Intelligent Strategies for Meta Multiple Criteria Decision Making Book

	List of Figures	ix
	List of Tables	xi
	Preface	xiii
	Foreword	xvii
1.	Introduction	1
1.	MCDM problems	1
2.	Solutions of MCDM problems	4
3.	Decision processes and the application of MCDM methods	5
4.	Concepts of 'correct' decision making in MCDM methods	9
5.	Summary and conclusions	14
2.	The Meta Decision Problem in MCDM	15
1.	Methodological criticism in MCDM	15
1.1	Criticism on single concepts and methods	15
1.2	The discussion on the descriptive orientation of MCDM	19
1.3	Foundations by axioms of rational behavior	22
2.	The meta decision problem in MCDM	24
2.1	Formulation and foundation of the problem	24
2.2	Criteria for method selection	25
2.2.1	The suitability for a type of problem	25
2.2.2	Criteria based on solution concepts	26
2.2.3	Criteria oriented towards implementation	28
2.2.4	Criteria based on the specific decision situation	30
2.3	Scalar and multicriteria meta decision problems	31
2.3.1	Scalar evaluations of MCDM methods	31
2.3.2	Method choice as an MADM problem	32
2.4	The meta decision problem as a problem of method design	34
2.4.1	Determining the parameters of an MCDM method	34
2.4.2	Formalization of MCDM methods	36
2.4.3	A parameter optimization model	37
2.5	The problem of information acquisition	39
2.5.1	Implicit information	40
2.5.2	Explicit information	41
3.	Summary and conclusions	44
3.	Neural Networks and Evolutionary Learning for MCDM	47
1.	Neural networks and MCDM	47
1.1	Introduction	47
1.2	The construction of neural networks working as traditional MCDM methods	49
1.3	Neural networks as an adaptive MCDM method	54
2.	Evolutionary learning	55
2.1	Evolutionary algorithms and neural networks	56
2.2	Evolutionary algorithms and MCDM	59
3.	Summary and conclusions	61
4.	On the Combination of MCDM Methods	63
1.	Introduction	63
2.	Properties of MCDM methods	69
3.	Properties of specific MCDM methods	71
4.	Properties of neurons and neural networks	73
5.	The combination of algorithms	74
6.	Neural MCDM networks	75
7.	Termination and runtime of the algorithm	76
8.	Summary and conclusions	77
5.	Loops--An Object Oriented DSS for Solving Meta Decision Problems	79
1.	Preliminary remarks	79
2.	Method integration, openness, and object oriented implementation	80
3.	A class concept for LOOPS	84
4.	Problem solving and learning from an object oriented point of view	84
5.	MADM methods in LOOPS	87
6.	Neural networks in LOOPS	89
7.	Neural MCDM networks in LOOPS	90
8.	Evolutionary algorithms in LOOPS	91
9.	An extended interactive framework	95
10.	Summary and conclusions	98
6.	Examples of the Application of LOOPS	99
1.	Some remarks on the application of LOOPS	99
2.	The learning of utility functions	100
3.	Stock selection	106
4.	Stock price prediction and the learning of time series	113
5.	Stock analysis and long-term prediction	121
6.	Method learning	124
7.	Meta learning	127
8.	An integrated proposal for the application of LOOPS	131
9.	Summary and conclusions	132
7.	Critical Resume and Outlook	135
	References	141
	Appendices	162
A	Some basic concepts of MCDM theory	163
1.	Relations	163
2.	Efficiency concepts and scalarizing theorems	165
3.	Utility concepts and other axiomatics	166
B	Some selected MCDM methods	169
1.	Simple additive weighting	169
2.	Achievement levels	169
3.	Reference point approaches	170
4.	The outranking method Promethee	171
C	Neural networks	173
1.	Introduction to neural networks	173
2.	Neural networks for intelligent decision support	178
D	Evolutionary algorithms	181
1.	Introduction to evolutionary algorithms	181
2.	The generalization of evolutionary algorithms	186
E	List of symbols	189
F	List of abbreviations	193
	Index	195