A comparison of genetic and conventional methods for the solution of integer goal programs

This paper discusses two different approaches to the solution of difficult Goal Programming (GP) models. An integer Goal Programming (IGP) solver and some genetically driven multi-objective methods are developed. Specialised GP speed up techniques and analysis tools are employed in the design and development of the solution systems. A selection of linear integer models of small to medium size with an internal structure that makes solution difficult are considered. These problems are solved by both methods in order to assess their computational performance over several criteria and to compare the differences between them. From the results obtained in this research, it is observed that genetic algorithms (GA) have performed in general less efficiently than the Integer Goal Programming system for the sample of problems analysed.