Multi-objective optimization of the hot rolling scheduling of steel using a genetic algorithm

The hot rolling process reduces a slab passing through a series of work-rolls to obtain a strip of target thickness. Developing robust, efficient, and accurate simulation methods improve the process. This research aims to minimize the hot rolling time, bending of work rolls, thermal crown, and wear of work rolls, subject to some process constraints. The problem solution is by using a multi-objective genetic algorithm with four function objectives. The second generation of the Non-dominated Sorting Genetic Algorithm was chosen to solve the problem of this research. A probed constitutive model has been incorporated into the algorithm to compute the flow stress as a function of the chemical composition of steels. The algorithm implemented to minimize the four objectives proposed obtained the optimal schedule and associated makespan.