|
1. |
Numerical solution for a class of open-loop differential games via the shifted Chebyshev polynomial approximation |
|
International Journal of Systems Science,
Volume 20,
Issue 2,
1989,
Page 171-179
JIANG JIONG,
ZHU SHAN-AN,
Preview
|
PDF (105KB)
|
|
摘要:
Shifted Chebyshev polynomial functions are used to solve a class of open-loop differential games described by linear state dynamics and quadratic objective functions. Using linear transformation of the variables, the necessary conditions that the open-loop differential games must satisfy are reduced to functional differential equations. The solutions of the functional equations are then obtained in a series of shifted Chebyshev polynomial approximations. The computational procedure is simple and available for computer implementation. It is also shown that the computational results provide satisfactory accuracy compared with those of the exact solutions.
ISSN:0020-7721
DOI:10.1080/00207728908910117
出版商:Taylor & Francis Group
年代:1989
数据来源: Taylor
|
2. |
Information-theoretic approach to classifying operators in conveyor systems |
|
International Journal of Systems Science,
Volume 20,
Issue 2,
1989,
Page 181-189
IKUO ARIZONO,
HIROSHI OHTA,
NORIYUKI NISHIYAMA,
Preview
|
PDF (112KB)
|
|
摘要:
It is important to select standard operators for the purpose of the smooth working of a conveyor system designed using standard time. An information-theoretic procedure is presented that is based on the Kullback-Leibler information for classifying operators in conveyor systems. The recommended procedure is compared with the statistical procedure, and is confirmed to be practical and relatively effective.
ISSN:0020-7721
DOI:10.1080/00207728908910118
出版商:Taylor & Francis Group
年代:1989
数据来源: Taylor
|
3. |
Doubling algorithm for continuous-time algebraic Riccati equation |
|
International Journal of Systems Science,
Volume 20,
Issue 2,
1989,
Page 191-202
MORISHIGE KIMURA,
Preview
|
PDF (175KB)
|
|
摘要:
A second-order convergent algorithm is presented which gives an approximation to the unique positive definite solution of the continuous-time algebraic Riccati equation (CARE). First the CARE is transformed into the discrete-time algebraic Riccati equation (DARE) using the transformation given by Hitz and Anderson (1972). Then the discrete-time doubling algorithm, whose initial values are expressed in forms suitable for computation, is applied to the DARE. Next, it is shown that this algorithm is convergent under the condition that the CARE has the unique positive definite solution. Finally an ‘inverse’ of the Hitz and Anderson (1972) transformation is presented, which transforms the DARE into the CARE. It is proved that the ‘inverse’ transformation preserves the stabilizability, controllability, detectability and observability of the DARE.
ISSN:0020-7721
DOI:10.1080/00207728908910119
出版商:Taylor & Francis Group
年代:1989
数据来源: Taylor
|
4. |
Measuring economic efficiency with stochastic input-output data† |
|
International Journal of Systems Science,
Volume 20,
Issue 2,
1989,
Page 203-213
JATIK. SENGUPTA,
Preview
|
PDF (182KB)
|
|
摘要:
Efficiency comparisons of two or more clusters under alternative information structures are attempted here using a new interpretation of the data envelopment analysis model. Mixed strategies and information theoretic distance measures are used here for the efficiency comparisons.
ISSN:0020-7721
DOI:10.1080/00207728908910120
出版商:Taylor & Francis Group
年代:1989
数据来源: Taylor
|
5. |
Mixed strategy and information theory in optimal portfolio choice |
|
International Journal of Systems Science,
Volume 20,
Issue 2,
1989,
Page 215-227
JATIK. SENGUPTA,
Preview
|
PDF (205KB)
|
|
摘要:
By formulating the portfolio choice problem as a two-person zero-sum game, it is shown that a saddle point exists in mixed strategies where the pay-off function includes entropy as an information measure. The search for optimal strategies (pure or mixed), when the probabilities of states of nature are unknown leads to a whole new field of research which can be related to the recent theory of capital market efficiency.
ISSN:0020-7721
DOI:10.1080/00207728908910121
出版商:Taylor & Francis Group
年代:1989
数据来源: Taylor
|
6. |
Optimal (n,m,N*) series-parallel systems with fixed repair facility |
|
International Journal of Systems Science,
Volume 20,
Issue 2,
1989,
Page 229-233
N. VENUGOPAL,
Y. KRISHNA REDDY,
Preview
|
PDF (80KB)
|
|
摘要:
Complex reliability systems have an important role in sophisticated systems like mass communication and computer-aided networks. Optimization methods for these systems thus have a large scope and potential. A procedure leading to the optimal (n, m, N*) series parallel systems with provision for a fixed repair facility is developed. The three entries respectively denote the fixed repair‐stage n, the number of units in the series m, and the optimal number of parallel units N* in the system. Under certain conditions and based on minimal cost considerations, the optimal N* is obtained. An illustration, specializing to a negative exponential law governing the stochastic failures, is provided.
