JID:FSS AID:6466 /FLA [m3SC+; v 1.184; Prn:4/02/2014; 8:27] P.

1 (1-8)
Available online at www.sciencedirect.com

ScienceDirect
Fuzzy Sets and Systems ••• (••••) •••–•••
www.elsevier.com/locate/fss

Fuzzy optimization: Milestones and perspectives

M.K. Luhandjula
Department of Decision Sciences, University of South Africa, PO Box 392, UNISA, Pretoria 0003, South Africa

Abstract
We provide a condensed and selective look at the known landscape of the theory and application of Fuzzy optimization, empha-
sizing Professors’ Tanaka and Asai contribution. Significant ideas are picked out in the trajectory of this field and projections are
made for its future developments. From our discussions, it unquestionably emerges that the above mentioned Japanese academics
have played a noticeable role in shaping the subject from its genesis to its maturation. Moreover, their research work will continue
to inspire prospective works in this field.
© 2014 Published by Elsevier B.V.

Keywords: Optimization; Mathematical programming; Uncertainty; Fuzzy sets

1. Introduction

Optimization is a very old and classical area which is of high concern to many disciplines. Engineering as well as
Management, Politics as well as Medicine, Artificial Intelligence as well as Operations Research and many other fields
are in one way or another concerned with optimization of designs, decisions, structures or information processes.
In addition to this, many concrete problems that may be cast into an optimization setting are rife with sources of
imprecision. Without any claim for exhaustivity, we may mention: uncertainty related to errors, sparsity of data,
subjectivity of expert judgements and ambiguity inherent to the range of parameter values. In this connection, the
noted philosopher Nietzche was quoted as saying: “No one is gifted with immaculate perception.” Moreover, false
certainty is bad science and it could be dangerous if it stunts articulation of critical choices.
It is well known that, in order to tackle a difficult problem, an algorithm has to incorporate all the available infor-
mation. Neglecting some inherent feature like imprecision would, according to the Computer Science rule “garbage
in, garbage out”, leave no other chance to the algorithm but to churn out meaningless outcomes.
The foregoing clearly indicates the persistent need to improve optimization models realism by making it possible to
incorporate uncertainty into mathematical programming frameworks. This has given rise to the realm of Mathematical
programming under uncertainty.
Mathematical disciplines like Stochastic programming, Robust optimization, Grey programming, Interval program-
ming, hybrid Rough set–Particle swarm optimization have arisen from the above mentioned need.

E-mail address: [email protected].

0165-0114/$ – see front matter © 2014 Published by Elsevier B.V.

http://dx.doi.org/10.1016/j.fss.2014.01.004
JID:FSS AID:6466 /FLA [m3SC+; v 1.184; Prn:4/02/2014; 8:27] P.2 (1-8)
2 M.K. Luhandjula / Fuzzy Sets and Systems ••• (••••) •••–•••

This paper considers a more recent approach to Optimization under uncertainty in which imprecision is modeled by
fuzzy relations and/or fuzzy parameters. It briefly surveys part of the known landscape of the theory and application
of Fuzzy optimization.
In such a short span of time and space, it is virtually an impossible task to do justice to the whole field of Fuzzy
optimization. We have then, deliberately, chosen to bring across some glimpses with emphasis to Professors’ Tanaka
and Asai contribution.
We take a more discursive approach, reflecting more on ideas rather than on technical details.
For a detailed picture of Fuzzy optimization, we refer the reader to several textbooks [1–3] and to references
therein. An interested reader may also consult authoritative overviews by Inuiguchi, Ichihashi and Tanaka [4], Rom-
melfanger [5], Inuiguchi and Ramik [6] and Delgado et al. [7].

