Abstract

Technology portfolio selection is one of the critical decision making process for any manager. This paper considers a fuzzy mixed portfolio selection with random fuzzy return and new hybrid algorithm approach for solving it. Fuzzy set theory is applied to model uncertain and flexible information. Since traditional technology evaluation methods is often unable to consider vague data gathered from environment, in this paper, the returns of each technology are assumed to be fuzzy random variables. In proposed hybrid algorithm we employ outputs of neural networks to produce initial solution for genetic algorithm and in order to reduce the computational work. Finally, we apply the model for investment case in carwash technologies selection. We concluded that the proposed approach can assist decision makers in selecting suitable technology portfolios, while there is a lack of reliable project information.

Key words: Technology portfolio, fuzzy random variable, neural network, chance programming, genetic algorithm, hybrid intelligence.