Embedding of linear programming in a simulated annealing algorithm for solving a mixed integer production planning problem

The present paper is concerned with the grouping of book covers on offset plates in order to minimize the total production cost. The mathematical formulation of the problem involves both binary and continuous variables. As exact methods are unable to provide solutions in reasonable time, a heuristic algorithm of the simulated annealing type is proposed. At each iteration, the values of the current solution binary variables are altered in order to yield a neighboring solution. To compute the corresponding values of the continuous variables and the value of the objective function, a linear programming routine is called at each iteration. This constitutes the main originality of the present approach and is in principle applicable in mixed integer programming problems. The procedure is tested on several examples.