2. Flexible programming

In solving their problems, Decision makers generally grapple with technological, environmental and competitive
factors which interact in a complicated fashion. In such a turbulent environment, the formulation of the problem in
terms of dichotomous “yes or no” statements yields often inconsistencies which are expressed by the vacuousness of
the feasible set.
More than this, a Decision maker may feel more comfortable formulating his optimization problem using expres-
sions of the “more or less” type.
The foregoing gives rise to the need for models that accept leeways on the achievement of the goal and the con-
straints in an optimization setting.
Fuzzy set theory is of particular importance in the formulation and solution of optimization problems, with relaxed
goals and constraints. As a matter of fact, such flexible goals and constraints may be properly modeled using the
language of this theory.
Making use of Bellman–Zadeh’s confluence principle [8], one may single out a satisfying solution in this context.
It is worth mentioning that in such a turbulent environment, the “optimum optimorum” no longer exists hence the
term “satisfying solution” suits better than “optimal solution”.
The above outlined approach has a long and active history dating back to the seminal paper by Tanaka, Okuda and
Asai [9] in mid-1970’s. In that paper a mathematical program with a fuzzy objective function and fuzzy constraints
was clearly formulated. Authors then exploited, with good reason, properties of α-level sets of involved fuzzy sets to
handle, theoretical, algorithmic and practical issues related to the fuzzy mathematical problem under scrutiny.
This line of research was further developed by the same authors [10,11] and by Zimmermann [12], giving birth to
the subfield of Fuzzy Optimization called Flexible programming.
These research works coupled with advances in Computing Technology sparked a massive flurry of interest in
Flexible programming in the 1980’s. This decade together with the next one saw also successful applications of Flex-
ible programming across a wide spectrum of domains including: water resource management, air-pollution reduction
and media selection.
It is worth mentioning that the main idea behind Flexible programming is to consider all goals and constraints
as flexible constraints. The use of the minimum operator, instead of the sum, enforcing every constraint or goal to
be somewhat respected. As a consequence, one may consider Flexible Optimization as pioneering soft constraint
satisfaction in Artificial Intelligence (see e.g. [13,14]).
This idea of symmetry between the goal and the constraints of a mathematical program, that is used in an essential
manner in Flexible programming, is not to everybody’s taste [15]. Some researchers are against this symmetric ap-
proach. They accept the idea of representing soft goals and constraints by appropriate fuzzy sets but they stop short
of taking the additional leap of considering the goal and the constraints as identical concepts. They advocate that
constraints should delineate the feasible set while the goal should remain on its role of ranking feasible alternatives.

3. Fuzzy robust programming

Another Fuzzy optimization model that came to the fore during the 1970’s is the fuzzy robust one [16]. Instead
of specifying the feasibility of an action by a set of inequalities, like in conventional optimization, in Fuzzy robust
JID:FSS AID:6466 /FLA [m3SC+; v 1.184; Prn:4/02/2014; 8:27] P.3 (1-8)
M.K. Luhandjula / Fuzzy Sets and Systems ••• (••••) •••–••• 3

programming the feasible region is defined via fuzzy set containment. A Fuzzy robust model can then be viewed as a
generalization of Soyster’s optimization model with set-inclusive constraints [17].
The study of Fuzzy robust model is essentially based on exploring the existing lattice isomorphism between the set
of fuzzy sets and the family of their α-level sets [18].
This isomorphism permits to convert the Fuzzy robust model into a crisp one. An important feature of the frame-
work is that, except for some particular cases, involving restrictive hypotheses, the resulting deterministic problem is
not computationally tractable. The only one resort is then to approximate involved fuzzy sets by crisp ones (in the
Chebycheff sense for instance), in a way to obtain a deterministic counterpart that can be tackled efficiently.
It is worth noticing that apart from the above described Fuzzy robust model, some variants have been developed.
The most common is the one where inclusion in the constraints is replaced by flexible inequality.
The modeling power of Fuzzy robust optimization is undisputable. Moreover, important theoretical results in con-
nection with the Fuzzy robust optimization model have been obtained through the Fuzzy set representation theorem.
Nevertheless, the Fuzzy robust programming model has neither the computational attractiveness nor the broad appli-
cability of its Flexible programming counterpart. In spite of this, efforts have been devoted to exploit ideas from robust
programming in a way to cope with possibilistic constraints (see e.g. [19]) and to deal with notions of optimality and
efficiency in a possibilistic optimization setting [20–23].
Attempts of rejuvenation of this subfield of Fuzzy Optimization is recently seen in the literature [24], with the goal
of propelling it into altogether new directions.

