ed.):1985,Design Optimization,Academic Press,New York.
Gero,J.S.and Jun,H.:1995,Emergence of shape semantics of architectural shapes,Technical Re-port,University of Sydney,NSW2006,Australia,Key Centre of Design Computing. Gero,J.S.and Kazakov,V.:1995,An exploration-based evolutionary model of a generative design process,Microcomputers in Civil Engineering(to appear).
Gero,J.S.,Louis,S.and Kundu,S.:1994,Evolutionary learning of novel grammars for design improvement,AIEDAM8(2):83–94.
Holland,J.:1975,Adaptation in Natural and Arti?cial Systems,University of Michigan,Ann Arbor. Kirkpatrick,S.,Gelatt,C.G.and Vecchi,M.:1983,Optimization by simulated annealing,Science 220(4598):671–680.
Pearson,W.and Miller,W.:1992,Dynamic programming algorithms for biological sequence com-parison,Methods in Enzymology210:575–601.
Simon,H.:1975,Style in design,in C.Eastman(ed.),Spatial Synthesis in Computer-Aided Building Design,Applied Science,London,287–309.
Stiny,G.:1980a,Introduction to shape and shape grammars,Environment Planning B7:343–351.
Stiny,G.:1980b,Kindergarten grammars:designing with Froebel’s building gifts,Environment Planning B7:409–462.
电大人力资源案例问答题 在国内，中高层管理人员供不应求的矛盾十分突出，工商管理硕士的职业发展前景非常看好，尤其是从国外学成回国的MBA 动。职业咨询中心建议他:在组织目标任务确定、管理规范的环境中从事事务性管理工作，目前不适合做高层管理工作，需要增强与人交往的兴趣。 问题:两个人都希望做高层管理者，为什么一个适合，另一个不适合呢?请用职业的选择理论来分析。 答:高层管理者不仅需要具备较高的管理技能，更需要具备一些重要的人格特征。通过对两从位先生的测评和咨询显示，他们都追求成功，有强烈的责任感，但是在人际沟通以及看待问题、做事的方式、职业兴趣上呈现较大的差异，正是这种差异导致了他们一个适合做高层管理人员，而另一个不适合。 从案例中可看出，有两方面是齐先生成为高层管理人员的主要障碍: 一方面是他缺乏宏观、整体意识，不能从组织整体的视野去制定发展战略和计划，关注的重心在于任务的完成和环节。显然，现在也
有很多高层管理者事必躬亲，以身作则，但是，这种行为风格在企业发展中的某个阶段是可以的，从未来发展的趋势看，高层管理者的创新意识、策划能力、对市场的敏感和把握，对企业的生存和发展才是至关重要的。另一方面，他缺乏人际沟通的兴趣。齐先生说自己不喜欢与人打交道，在工作中似乎也找不到合适的同事和助手。 人际沟通对于高层管理人员是非常重要的。一位管理学家说“所谓管理费用就是使人完成工作。”他强调了“人”在管理工作中的重要性。研究表明，中层管理者把80%-90%的时间用在与别人交往上，中上层的管理人员，花在与别人交谈的时间也高达67%。国外曾对管理人员的特点进行调查分析，有100%的人认为“指导能力”是管理者的一个重要特征，有47%的人认为“亲和力”是管理者不可缺少的。作为管理者，最主要的和最大量的工作是与组织内外的各类人打交道，对人际不关心的管理者是不合格的。 MBA教育提供的是管理的理念、技术，但是给自己准确定位，选择适合自己的发展道路，是更值得思考的事情，高层管理者并非唯一出路。 (2)天龙航空食品公司的员工考评 罗芸在天龙航空食品公司担任地区经理快一年了。他分管10家
国开(中央电大)专科《财务管理》十年期末考试题库 (电大期末纸质考试必备资料) 说明：1.试卷号码：2038； 2.适用专业及层次：会计学(财务会计方向)和金融(金融与财务方向)、专科； 3.资料整理于2020年1月10日，收集了2010年1月至2019年7月中央电大期末考试的全部试题及答案。 2019年7月试题及答案 一、单项选择 1.下列各项企业财务管理目标中，能够同时考虑资金的时间价值和投资风险因素的是(A)。 A.产值最大化 B.利润最大化 C.每股收益最大化 D.股东财富最大化 2.某种股票为固定成长股票，股利年增长率6%，预计第一年的股利为8元／股，无风险收益率为10%，市场上所有股票的平均收益率为16%，而该股票的贝塔系数为1.3，则该股票的内在价值为(B)元。 A.65.53 B.67.8 C.55.63 D.71.86 3.