ISSN:0020-7721
DOI:10.1080/00207728908910122
出版商:Taylor & Francis Group
年代:1989
数据来源: Taylor
|
7. |
New results on stochastic systems excited by white noise powers |
|
International Journal of Systems Science,
Volume 20,
Issue 2,
1989,
Page 235-250
GUY JUMARIE,
Preview
|
PDF (205KB)
|
|
摘要:
Many systems of practical interest are described by stochastic differential; equations excited by powers of white noise, which can be either linear or non-linear. It is shown that, in some instances of importance, these dynamics can be meaningfully analysed by means of the central limit theorem, and moreover can be reduced to models driven by the Itô stochastic differential equations. The technique is described at length for a one-dimensional system and some applications are given, for instance, for stochastic systems with multiplicative white noises. Then the simple linear oscillator excited by gaussian white noise powers is considered, and finally, the method is applied to a special class of multivariable systems. The approach is quite consistent with the maximum entropy principle.
ISSN:0020-7721
DOI:10.1080/00207728908910123
出版商:Taylor & Francis Group
年代:1989
数据来源: Taylor
|
8. |
Optimal flow control of aG/G/1 queue |
|
International Journal of Systems Science,
Volume 20,
Issue 2,
1989,
Page 251-265
D. D. KOUVATSOS,
A. T. OTHMAN,
Preview
|
PDF (208KB)
|
|
摘要:
The problem of the optimal flow control of a G/G/l queue at equilibrium is investigated by approximating the general (G-type) distributions by a maximum entropy model with known first two moments. It is shown that the flow control mechanism maximizing the throughput, under a bounded time delay criterion, is of window type (bang-bang control). The maximum number of packets in transit within the system (i.e., sliding window size) is derived in terms of the maximum allowed average time delay. Furthermore, the relationship between the maximum throughput and maximum time delay is determined. Numerical examples provide useful information on how critically the optimal throughput is affected by the distributional form of the interarrival and service patterns and a conjecture on performance bounds is made.
ISSN:0020-7721
DOI:10.1080/00207728908910124
出版商:Taylor & Francis Group
年代:1989
数据来源: Taylor
|
9. |
Some replacement policies with minimal repairs and repair cost limit |
|
International Journal of Systems Science,
Volume 20,
Issue 2,
1989,
Page 267-279
P. K. KAPUR,
R. B. GARG,
N. L. BUTANI,
Preview
|
PDF (141KB)
|
|
摘要:
Some extended replacement policies based on the number of failures, incorporating the concept of repair cost limit are discussed. Three models are considered as follows: (a) a unit is replaced at the nth failure, or when the estimated minimal repair cost exceeds a particular limitc; (b) a unit has two types of failures and is replaced at the nth type 1 failure, or type 2 failure, or when the estimated repair cost of type 1 failures exceeds a predetermined limit c—type 1 failures are minimal; failures, type 2 failures are catastrophic failures and both occur with constant probability; (c) a unit has two types of failures and the type 1 and type 2 failures are age dependent—the unit is replaced at the nth type 1 failure, type 2 failure, or when the estimated repair cost due to type 1 failures exceeds a predetermined limit c. Introducing costs due to replacements, inspections, and minimal repairs, an optimal number of minimal repairs before replacement is obtained, which minimizes the expected cost rate. Some particular cases are also derived. Finally, the application of these models to computer science is discussed.
ISSN:0020-7721
DOI:10.1080/00207728908910125
出版商:Taylor & Francis Group
年代:1989
数据来源: Taylor
|
10. |
Algorithm for minimal realization of linear multivariable systems |
|
International Journal of Systems Science,
Volume 20,
Issue 2,
1989,
Page 281-295
O. A. SEBAKHY,
M. EL SINGABY,
Preview
|
PDF (161KB)
|
|
摘要:
Based on an iterative algorithm developed by the authors for computing the invariant zeros of linear systems as eigenvalues, a new algorithm for minimal realization of linear multivariable control systems is presented. This method of realization gives new insight into the problem of constructing a linear multi-variable system by using its pole-zero characterization. Two numerical examples are presented to show the feasibility of the proposed method.
ISSN:0020-7721
DOI:10.1080/00207728908910126
出版商:Taylor & Francis Group
年代:1989
数据来源: Taylor
|
|