4. Mathematical programming with fuzzy data – possibilistic programming

A common paradigm in application of optimization models is that all involved parameters are fixed known data. In
many practical situations, such an assumption turns out to be unreliable.
Substantial progress on analysis and arithmetic of fuzzy quantities, through Zadeh’s extension principle, in the
1980s and 1990s [25,26], served as a catalyst to an important milestone on the Fuzzy Optimization trajectory. Namely,
the drawn up of theoretical insights, description of methodological approaches, development of exact algorithms,
heuristics and metaheuristics along with applications in a wide spectrum of domains, in the subfield of mathematical
programming with fuzzy parameters.
Undoubtedly, the study of mathematical programming with fuzzy data epitomized many important features and
mathematical difficulties that arise in the field of Fuzzy optimization.
It is worth to note that when, in an optimization setting, a possibility interpretation of involved fuzzy quantities is
at hand, the term possibilistic programming is preferred to that of Fuzzy optimization.
Here again the optimum optimorum does not exist and one has to resort to the bounded rationality principle for
solution concepts.
Among theoretical results obtained we may mention the formulation of solution concepts and their mathematical
characterization, the transformation of a possibilistic program into deterministic terms, useful insights on the stability
of an optimization problem with fuzzy parameters.
The fundamental idea underlying extant methods for dealing with possibilistic optimization problems is to seek to
immunize the solution, in some sense, from the uncertainty.
The most commonly used approaches include methods based on deffuzification of involved fuzzy quantities, tech-
niques relying on comparison of fuzzy numbers, methods based on exploration of possibility and necessity measures.
A formidable achievement of Tanaka and Asai is to have made pioneering and fundamental contribution to several
of these developments.
In [27,28] they magnificiently showed how to deal with a mathematical program with an interval objective function.
This approach is often used on deffuzification methods based on α-level cuts of involved fuzzy quantities. Their
paper [29] is devoted to how to deal with a linear programming problem with fuzzy numbers by comparing involved
fuzzy quantities. These articles may be regarded as precursors of approaches based on exploring comparisons between
fuzzy quantities. In [30] they used possibility and necessity measures for pattern classification purposes. This paper
has influenced the use of these two measures in Decision making in general and in Optimization in particular.
Readers interested on these issues might refer to the following early papers co-authored by Tanaka [31,32].
Beyond the standard formulation of mathematical programming problems with fuzzy parameters, a number of
variants have been studied. For instance, a mathematical program with fuzzy coefficients and flexible relationships.
JID:FSS AID:6466 /FLA [m3SC+; v 1.184; Prn:4/02/2014; 8:27] P.4 (1-8)
4 M.K. Luhandjula / Fuzzy Sets and Systems ••• (••••) •••–•••

Tanaka and Asai [33] have also pondered the relevant issue of value of information in a fuzzy optimization model.
They have rigorously established that if the fuzziness on coefficients of a fuzzy linear program is reduced by using
information about these coefficients, a more satisfying solution is obtained.
It is natural that a solution of a Fuzzy mathematical program be fuzzy to reflect the fuzzy nature of the problem.
Tanaka and Asai have pioneered the formulation and the investigation of properties of fuzzy solutions for Fuzzy
optimization problems [34]. Their approach, based on the max–min strategy represents a step in the right direction for
this crucial problem. Another spectacular achievement of Tanaka, in this regard, is that of having obtained powerful
algorithms to get the joint possibility distribution of decision variables of a Possibilistic program [35].
Applications of Possibilistic programming are meaningful, interesting and easy to grasp; diet formulation, opti-
mization of project networks, reliability problems, manufacturing cell formation and the list is endless.