某人分期购买一辆汽车，每年年末支付10 000元，分5次付清，假设年利率为5%，则该项分期付款相当于现在一次性支付(C)元。 A.55 256 B.40 259 C.43 295 D.55 265 4.某投资方案的年营业收入为10 000元，年付现成本为6 000元，年折旧额为1 000元，所得税率为25%，该方案的每年营业现金流量为(C)元。 A.1 680 B.2 680 C.3 250 D.4 320 5.与其他短期融资方式相比，下列各项属于短期融资券缺点的是(D)。 A.融资成本低 B.融资数额大 C.可提高公司影响力 D.风险大 6.某公司普通股目前的股价为10元／股，融资费率为8%，刚刚支付的每股股利为2元，股利固定增长率3%，则该股票的资本成本为(D)。 A.22.39% B.21.74% C.24.74% D.25.39% 7.某公司近年来经营业务不断拓展，目前处于成长阶段，预计现有的生产经营能力能够满足未来10年稳定增长的需要，公司希望其股利与公司盈余紧密配合。基于以上条件，最为适宜该公司的股利政策是(C)。 A.剩余股利政策
人力资源管理 作业1 一、选择题 1．具有内耗性特征的资源是( )。 A．自然资源 B．人力资源 C．矿产资源 2．人力资源管理与人事管理的关键区别体现在( ) A．内容上B．观念上C．工作程序上D．工作方法上 3．以人性为核心的人本管理的主体是何种?( ) A．职工 B．环境C．文化 D．价值观 4．某企业对10名新招来的员工进行上岗前培训，从讲课到实习一共花了5000元费用，请问这笔费用应从人力资源成本的哪个工程中列支?( ) A.获得成本B．开发成本C．使用成本D．保障成本 5．预测由未来工作岗位的性质与要求所决定的人员素质和技能的类型，这是制定人力资源规划时哪一个步骤?( )A．预测未来的人力资源供给B．预测未来的人力资源需求C．供给与需求的平衡 D．制定能满足人力资源需求的政策和措施 6．从现实的应用形态来看；下面哪个方面不是人力资源所应包含的内容? ( ) A．体质B．智力C．思想D．技能7．把“员工视为活动主体、公司主人”是哪一种人力资源管理模式?( ) A. 自我中心式、非理性化家族管理B．以人为中心、非理性化家族管理 C．以人为中心、理性化团队管理D．自我中心式、理性化团队管理 8．每个员工都明确企业发展目标，团结协作努力实现企业目标，这反映了“以人为中心、理性化团队管理”模式的什么特点?( ) A．封闭式的自危表现B．开放式的悦纳表现!C．封闭式的悦纳表现 D．开放式的自危表现9．下面哪一项不是人本管理的基本要素?( ) A．企业人B．环境C．文化D．产品10．与员工同甘共苦、同舟共济，反映了人本管理哪方面的基本内容?( ) A．人的管理第一B．以激励为主要方式C．积极开发人力资源 D．培育和发挥团队精神11．明确目标责任，使其竞争，是进行人本管理的哪种运行机制?( ) A．动力机制B．压力机制C．约束机制D．环境影响机制12．通过检查人力资源目标的实现程度，来提供关于人力资源计划系统的反馈信息。这是人力资源规划系统的哪项活动?( ) A．人员档案资源B．人力资源预测C．行动计划D．控制与评价13. 工作分析中方法分析常用的方法是( ) A．关键事件技术 B．职能工作分析C．问题分析D．流程图 14. 管理人员定员的方法是( ) A．设备定员法 B．效率定员法 C．职责定员法 15. 依据个体的经验判断，把所有待评价的职务依序排列，由此确定每种工作的价值的方法是( )。 A．因素分解法B．因素比较法 C. 经验排序法D．因素评分法 二、案例分析1：贾厂长的管理模式
“电大在线”选课、学习、发帖、考试平台介绍及操作步骤详解一、“成都电大在线”操作步骤 （一）、选课 登录https://www.doczj.com/doc/491107746.html,（成都电大在线首页）在页面左侧“用户登录”栏 输入账号（学号）、密码（身份证8位生日号码如：19800101）点 “登录”。进入后在右上方分本、专科选你所学专业，如 （若过去已选过课，则直接在“我的学习课程列表”下方点击“选课”，再选专业） 就会出显课程列表，如： 然后对照本期课表，勾选本学期所要选的课（在框里打“”），如： 勾选完后，点击最下面“保存选课”，这样课就选好了。 （二）、网上学习（要求：每门课程在网上学习时间达到60分钟以上，点击浏览或下载视频、课件、文字等学习资源25次以上；发帖或回帖、跟帖1次以上。）
点击“我的学习课程”，左下方会出显示“我的课程列表”， 点击你要学习的课程，如：即进入该课程的课 程学习页面：这些就是所需学习的内容，点击这些内容进行学习即可。 【特别注意：1、进入课程学习页面后系统便开始自动计录你的学习活动（包括在线学习时间、 点击浏览和下载学习资源的次数等），但是，如1小时内无学习活动（点击浏览或下载学习资源），系统将自动退出，且不计录学习时间；2、点击打开一个学习资源后必须停留5分钟以上才能继续打开第二个资源，每次学习活动最多只能同时打开三个学习资源。每个学习资源打开后的停留时间不能超过60分钟，且文字类学习资源打开后30分钟无动作（鼠标移动、翻动页面等）或视频类资源播放完后30分钟不关闭的系统都将视其学习活动无效，不予记录；3、在完成某项内容学习后，点击下面
