Annealing penalty hybrid genetic algorithm for solving production-scheduling problem

The lot size production-scheduling problem with the objective of minimizing production cost is considered and a mixed integer-programming model is given. Firstly, the mathematical description of this problem is presented based on the characteristic of single-level and multi-resource production scheduling. Then, in view of complexity of this problem, the optimization solutions are obtained based on a random searching and evolution process of the genetic algorithm. In order to avoid the premature convergence of the fitness functions, an annealing penalty factor is introduced to deal with the fitness function, and the method increases the probability of achieving the whole level optimization solution. Experimental results show its validity and the method can be applied to solve the production-scheduling problems.