5. Extensions

Ideas discussed in previous sections have been extended in four important directions. First, attempts have been
made to incorporate simultaneously randomness and fuzziness in an optimization framework. Second, efforts have
been devoted to take the best of Fuzzy Optimization in a way to come to grips with complexity inherent to the
presence of several (deterministic or fuzzy) objective functions. Third, attention has been paid to handle multistage
processes under fuzziness. Fourth, the methodology of Fuzzy mathematical programming has also been applied to
multilevel optimization settings.
In this section, we offer a glimpse at these generalizations of the basic Fuzzy Optimization models.

5.1. Fuzzy Stochastic Optimization

In recent years, the scientific community has increased his interest on mathematical programming under fuzziness
and randomness (see e.g. the books [36,37] and references therein). This trend can be attributed to two converging
factors:
(a) the availability of powerful tools that blend ideas from Probability theory and Fuzzy set theory (probability of a
fuzzy event, probabilistic set, fuzzy random variables, random fuzzy variables, to mention but a few);
(b) the need, in solving concrete real-life problems, to base decision on information which is both fuzzily imprecise
and probabilistically uncertain.
An important milestone in this effort to tackle optimization problems involving both randomness and fuzziness is the
publication of the paper [38] by Tanaka and Asai.
This paper offered useful insights into the question of how to combine the two kinds of imprecision in a decision
making context. It is also worth mentioning that Tanaka and Asai research on fuzzy regression [39] and fuzzy data
analysis [40] also shed light on mathematical tools for taking an intellectual step towards situations where fuzziness
and randomness are in the state of affairs.
Indeed, fuzzy regression and fuzzy data analysis are nothing but processes of integrating probability and possibility
information under one roof.
Building on the above mentioned hybrid tools for combining fuzziness and randomness, several fuzzy stochastic
models have been introduced: Flexible stochastic optimization model, mathematical programming model with fuzzy
random or random fuzzy coefficients, optimization problem with random variables and fuzzy numbers, mathematical
programming problems with random variables having fuzzy parameters, etc.
Techniques for solving these models have also been described and algorithms ranging from simple ones (Sim-
plex, Gradient descent) to complex ones (Hybrid intelligent algorithms) via metaheuristics (Genetic algorithms, Tabu
search, Simulated annealing) have been pushed forward.
Moreover, many applications of the Fuzzy stochastic optimization methodology including financial, industrial,
marketing resource, portfolio selection [41] are reported in the literature.
An interested reader is referred to [42,43] and in references therein for more detail in Fuzzy Stochastic Optimiza-
tion.
JID:FSS AID:6466 /FLA [m3SC+; v 1.184; Prn:4/02/2014; 8:27] P.5 (1-8)
M.K. Luhandjula / Fuzzy Sets and Systems ••• (••••) •••–••• 5

5.2. Fuzzy multiobjective programming

First and foremost, Flexible programming (see Section 2) has provided a gateway to approaches for tackling de-
terministic multiobjective programming problems [44]. This is basically done by viewing objective functions of the
considered multiobjective problem as soft goals and then take advantage of the computational convenience offered by
Flexible programming. If membership functions of soft goals as well as the operator used to represent the semantic
meaning of the connective “and” are judiciously chosen, then the solution obtained by this way is Pareto optimal for
the original multiobjective program.
A very important part of research effort in this realm has been the generalization of ideas for dealing with the single
objective possibilistic programming to the multiobjective case.
Many researchers took up the challenge of extending the standard notion of Pareto optimality [45] before describing
algorithms for singling out the sought solution.
It is worth mentioning that Murata, Ishibuchi and Tanaka were among the first to cope with multiobjective pro-
gramming problems using metaheuristics [46].
Ongoing research include approximation of involved fuzzy quantities by their nearest interval approximations, the
use of Embedding theorem of fuzzy numbers to obtain equivalent deterministic counterparts and the exploration of
gH-differentiability to obtain a solution through Karush–Kuhn and Tucker conditions for Pareto optimality [47].
Fuzzy approaches for deterministic and possibilistic multiobjective programming constitutes a research field which
is of high practical relevance and which is still evolving dynamically.