的“关闭页面”，可继续点击打开其他内容学习；4、在退出课程学习页面时一定要从该页面右上的 “退出系统”退出，不能直接关闭课程学习页面，否则你的学习活动也将不被记录！！！】 （三）、发帖、回帖、跟帖（要求：每门课程1次以上） 1.发帖：学员可根据需要在多个“论坛”发帖（亦可回帖、跟帖） （1）、在课程论坛发帖： 【特别注意：所发帖子必须为有效帖！即：帖子内容必须与该门课程学习的内容、方法等有关：如学习中的问题、体会、感悟、学习方法交流以及实时导学签到等。反之：凡无意讨论课程内容，直接征求作业和考试答案；或纯粹打招呼、呼朋唤友；或复制粘贴别人的帖子以及发布与本课学习无关的帖子，如询问考试成绩、学籍学务问题等，都将视为无效帖，不予记录。】 方法：按前面“二、”操作进入课程学习页面。在学习页面上面的“论坛帖子”中点击右下方的“更多课程帖子” 即进入该门课程的 “互动论坛”再点击 “发新帖”即进入发帖页面：在“帖子主题”栏内简明扼要写出你所发帖子的主题（如：“管理和领导是一回事吗？”）； 如你觉得有必要的话，还可在下面的框内对主题进一步说明或阐述（也可什么都不说，直接发主题） 内容写好后，点“发帖”即可。（这时回到“互动论坛”页面便可看见你发的帖。辅导教师或其他学员将回答你的问题或参与讨论。） （2）、参加“成都电大网上导学”发帖： （说明：1、这种发帖的形式与（1）是一样的，也是在“课程互动论坛”上进行。所不同的是它是由专业课教师在规定的时间里，就规定的内容在线组织大家发帖讨论、答疑。属教师在线辅导性质。为此，学员需先查询导学时间和内容安排并提前做好学习准备以便届时参与发帖、讨论；2、参与导学时的“签到贴”亦视为有效贴）。
第1题: 广义的人力资源规划实质上是( )。（D）所有人力资源规划的总称 第2题: 长期的人力资源规划一般在( )以上。（A）五年 第3题: 在企业中，( )是一种集权和分权相结合的组织结构形式。（B）直线职能制 第4题: 直线制是一种最简单的集权式组织结构形式，又称( )。（C）军队使结构 第5题: 法定休假日安排劳动者工作的，应支付不低于工资( )%的工资报酬。（C）300 第6题: 岗位评价的结果可以是分值形式、等级形式，也可以是排序形式，但是我们最为关心的是( )。（C）岗位与薪酬的对应关系 第7题: 岗位评价方法中成本相对较低的是( )。（A）排列法 第8题: 关于分类法的不正确描述是( )。（D）对精度要求高 第9题: 工作岗位分析与评价的目的在于明确每个岗位的( )。（B）相对价值 第10题: 劳动者实际劳动给付的对象是( )。（B）接受单位 第11题: 受派遣劳动者的工资和社会保险是由( )支付。（C）劳动者派遣机构 第12题: ( )是人力资源开发的最高目标。（A）人的发展 第13题: ( )不属于人力资源创新能力运营体系。（C）创新能力结构体系 第14题: ( )不属于人力资源组织开发的方法。（B）美国模式 第15题: ( )不属于人力资源管理开发的手段。（C）科技手段 第16题: 在管理内容上，现代人力资源管理( )。（C）以人为中心 第17题: 在管理形式上，现代人力资源管理是( )。（C）动态管理 第18题: 在管理方式上，现代人力资源管理采取( )。（C）人性化管理 第19题: 在管理策略上，现代人力资源管理是( )。（B）战略与战术相结合 第20题: 在管理技术上，现代人力资源管理( )。（B）追求科学性与艺术性 第21题: 在管理体制上，现代人力资源管理属于( )。（A）主动开发型 第22题: 在管理手段上，现代人力资源管理( )。（A）以计算机为主 第23题: 在管理层次上，现代人力资源管理部门( )。（B）处于决策层 第23题: 在管理层次上，现代人力资源管理部门( )。（C）人的管理 第25题: ( )不属于现代人力资源管理的三大基石。（D）员工的引进与培养 第26题: 校园招聘的优点是( )。（A）学生的可塑性强 第27题: 所谓( )，就是在招聘广告中不出现招聘企业的名称。（C）遮蔽广告 第28题: 准备在北京地区招聘90名超市收银员，最合适的招聘信息发布渠道是( )。（A）报纸 第29题: 选择报纸刊登广告的好处是( )。（A）为公司做了广泛宣传 第30题: 下列描述不正确的是( )。（B）同一单位招聘申请表项目是相同的 第31题: ( )是根据企业过去的统计资料，或者由社会权威机构对应聘者的条件，按照重要程度确定相应的权数，从而对应聘者自身条件进行综合评价分析的一种表格形式。（C）加权招聘申请表 第32题: 狭义的人力资源规划实质上是( )。（D）企业各类人员需求的补充计划 第33题: 人力资源规划在整个人力资源管理活动中占有( )。（B）重要地位 第34题: 岗位分析为企业员工的考核、晋升提供了( )。（C）基本依据 第35题: 岗位分析的最终成果是制作出岗位说明书和( )。（B）岗位规范 第36题: ( )能使员工完成任务的内容、形式和手段发生变化。（C）工作扩大化 第37题: 岗位设计工作的人手点不包括( )。（D）劳动关系的改善 第38题: 人力资源管理的基础是( )。（E）工作分析 第39题: 设置岗位的基本原则是( )。（B）因事设岗 第40题: 根据生产总量核算定员人数属于( )。（C）按劳动效率定员
中央广播电视大学2009——2010学年度第二学期“开放本科”期末考试知识产权法试题 一、填空题(每空 1 分，共10分) 1. 根据是否以他人商标权为前提，可将商标权取得的方式分为 ________ —与___________ 。 2. __________________________________________ 我国商标法规定，注册商标的有效期为______________________________________ ，自___________ 之日起计算。 3. 我国专利法实施细则规定，发明是指对——————、——————或者其改进所提出的新的技术方案。 4. 我国著作权法规定，依法禁止——————、——————的作品，不受法律保护。 5. 作品的构成条件包括——————、范围有限性、可感知性和——————。 二、单项选择题(每小题 2 分，共 2 0 分) 1. 提供服务的经营者在其服务项目上所使用的，用以区别于其他服务者所提供的服务项目的显著性标志，叫做( )。 A. 制造商标 B.销售商标 C.服务商标 D.图形商标 2. 甲公司对乙公司的注册商标有异议，其申请注册商标争议裁定的法定期限为( )。 A. 1 年 B. 2 年 C. 3 年
