Hybrid approach to supporting decision making processes in companies

This article presents the advantages of hybrid approach to the support decision making by analyzing three areas of business decision problems, solved by combination of well-known algorithms into the new hybrid constructions: cascade optimization hybrid, parallel classification hybrid and hybrid multicomponent attribute selection. Each of them solved a different problem: the cascade optimization hybrid allowed for finding an extreme of a composite objective function, the parallel classification hybrid was used to choose a proper class through voting, the multicomponent attribute selection robustly chose significant decision variables. A hybrid approach to the problem of supporting the decision making processes is more effective than using each of the component methods alone, even for the sophisticated ones. A combination of several methods with different characteristics and performance makes it possible to take advantages of their strong sides and simultaneously eliminate the weak ones, resulting in a better computational support of decision making.