5.3. Fuzzy dynamic programming

It is well known that the two key dimensions of information in an optimization context are its quality and its
evolution. In this paper, we assume that available information is not of good quality as it is tainted with vagueness.
Regarding the second dimension, we suppose, in this section, that available information may change during the deci-
sion process. This calls for another approach that deemphasizes the statical nature of the decision problem in favor of
the dynamical one. A strand of the literature on Mathematical programming under fuzziness has pursued this dynami-
cal avenue. Efforts have been devoted to making standard Dynamic programming model more realistic by a relaxation
of the often artificial assumptions of precision as to the goal, constraints, state, etc.
For a review of basic problem classes and developments of fuzzy dynamic programming, which is a powerful tool
for dealing with multistage decision making and optimization problems under fuzziness, we refer the reader to the
book by Esogbue and Kacprzyk [48]. An interested reader may also consult [49] where thorny issues related to Fuzzy
dynamic programming, like its computational efficiency, are addressed. The methodology of multistage programming
under fuzziness has applications in a vast spectrum of domains including: Optimization of Regional Structures, Water
resource allocation and Risk Analysis.
Impact of Bellman, Zadeh, Tanaka and Asai on the development of Fuzzy dynamic programming is undeniable.
In [8], Bellman and Zadeh provided a general and intuitive principle for both static and dynamic decision making
under fuzziness. Tanaka and Asai exploited this principle to cope with a Fuzzy mathematical program, mentioning
explicitly, for the first time, a sequential decision problem under fuzziness namely, a fuzzy control problem [9].

5.4. Fuzzy Multilevel Optimization

Another important landmark in the evolution of Fuzzy optimization is the publication of research work on Fuzzy
multilevel programming. In related papers, authors made good use of Fuzzy optimization to deal with decentralized
planning problems with multiple decision makers in a hierarchical organization.
Hierarchical decision making process is extremely practical to such decentralized systems as agriculture, govern-
ment policy, economic systems, finance, etc.
It is part of the Multilevel programming (MLP) folklore, repeatedly observed in practice, that most existing method-
ologies proposed for solving MLP problems turn out to be computationally inefficient [50]. This unfortunate situation
has lead some researchers to deviate from the traditional path based on concepts of vertex enumeration to approach
the problem through Flexible programming paradigm.
JID:FSS AID:6466 /FLA [m3SC+; v 1.184; Prn:4/02/2014; 8:27] P.6 (1-8)
6 M.K. Luhandjula / Fuzzy Sets and Systems ••• (••••) •••–•••

An interested reader may consult [51] where a fuzzy approach for bilevel programming problems is outlined.
Generalization of this approach to problems with greater levels may be found elsewhere [52]. A step up the ladder
of the above mentioned approach for MLP problems is its extension to the case of Multilevel optimization problems
with fuzzy data [53].
Solution concepts, including the Stackelberg one, have been elaborated and interesting theoretical results have been
obtained. Furthermore powerful algorithms for singling out desired solutions have been pushed forward.
Another issue that researchers are busy pondering is that of tailoring the above outlined fuzzy approaches for
deterministic and Fuzzy multilevel programming to the case where each decision maker has several objective func-
tions [54].

6. Concluding remarks and perspectives

We have indicated important milestones in the development of Fuzzy optimization emphasizing the macro-level
view and highlighting Tanaka and Asai contribution. It clearly surfaces, from our discussion, that tangible progress
has been made in this field. Moreover, by their rich contribution, Tanaka and Asai have helped raise the intellectual
respectability of Fuzzy optimization. By analyzing the current state of the art of Fuzzy mathematical programming,
some directions can be drawn for future research in this field.