D. 5年 3．我国著作权法规定，著作权自( )之日起产生。 A. 作品发表 B.作品创作完成 C.作品登记 D.作品公告 4. 我国商标法规定，未经注册商标所有人的许可，在同一种或者类似商品或者服务上使用与其注册商标相同或者近似的商标的行为属于商标侵权，这类行为称为( )。 A. 使用侵权 B.销售侵权 C.标识侵权 D.反向假冒侵权 5. 我国专利法规定，对产品的形状、图案、色彩或者其所作出的富有美感并适合于工业上应用的新技术称为( )。 A. 发明 B.实用新型 C.外观设计. D.新技术 6. 授予专利权的发明和实用新型应具备三个条件，其中在申请日以前没有同样的发明或实用新型在国内外出版物上公开发表过、在国内公开使用过或者以其他方式为公众所知，也没有同样的发明或实用新型由他人向专利局提出过申请并且记载在申请日以后公布的专利申请文件中，这属于三个条件中的( ) A. 新颖性 B.创造性 C.实用性 D.推广性 7. 发明专利的保护期为( )。 A. 5 年 B. 10 年 C. 20年 D. 30 年 8．认为专利权的授予不符合法律规定，可以宣告专利权无效的请求人是( )。
第一章自测题 一、单选题 1. 一个国家或地区的人口总体的数量表现，是指（）。 B A. 人力资源 B.人口资源 C.劳动力资源 D.人才资源 2.（）是人力资源的基础。 B A. 人力资本 B.人口资源 C.劳动力资源 D.人才资源 3.具有内耗性的资源是（）。B A.物质资源 B.人力资源 C.矿产资源 D.信息资源 4.人力资源管理理解正确的是（）。 D A.以事为中心 B.消费性部门 C.视人为物 D. 用人看重潜能 5.人力资源的形成不是一个自然而然的过程，它需要有计划有组织地去培养和发掘，这一点体现出人力资源（）特点。 D A. 再生性 B.资本性 C.时效性 D.可控性 6.主张集体奖，而不主张个人奖。这是哪种假设的思想？（） B A. “经济人”假设 B.“社会人”假设 C.“自我实现的人”假设 D.“复杂人”假设 7.（）是假设人的行为动机就是为了满足自己的私利，工作是为了得到经济报酬的一种人性理论。 C A.复杂人 B.社会人 C.经济人 D.自我实现人 8.“复杂人”假设是由（）提出的。 D A.马斯洛 B. 泰罗 C.法约尔 D.薛恩 9.与人本心理学时期相关的的人力资源理论是（）。 D A.人际关系学说 B. 科学管理 C.《道德经》 D.期望理论 10. 二、多选题 1.下列关于人力资源管理特点理解正确的是（）。ABD A. 以人为中心 B.效益性部门 C.用人看重经验 D.视人为资源 E.视人为工具
2.（）是人本管理的基本要素。ACDE A. 企业人 B.产品 C.文化 D.环境 E.价值观 3.人本管理的机制有（）。ABCDE A. 动力机制 B.压力机制 C.约束机制 D.保证机制 E.选择机制 4.下列关于人才资源理解正确的包括（）。AD A. 侧重于人的质量 B.突出人的数量和劳动者的数量 C.注重人口数量和质量的统 一 D.反映了一个民族的素质 E. 我国在人才资源的整体水平上居世界首位 5.影响人力资源数量的因素包括（）。ABC A. 人口总量及其生产状况 B.人口年龄结构及其变动 C.人口迁移 D. 劳动者的身体素质 E. 劳动者的智能水平 6.下列关于“操作层面”的人力资源管理理解正确的包括（）。ABCD A. 以事务工作为主 B.以岗位管理作为组织实施人力资源管理的基础或出发点 C.带有极强的机械主义色彩 D. 强调人与岗位是相对应的 E. 属于人力资源管理的“后技术时代” 7.关于当代中国人力资源管理理解正确的包括（）。BCD A.已经基本完成“后技术时代”人力资源管理的主要部分工作 B.需要加强人力资源相关制度的的修正.完善 C.属于以“战略层面“和“终极层面”为主的人力资源管理时期 D. 还有一些组织处于人力资源管理的“技术时代” E. 组织更注重“非确定性情景”下的人力资源管理 8.关于人际关系时期，下列说法错误的包括（）。ADE A.泰勒为代表人物 B.理论基础为需要层次理论 C.提倡实施针对性的管理 D. 与全面质量管理时期相同步 E. 体现根据不同人采取不同管理方式的“权变管理” 三、判断题 1.人力资源侧重于人的质量。错误 A. 正确 B.错误 2.小王一周工作很累，经过星期六与星期天的休息，体力得到恢复。这体现着人力资源的时效性与再生性。错误 A.正确 B.错误 3.人事管理用人注重经验，人力资源管理用人看重潜能。正确 A. 正确 B.错误 4.人力资源管理的根源最早可以追溯至中国古代的《管子》。错误
试卷代英语Ⅱ国家开放大学中央电大期末考试 试卷及答案 SANY标准化小组 #QS8QHH-HHGX8Q8-GNHHJ8-HHMHGN#
