ABSTRACT
One of the traditional tools for information aggregation is the weighted average method, for example, a linear integral or the Lebesgue integral (Lebeggue 1966). These methods assume that the information sources involved are non-interactive/ independent and, hence, their weighted effects are viewed as additive type. However, this assumption is not realistic in many real-world applications. Due to some inherent interaction/interdependencies among diverse information sources, the weighted average method does not work well in many real problems. Instead of the weighted average method, the Choquet integral can be used. The Choquet integral can be applied to multiattribute evaluation such as Grabisch (1995, 1996), Lee, Liu, and Tzeng (2001), Chen and Tzeng (2001), etc. Fuzzy measures and fuzzy integrals can analyze the human evaluation process and specify decision-makers’ preference structures.