– Complementarity and synergy between the constituent methodologies of mathematical programming under un-
certainty (namely, Stochastic optimization, Robust programming, Grey programming, Fuzzy optimization) should
be fully exploited to achieve higher levels of capability in the conception design and utilization of uncertainty, in
an optimization framework. To assert that it is more useful to conceive imprecision in an optimization setting as
a variegated whole, is not to minimize importance of research works that have been done in specific areas. It is
instead to assert that new perspective may be gained by integration of both approaches rather than exclusion.
– It is well known that Decision makers are characterized by narrow-mindedness that leads them to turn uncertain-
ties into certainties and ambiguities into clarity. This should be done without caricaturing the reality. So we find it
wise that corrective actions be taken for Fuzzy optimization techniques, that yield crisp solutions without striking
a balance between efficiency and effectiveness. Moreover ways for generating fuzzy solutions, when crisp solu-
tions are questionable, should be investigated. Tanaka and Asai [34] took the challenge to compute fuzzy solutions
of some specific problems. More is needed for the general case, mostly in terms of mathematical characterization
and algorithmic search.
– Decision makers would perhaps best be served if from a deep comparative analysis between main Fuzzy mathe-
matical programming methods, a user-friendly Decision Support System (DSS) could be pushed forward. Such
a DSS would be of great help advising a Decision maker with the choice of the most suitable method for his
problem.

It is worth mentioning that the above mentioned perspectives reflect just our own point of view. We do not claim that
they are the only possible directions. More than this, there are clearly other aspects of Tanaka and Asai’s contribution
on the field of Fuzzy optimization that are not covered in this paper. It is our hope, however, that all these ideas
will continue to trigger research work on Fuzzy optimization so that the vision of its powerful role as projected by
Professor Tanaka in a foreword of a special issue of the Journal of Fuzzy optimization and Decision making [54]
becomes a reality.

References

[1] W.A. Lodwik, Fuzzy Optimization, Recent Advances and Applications, Springer, 2010.
[2] C. Carlsson, R. Fullér, Fuzzy Reasoning in Decision Making and Optimization, Studies in Fuzziness and Soft Computing, Physica-Verlag,
2002.
[3] M. Sakawa, Genetic Algorithms and Fuzzy Multiobjective Optimization, Kluwer Academic Publishers, Boston, 2001.
[4] M. Inuiguchi, H. Ichihashi, H. Tanaka, Fuzzy programming: a survey of recent developments, in: R. Slowinski, J. Teghem (Eds.), Stochastic
Versus Fuzzy Approaches to Multiobjective Mathematical Programming Under Uncertainty, Kluwer Academic Publishers, Dordrecht, 1990,
pp. 45–68.
[5] H. Rommelfanger, Fuzzy linear programming and applications, Eur. J. Oper. Res. 92 (1996) 512–527.
JID:FSS AID:6466 /FLA [m3SC+; v 1.184; Prn:4/02/2014; 8:27] P.7 (1-8)
M.K. Luhandjula / Fuzzy Sets and Systems ••• (••••) •••–••• 7