试卷代号：1162 中央广播电视大学2009-2010学年度第一学期 “开放本科”期末考试英语Ⅱ(2) 试题 2010年1月 第一部分交际用语（共计10分，每小题2分） 1-5小题：阅读下面的小对话，从A、B、C、D四个选项中选出一个能填入空白处的最佳选项， 并在答题纸上写出所选的字母符号。 1. - How was the journey to London ----------- A. It was a nirw-hour plane journey B. It went very wellC. I flew there D. I was very well 2. - Can you help me clear up the mess-------------------- A. Tell me who made it B. No trouble at allC. Yes, that'll be all right D. No problem 3. - What's the fare to the museum-------------------- A. Five hours B. Five o'clock C. Five miles D. Five dollars 4. - Is it going to be warm next week------------------- A. Yes, it is B. I don't believe it C. No, it hasn't D. It changes all the time 5. - Here you are, . Yes, I'm here B. Yes, here we are C. Thank you very much D. That's all right 第二部分词汇与结构（20分，每小题2分） 6 - 15小题：阅读下面的句子，从A、B、C、D四个选项中选出一个能填入空白处的最佳选项， 并在答题纸上写出所选的字母符号。 6. Unfortunately the poor girl can't do anything but ------- . all her belongings at alow price. A. to sell B. selling C. sell D. sold 7. They were asked to avoid _ any water which had not been boiled. A. drinking B. to drink C. having D. not to be drunk 8. John was bored, _ he left his life in England. . A. in addition B. so C. however D. furthermore 9. She is very careful. She _ very few mistakes in her work. A. does B. takes C. makes D. gets 10. The workers are busy _ models for the exhibition. A. to make B. with making C. being making D. making 11. --------- you change your mind, I won't be able to help you. A. When B. Unless C. While D. So 12. We are going to have our office _ to make room for a new engineer. A. to rearrange B. rearrange D. rearranging 13. Since this road is wet and slippery this morning, it _--------- last night. A. must rajn B. was raining C. may be raining D. must have rained 14. Do you think Tommy is--------- the truth A. sayingB. tellingC. speaking D. talking 15. John fell asleep _ he was listening to the music. A. while B. before C. after D. as soon as 第三部分完形填空（每题2分，共20分） 16 - 25小题：阅读下面的短文，从短文后所给的A、B、C、D四个选项中选出能填入相应空白 处的最佳选项，并在答题纸上写出所选的字母符号。 Genetic Engineering Many people _ 16 unaware that a lot of the foods they eat every day, 17 bread,ham and cheese, have been altered by using new technology. Food can be changed. It can bemade to taste different or to look different - carrots can be
第1题: 广义的人力资源规划实质上是(所有人力资源规划的总称)。 第2题: 长期的人力资源规划一般在(五年)以上。 第3题: 在企业中，(直线职能制)是一种集权和分权相结合的组织结构形式。 第4题: 直线制是一种最简单的集权式组织结构形式，又称(军队式结构)。 第5题: 法定休假日安排劳动者工作的，应支付不低于工资(300 )%的工资报酬。 第6题: 岗位评价的结果可以是分值形式、等级形式，也可以是排序形式，但是我们最为关心的是(岗位与薪酬的对应关系) 第7题: 岗位评价方法中成本相对较低的是( 排列法)。 第8题: 关于分类法的不正确描述是( 对精度要求高)。 分类法使用于大企业管理岗位；划分类别是关键；成本相对较高。 第9题: 工作岗位分析与评价的目的在于明确每个岗位的( 相对价值)。 第10题: 劳动者实际劳动给付的对象是( 接受单位)。 