论文信息 - Randomized interpolation: a crossover operator producing children at a constraint transition

Randomized interpolation: a crossover operator producing children at a constraint transition

We introduce a type of genetic crossover operation between an arbitrary feasible solution and any infeasible solution to a constrained optimization model (COModel). This crossover operator produces a feasible-infeasible pair on the boundary, that is, the individuals in the resulting pair of solutions are minimally different.

Steven Orla Kimbrough | David Harlan Wood | S. Kimbrough | D. Wood

[1] Steven Orla Kimbrough,et al. On a Feasible-Infeasible Two-Population (FI-2Pop) genetic algorithm for constrained optimization: Distance tracing and no free lunch , 2008, Eur. J. Oper. Res..

[2] Tomasz Dominik Gwiazda,et al. Crossover for single-objective numerical optimization problems , 2006 .

[3] Steven Orla Kimbrough,et al. On repair by binary interpolation , 2007, GECCO '07.

[4] Steven O. Kimbrough,et al. On using genetic algorithms to support post-solution deliberation in the generalized assignment problem , 2009 .

[5] Hoong Chuin Lau,et al. Effective heuristic methods for finding non-optimal solutions of interest in constrained optimization models , 2010, GECCO '10.

[6] S. Kimbrough,et al. On Repair by Binary Interpolation : A Genetic Operator Having Offspring at a Constraint Transition ∗ , 2007 .

[7] Steven Orla Kimbrough,et al. Exploring the Evolutionary Details of a Feasible-Infeasible Two-Population GA , 2004, PPSN.