[6] M. Inuiguchi, J. Ramik, Possibilistic linear programming: a brief review of fuzzy mathematical programming and a comparison with stochastic
programming in portfolio selection problem, Fuzzy Sets Syst. 111 (2000) 3–28.
[7] M. Delgado, J.L. Verdegay, M.A. Vila, Relating different approaches to solve linear programming problems with imprecise costs, Fuzzy Sets
Syst. 37 (1) (1990) 33–42.
[8] R. Bellman, L.A. Zadeh, Decision making in a fuzzy environment, Manag. Sci. 17 (1970) 141–164.
[9] H. Tanaka, T. Okuda, K. Asai, On fuzzy mathematical programming, J. Cybern. 3 (4) (1974) 37–46.
[10] H. Tanaka, T. Okuda, K. Asai, Decision making and its goals in a fuzzy environment, in: L.A. Zadeh, et al. (Eds.), Fuzzy Sets and Their
Applications to Cognitive and Decision Processes, 1975, pp. 257–277.
[11] H. Tanaka, T. Okuda, K. Asai, Decision making in fuzzy environment: fuzzy information and decision making, Int. J. Prod. Res. 15 (60)
(1977) 623–635.
[12] H.J. Zimmermann, Description and optimization of fuzzy systems, Int. J. Gen. Syst. 2 (1976) 209–215.
[13] D. Dubois, H. Fargier, H. Prade, Possibility theory in constraint satisfaction problems: Handling priority, preference and uncertainty, Appl.
Intell. 6 (1996) 287–309.
[14] P. Meseguer, F. Rossi, T. Schiex, Foundations of Artificial Intelligence, vol. 2, 2006, pp. 281–328 (Chapter 9).
[15] J.L. Verdegay, Fuzzy mathematical programming, in: M.M. Gupta, E. Sanchez (Eds.), Fuzzy Information and Decision Processes, 1982,
pp. 231–237.
[16] C.V. Negoita, Management Applications of System Theory, Birkhäuser, 1975.
[17] A.L. Soyster, Convex programming with set constraints and application to inexact programming, Oper. Res. 21 (1973) 1154–1157.
[18] C.V. Negoita, D.A. Ralescu, On fuzzy optimization, Kybernetes 6 (3) (1976) 193–195.
[19] X. Chen, J. Li, Robust Fuzzy optimization, in: Proceedings of the International Conference on Computer Science and Information Technology,
Singapore, 2008, pp. 911–914.
[20] M. Inuiguchi, H. Ichihashi, Y. Kume, Relationships between modality constrained programming problems and various fuzzy mathematical
programming problems, Fuzzy Sets Syst. 49 (1992) 243–259.
[21] M. Inuiguchi, M. Sakawa, Robust Optimization under softness in a fuzzy linear programming problem, Int. J. Approx. Reason. 18 (1998)
21–34.
[22] M. Inuiguchi, T. Tanino, H. Tanaka, Optimization approaches to possibilistic linear programming problems, in: Proceedings of Joint 9th IFSA
World Congress and 20th NAFIPS International Conference, 2001, pp. 2724–2729.
[23] M. Inuiguchi, Robust optimization by fuzzy linear programming, managing safety of heterogenous systems, in: Y. Ermoliev, M. Makowski,
K. Marti (Eds.), Decisions Under Uncertainty and Risks, in: LNEMS, vol. 658, Springer-Verlag, Berlin, 2012, pp. 219–239.
[24] M. Timonin, Robust optimization of the Choquel integral, Fuzzy Sets Syst. 213 (2012) 27–46.
[25] D. Dubois, H. Prade, Fuzzy Sets and Systems: Theory and Applications, Academic Press, 1980.
[26] J. Ramik, J. Rimanek, Inequality relation between fuzzy numbers and its use in fuzzy optimization, Fuzzy Sets Syst. 16 (1985) 123–138.
[27] H. Ishibuchi, H. Tanaka, Formulation and analysis of linear programming with interval coefficients, J. Jpn. Ind. Manage. Assoc. 40 (1989)
320–329.
[28] H. Ishibuchi, H. Tanaka, Multiobjective programming in optimization of the interval objective function, Eur. J. Oper. Res. 48 (1990) 219–225.
[29] H. Tanaka, K. Asai, Fuzzy linear programming with fuzzy numbers, Fuzzy Sets Syst. 13 (1984) 1–10.
[30] H. Ishibuchi, R. Fujioka, H. Tanaka, Possibility and necessity pattern classification using neural networks, Fuzzy Sets Syst. 48 (3) (1992)
331–340.
[31] M. Inuiguchi, H. Ichihashi, H. Tanaka, Decision procedures based on modal concept and its application to fuzzy multi-objective linear pro-
gramming problem, J. Oper. Res. Soc. Jpn. 30 (1987) 449–471 (in Japanese).
[32] M. Inuiguchi, H. Ichihashi, H. Tanaka, Possibilistic linear programming with measurable multiattribute value functions ORSA, J. Comput. 1
(1989) 146–158.
[33] H. Tanaka, H. Ichihashi, K. Asai, A value of information in fuzzy linear programming problems via sensitivity analysis, Fuzzy Sets Syst.
18 (2) (1986) 119–129.
[34] H. Tanaka, K. Asai, Fuzzy solution in fuzzy linear problems, IEEE Trans. Syst. Man Cybern. 14 (2) (1984) 325–328.
[35] H. Tanaka, P. Guo, H.J. Zimmermann, Possibility distributions of fuzzy decision variables obtained from possibilistic linear programming
problems, Fuzzy Sets Syst. 113 (2000) 323–332.
[36] S. Wang, J. Watada, Fuzzy Stochastic Optimization: Theory, Models and Applications, Springer, 2012.
[37] M. Sakawa, I. Nishizaki, H. Katagiri, Fuzzy Stochastic Multiobjective Programming, Springer, 2011.
[38] H. Tanaka, T. Okuda, K. Asai, A formulation of fuzzy decision problems with fuzzy information using probability measures of fuzzy events,
Inf. Control 38 (1978) 135–147.
[39] H. Tanaka, S. Uejima, K. Asai, Linear regression analysis with fuzzy model, IEEE Trans. Syst. Man Cybern. 12 (6) (1982) 903–907.
[40] H. Tanaka, Fuzzy data analysis by possibilistic linear models, Fuzzy Sets Syst. 24 (1987) 363–375.
[41] H. Tanaka, P. Guo, I.B. Turksen, Portfolio selection based on fuzzy probabilities and possibility distributions, Fuzzy Sets Syst. 111 (2000)
387–397.
[42] M.K. Luhandjula, Fuzzy stochastic linear programming: Survey and future, research directions, Eur. J. Oper. Res. 174 (2006) 1353–1367.
[43] Y.K. Liu, B. Liu, A class of fuzzy random optimization: Expected value models, Inf. Sci. 155 (2003) 89–102.
[44] H.J. Zimmermann, Fuzzy programming and linear programming with several objective functions, Fuzzy Sets Syst. 1 (1978) 45–55.
[45] M. Sakawa, Fuzzy Sets and Interactive Multiobjective Optimization, Plenum Press, New York, 1993.
[46] T. Murata, H. Ishibuchi, H. Tanaka, Multiobjective genetic algorithm and its application to flow shop scheduling, Comput. Ind. Eng. 30 (1996)
957–968.
[47] H.C. Wu, The Karush–Kuhn–Tucker optimality conditions for multiobjective programming problems with fuzzy valued objective functions,
Fuzzy Optim. Decis. Mak. 8 (2009) 1–28.
JID:FSS AID:6466 /FLA [m3SC+; v 1.184; Prn:4/02/2014; 8:27] P.8 (1-8)
8 M.K. Luhandjula / Fuzzy Sets and Systems ••• (••••) •••–•••