第11题: 受派遣劳动者的工资和社会保险是由( 劳动派遣机构)支付。 第12题: (人的发展)是人力资源开发的最高目标。 第13题: (创新能力结构体系)不属于人力资源创新能力运营体系。 人力资源创新能力运营体系包括创新能力配置体系、创新能力激励体系、创新能力开发体系。 第14题: (美国模式)不属于人力资源组织开发的方法。 莱维特模式、利温模式、格雷纳模式属于人力组织开发的方法。 第15题: (科技手段)不属于人力资源管理开发的手段。 经济手段、行政手段、法律手段属于人力资源管理开发的手段。 第16题: 在管理内容上，现代人力资源管理( 以人为中心)。 第17题: 在管理形式上，现代人力资源管理是( 动态管理)。 第18题: 在管理方式上，现代人力资源管理采取(人性化管理)。 第19题: 在管理策略上，现代人力资源管理是( 战略与技术相结合的)。 第20题: 在管理技术上，现代人力资源管理( 追求科学性和艺术性)。 第21题: 在管理体制上，现代人力资源管理属于( 主动开发型)。 第22题: 在管理手段上，现代人力资源管理(以计算机为主)。 第23题: 在管理层次上，现代人力资源管理部门( 处于决策层)。 第24题: 企业管理的核心是( 人的管理)。 第25题: ( 员工的引进与培养)不属于现代人力资源管理的三大基石。 员工的技能开发、员工的绩效管理、定编定岗定员定额属于现代人力资源管理的三大基石。第26题: 校园招聘的优点是( 学位的可塑性强)。 第27题: 所谓( 遮蔽广告)，就是在招聘广告中不出现招聘企业的名称。 第28题: 准备在北京地区招聘90名超市收银员，最合适的招聘信息发布渠道是( 报纸)。第29题: 选择报纸刊登广告的好处是( 为公司做了广泛宣传)。 第30题: 下列描述不正确的是( 同一单位招聘申请表项目是相同的)。 同一单位招聘申请表项目不一定相同； 招聘申请表设计得好，可降低成本； 设计申请表，要考虑政策要求。 第31题: ( 加权招聘申请表)是根据企业过去的统计资料，或者由社会权威机构对应聘者的条件，按照重要程度确定相应的权数，从而对应聘者自身条件进行综合评价分析的一种表格形式。 第32题: 狭义的人力资源规划实质上是( 企业各类人员需求的补充规划)。
2018年中央电大《建筑结构》期末考试试题附全答案 一、填空题（每小题2分，共20分） 1．对于有明显流幅的钢筋(俗称软钢)，一般取（屈服强度）作为钢筋设计强度的依据。 2．混凝土强度等级的表示方法为，符号C代表（混凝土），C后面的数字表示以（）为单位的立方体抗压 强度标准值。 3．结构的极限状态分（承载能力极限状态）和（正常使用极限状态）两种。 4．混凝土和钢筋的强度设计值，定义为强度标准值（除以）相应的材料强度分项系数。 5．（纵向受力钢筋）的外边缘至（混凝土）表面的垂直距离，称为混凝土保护层厚度，用c表示 6、钢筋混凝土从加荷至构件破坏，梁的受力存在着三个阶段，分别为弹性工作阶段、（带裂缝工作阶段） 和（破坏阶段）阶段。 7、受拉钢筋首先到达屈服，然后混凝土受压破坏的梁，称为（适筋）梁，这种破坏称为（延性） 破坏。 8、影响有腹筋的斜截面破坏形态的主要因素是（剪跨比）和（配箍率）。 9、钢筋混凝土梁的斜截面破坏的三种形式中，只有（剪压）破坏是斜截面承载力计算的依据。 10、T型截面梁按中和轴位置不同分为两类：第一类T型截面中和轴位于（翼缘内）,第二类T型截面中和 轴位于（梁肋内）。 7.在截面设计时，满足下列条件（D）则为第二类T形截面。 A．B．)C．D． 8．当剪跨比适中，且腹筋配置量适当时，常发生( B)。 A．适筋破坏B．剪压破坏C．斜压破坏D．斜拉破坏 9．在钢筋混凝土梁斜截面承载力计算中，若，则应采取的措施是（A）。 A．加大截面尺寸B．增加箍筋用量C．配置弯起钢筋D．增大纵筋配筋率 10.条件相同的无腹筋梁，发生斜压、剪压、斜拉破坏形态时，梁的斜截面抗剪承载力的大致关系是（A）。 A．斜压破坏的承载力>剪压破坏的承载力>斜拉破坏的承载力 B．剪压破坏的承载力> 斜压破坏的承载力>斜拉破坏的承载力 C．斜拉破坏的承载力>斜压破坏的承载力>剪压破坏的承载力 D．剪压破坏的承载力>斜拉破坏的承载力>斜压破坏的承载力 二、选择题 1．对于无明显屈服点的钢筋，其强度标准值取值的依据是（D） A.最大应变对于的应力 B.极限抗拉强度 C. 0.9倍极限抗拉强度 D.条件屈服强度 2.我国规范采用（A ）作为混凝土各种力学指标的代表值。 A．立方体抗压强度B．轴心抗压强度C．轴心抗拉强度D．劈拉强度 3. 建筑结构在其设计使用年限内应能满足预定的使用要求，有良好的工作性能，称为结构的（B）A．安
一、判断题 1、人力资源得基础就是人得知识与技能。(N) 2、围绕西部开发、振兴东北老工业基地、中部崛起等国家得战略部署与重点任务,我们应大力开发自然资源,并以此促进人类社会得长远发展。(N) 3、人力资源规划就就是要保障组织发展将来所需得人力资源。(N 4、人力资本关注得就是收益问题,人力资源关注得就是价值问题。(Y 5、我们可以用工作活动流程图得形式来揭示工作任务得操作要素与流向。(Y) 6、影响企事业组织招聘得内部因素有很多,其中企事业组织得性质就是首要因素。(N 7、员工培训得基本程序得第一步就是制定培训计划。N) 8、员工考评就是人力资源管理过程中最为核心得环节。(Y) 9、员工薪酬就就是指发给员工得工资。(N) 10,职业发展就是指组织或者员工本人确定得一个职业生涯得发展目标。(Y) 11、即便用人单位及管理人员违章指挥、强令冒险作业,但为了严明劳动纪律,员工也必 须服从执行,无权拒绝。(N) 12、根据劳动合同法得相关规定,组织与员工之间建立劳动关系,不就是必须签订劳动合同得。(N) 1.