Satisficing-based formulation of fuzzy random multi-criteria programming models in production applications

Arbaiy, Nureize (2012) Satisficing-based formulation of fuzzy random multi-criteria programming models in production applications. PhD thesis, Waseda University.

[img]
Preview
PDF
339Kb

Abstract

In practice of various real-life applications, mathematical programming plays a pivotal role in finding the solution of their optimization problems. Conventionally, mathematical programming is set with numerical values although it is troublesome for decision makers to provide rigid values in presence of uncertainties in decision making process. Building mathematical programming model with crisp and precise values sometimes generates infeasible or improper solution. Besides that, when the real-life application faces hybrid situation of simultaneous fuzziness and randomness, or ambiguous and vague information, it makes the existing multicriteria evaluation model incapable of handling such uncertainties. Satisficing based optimization is used as underlying concept, that is to realize the reality of decision making process which seeks for satisficing based solution rather that optimal solution. Hence, based on different multicriteria evaluation scheme and requirement, the objective of this study is to propose three kinds of mathematical programming model: (1) multi-attribute evaluation model, (2) satisficing based multi-objective evaluation model, and (3) possibility based multi-objective evaluation model. The initial model-setting of all is done by fuzzy random regression analysis, which alleviates the difficulties to determine the model’s coefficients in fuzzy random circumstances. The algorithms presented herein are accompanied with numerical experiments where data are taken from the industry application of oil palm production. The analytical results of the proposed methods reveal the improvement of conventional decision making approaches to appropriately handle inherent uncertainties contained in the real-world situation. The implementation of the proposed method shows the significant capabilities to solve real application problem.

Item Type:Thesis (PhD)
Subjects:Q Science > QA Mathematics
ID Code:4652
Deposited By:Normajihan Abd. Rahman
Deposited On:30 Dec 2013 10:26
Last Modified:30 Dec 2013 10:26

Repository Staff Only: item control page