[48] A.O. Esogbue, J. Kacprzyk, Fuzzy dynamic programming, in: Fuzzy Sets in Decision Analysis, OR and Statistics, in: The Handbooks of
Fuzzy Sets, 1998, pp. 281–307.
[49] A.O. Esogbue, Computational and data acquisition issues in fuzzy dynamic programming, in: Proc. ORSA/TINS Conference, Anaheim, USA,
1991.
[50] H.S. Shih, Y.J. Lai, E.S. Lee, Fuzzy approach for multilevel programming problems, Comput. Oper. Res. 23 (1996) 73–91.
[51] M. Sakawa, I. Nishizaki, Interactive fuzzy programming for two-level nonconvex programming problems with fuzzy parameters through
genetic algorithms, Fuzzy Sets Syst. 127 (2002) 185–197.
[52] M. Sakawa, Fuzzy multiobjective and multilevel optimization, in: International Series in Operations Research and Management Sciences,
vol. 52, 2003, pp. 171–226.
[53] V.V. Prokhorov, D.V. Smirnov, A model of estimation and optimization of expenditure in multilevel fuzzy environments, Program. Comput.
Softw. 27 (2001) 252–259.
[54] H. Tanaka, Foreword, Fuzzy Optim. Decis. Mak. 3 (2004) 289–290.