我国超过法定退休年龄得人不属于现实得人力资源。(×) 2.组织中任何战略规划得实施都离不开人力资源战略得支撑。( √) 3.为了评估规划得有效性,规划人员有必要首先确定评估标准。(√) 4.人力资源成本会计既要研究如何计量在获得与开发人力资源方面组织得投资,又要研究如何计量目前录用人员得重置成本。(√) 5.工作分析作为一种活动,其主体就是工作分析者,客体就是工作环境。( ×) 6.招聘程序得第一步就是招募。( √) 7.员工培训得内容主要有两个方面:即业务技能与业务知识。( ×) 8.员工考评只能由员工得主管对其进行考评。( ×) 9.在贯彻按劳取酬原则时,需要综合考虑三种劳动形态,即潜在形态、流动形态与物化形态,应以潜在劳动为主要依据,同时考虑物化劳动与流动劳动来进行分配。( ×) 10.职业生涯管理只就是员工对自己得职业工作经历进行设计与规划得过程。(×) 11.要搞好员工保障管理体系建设,就必须保障人权,满足社会成员基本生活需求。(√) 12、劳动关系就是指劳动者为其组织提供劳动,并从组织获得报酬形成得一种社会关系。(√) 1.人力资源得基础就是人得体力与智力。( √) 2.人力资源管理战略与组织战略之间没有任何关系。(N ) 3.人力资源规划工作得内容包括人力资源组织、生产、营销等方面得计划。( N ) 4.人力资源价值会计研究人力资源对组织产生得价值。( √) 5.效率定员计算法适用于一切能用劳动定额表现生产工作量得工种或岗位。( √) 6.面试方法可以全面测评个体得任何素质。(√) 7.员工培训只对员工进行专业知识技能得培训。( N) 8.员工考评指标得设计直接影响到整个员工考评得质量。( √) 9.技能工资制适用于生产专业化、自动化程度较高得生产流水作业,以及分工细、同一岗位技能要求差别不大得企业与工种。( √) 10.员工可以通过组织目标得实现达成自己得职业发展目标。(N) 11.社会保障行政管理与基金运营由同一机构负责。N 12.签定劳动合同就是建立劳动关系得具体方式。( √)
【软件工程】形成性考核册答案 电大【软件工程】形考作业一： 第一、二章 一、填空题 1. 软件工程学的内容可包括（理论）、结构、（方法）、（工具）、环境、管理、规范等。 2. 软件生存周期一般可分为问题定义、（可行性研究）、（需求分析）、设计、（编码）、测试、运行与维护阶段。 3. 可行性研究的目的是（用最小的代价在尽可能短的时间内确定问题是否能够解决）。 4. 系统流程图是（描绘物理系统）的传统工具。 5. 在可行性研究中，（技术可行性）是系统开发过程中难度最大，最重要的一个环节。 二、判断题 1. 软件工程采用的生存周期方法就是从时间角度对软件的开发和维护这个复杂问题进行分解，将软件生存的时期分为若干阶段。（√） 2. 螺旋模型是利用笛卡尔坐标的四个象限上分别代表四个方面的活动来表示。（√） 3. 原型模型适用于嵌入式软件。（╳） 4. 面向对象技术是构件组装模型的基础。（√） 5. 在可行性研究中最难决断和最关键的问题是经济可行性。（╳） 6. 系统流程图表达的是部件的信息流程，还表示对信息进行加工处理的控制过程。（╳） 7. 自底向上成本估计不是从整体开始，而是从任务单元开始。（√） 三、单项选择题 1. 下列哪个阶段不是软件生存期三个阶段中的内容（C ）。 A．计划阶段 B．开发阶段 C．编码阶段 D．维护阶段 2. 下列关于瀑布模型的描述正确的是（C ）。
A. 瀑布模型的核心是按照软件开发的时间顺序将问题简化。 B. 瀑布模型具有良好的灵活性。 C. 瀑布模型采用结构化的分析与设计方法，将逻辑实现与物理实现分开。 D. 利用瀑布模型，如果发现问题修改的代价很低。 四、简答题 1. 软件的定义与特点。 参考答案： 软件是计算机系统中与硬件相互依存的另一部分，它是包括程序，数据结构及其相关文档的完整集合。它包括三个方面的内容： ⑴程序：在运行时，能提供所希望的功能和性能的指令集。 ⑵数据结构：使程序能够正确运行的数据结构。 ⑶文档：描述程序研制过程、方法及使用的图文材料。 软件产品具有以下一些特性： ①软件是一种逻辑实体，而不是具体的物理实体，因而它具有抽象性。②软件是通过人们的智力活动，把知识与技术转化成信息的一种产品，是在研制、开发中被创造出来的。③在软件的运行和使用期间，没有硬件那样的机械磨损、老化问题。④软件的开发和运行经常受到计算机系统的限制，对计算机系统有着不同程度的依赖性。⑤软件的开发至今尚未完全摆脱手工的开发方式。⑥软件的开发费用越来越高，成本相当昂贵。 也可概括为： ⑴抽象性：逻辑实体，可记录，但看不到。 ⑵可复制性：与开发成本相比，复制成本很低。 ⑶无折旧。 ⑷受硬件制约。 ⑸未完全摆脱手工工艺。 ⑹开发费用高。 2. 什么是软件工程。 参考答案：
试卷号：1044 2019年7月国开（中央电大）法学本科《合同法》期末考试试题及答案 说明：资料整理于2019年12月9日；该课程考试方式为闭卷，考试时间90分钟，形考手段纸质（30%），终考手段为纸质（70%）。 一、单项选择题 1.下列合同中，应当采取书面形式的是(A)。 A.技术转让合同 B.买卖合同 C.租赁合同 D.保管合同 2.合同成立的一般时间为(B)。 A.要约发出时间 B.承诺生效时间 C.合同签字时间 D.签订确认书时 3.甲对乙声称：“我打算卖掉家中祖传的一套家具”，乙立即向甲表示：“我愿意以10万元价值购买此套家具”。下列判断正确的是(D)。 A.甲对乙的表示构成要约 B.乙对甲的表示构成承诺 C.甲对乙的表示构成承诺 D.乙对甲的表示构成要约 4.债权人依照合同法的规定提起代位权诉讼的(B)。 A.由原告住所地人民法院管辖 B.由被告住所地人民法院管辖 C.由原告户籍所在地人民法院管辖 D.由被告户籍所在地人民法院管辖 5.债务人将合同的义务全部或者部分转移给第三人，(D)。 A.应当通知债权人 B.不必通知债权人 C.不必经债权人同意 D.应当经债权人同意 6.供用水合同主要适用于(C)。 A.乡镇的供水 B.农村的供水 C.城市的供水 D.有居民居住的地方的供水 7.可以任意撤销的赠与合同，(A)。 A.仅限于一般的赠与合同 B.可以是所有类型的赠与合同 C.可以是经过公证的赠与合同 D.可以是具有道德义务性质的赠与合同 8.我国《合同法》规定租赁合同的当事人约定的租赁期限不得超过(D)。 A.l年 B.5年 C.10年 D. 20年 9.在建设工程合同中，若总承包人合法地将承包工作交分包人完成，则分包人完成的工作成果应由(C)向发包人承担责任。 A.总承包人 B.分包人
中央电大2018春开放教育学习指南_0001 试卷总分：100 测试时间：60 单项选择题判断题 一、单项选择题（共20 道试卷，共50 分。） 1. 学生注册中央广播电视大学开放教育学习前需要参加（） A. 入学水平测试 B. 全国成人高考 C. 全国普通高考 2. 中央广播电视大学开放教育试点通过教育部总结性评估的时间是（） A. 2004年 B. 2007年 C. 2008年 3. 下列载体形式中属于录像教材的是（） A. IP课件 B. DVD C. VOD 4. 统一开设开放教育专业和主干课程的单位是（） A. 中央广播电视大学 B. 省级广播电视大学 C. 教案点 5. 截至2009年，分布在全国电大系统的县级电大教案点的数量是（） A. 1000多个 B. 2000多个 C. 3000多个 6. 学生使用课程教案资源时，为了提高学习效率，一般应该（）
A. 选择该课程全部教案资源，逐一学习 B. 根据自身条件和学习习惯选择资源，综合运用 C. 没有必要看学校提供的任何资源 7. 关于开放教育学生的毕业及毕业证书，正确的说法是（） A. 学生只要累积的学分达到最低毕业学分数要求就可以毕业 B. 毕业证书由中央广播电视大学颁发并报教育部进行电子注册 C. 毕业证书由省级广播电视大学颁发并报当地教育厅进行电子注册 8. 下列不属于开放教育学生权利和义务的是（） A. 在学校学习的资格，如上课、考试、上网学习等 B. 受学校规章制度、纪律的约束和管理，如按时缴纳学费、参加学习和考试等 C. 要求学校保证颁发毕业证书 9. 下列情况中，不属于开放教育学生学籍档案主要内容的是（） A. 学生的基本信息，学期、课程和考试注册情况与考试成绩 B. 学习过程的奖惩情况 C. 学生网上学习记录 10. 下列关于开放教育教师和学生关系的说法，不正确的是（） A. 教师是学生“意义建构”的帮助者、促进者 B. 教师是简单的知识传授者与灌输者 C. 学生是对新知识进行“加工”并与已有知识建立联系的主体 11. 修读开放教育专科或本科（专科起点）的学生需要学习的课程总门数大约有（） A. 10门 B. 15门 C. 20门 12. 开放教育的学习方式主要是（） A. 自主学习
【选择题题库】 01．提出“教育是生活的过程，而不是将来生活的预备。教育是经验的改造或改组”的教育家是()。 B．杜威 02．教育的目的主要受( )因素的制约。 D．政治制度 E．生产力发展水平 03．我国现行的学校教育系统，在类别上包括( )。 A．基础教育体系 B．职业技术教育体系 C．高等教育体系 D．成人继续教育体系 E．特殊教育学校系统04．按照评价的不同目的，课程评价可分为( )。 C．决策性评价 D．研究性评价 E．工作性评价 5．“学为人师，行为世范”体现了教师工作的()。 C．示范性05．教学方法、教学手段及组织形式主要受( )的制约。E．生产力发展水平 06．有的放矢、因材施教等，都是指要根据学生身心发展规律中的( )进行教育。D．个别差异性 07．我国现行的学校教育系统的级别层次包括( )。 A．学前教育 B．初等教育 C．中等教育 D．高等教育 08．地方课程开发中应把握的特征包括( )。 B．地域性 C．针对性 D．社会性 E．探究性 5．在德育教育中，“抑其所短，扬其所长”是( )德育原则的要求。B．长善救失 09．作为进步教育的代表，杜威提出的基本主张包括()。A．学校即社会 B．教育即生活 D．儿童中心 10．我国现行的学校教育系统的级别层次包括()。 A．学前教育 B．初等教育 C．中等教育 D．高等教育
11．以语言传递为主的教学方法包括( )。 A．讲授法 B．谈话法C．讨论法 D．读书指导法 12．“学为人师，行为世范”体现了教师工作的( C．示范性 13．学校管理工作一定要按客观规律办事，正确处理主观与客观、理论与实际、传统经验与现代管理科学之间的关系，这是学校管理原则中的( )的要求。D．科学性原则 14．标志着教育学作为一门独立的学科开始形成的教育论著是( )。C．《大教学论》15．教育教学中搞“一刀切”、“齐步走”，违背了个体身心发展的( )。B．个别差异性 16．教育目的是教育活动的出发点和归宿，它的功能有( )。A．定向功能 B．调控功能 C．激励功能 D．评价功能 17．具有信息时代特征的教师职业形象有( )。 B．伴奏者 C．协助者 D．领导者 E．促进者 18．学校管理工作必须讲求效益，做到低消耗、高质量、高速度，力求用尽可能少的人力、物力、财力和时间，取得尽可能好的工作效果，这是学校管理原则中的( )的要求。B．有效性原则 19．作为进步教育的代表，杜威提出的基本主张包括()等。A．学校即社会 B．教育即生活 20．教学方法、教学手段及组织形式主要受( )的制约。E．生产力发展水平 21．以语言传递为主的教学方法包括( )。 A．讲授法 B。谈话法C．讨论法 D．读书指导法