عنوان مقاله [English]
Supply chain management and integration of its components are a key issue for sustainable economy. One of the most important in optimization supply chain modeling is production- distribution planning problem. Several authors have developed models for the production-distribution problem when only a percentage of solution procedure is in exact area. Most of these models were solved with
the meta-heuristic method. In this paper, we are extended a production-distribution nonlinear programming problem in a two-echelon supply chain network, including manufacturers and distributors, and are solved with a mixed of exact solution and a meta-heuristic algorithm. The aim of this research is to determine the value of products delivered and the carrying amount of each vehicle such that the profit average, including sales price, production costs and transportation costs, is maximized. The model is for multiple distributors and all manufacturers in which all manufacturers are produced a type of product and are sent it to distributors. The mathematical model of the production-distribution problem is derived for which the objective function is proved to be convex, and the constraints being in linear forms are convex too. So, the proposed model is a convex nonlinear programming problem and its local maximum is the global maximum. Then, the proposed nonlinear programming problem is solved by two methods of a genetic algorithm and, Sequential Unconstrained Minimization Technique (SUMT) approach along with steepest descent method. The SUMT is the usual way in which constrained
problems are converted to an unconstrained form and solved that way. It makes
use of barrier methods as well to find a suitable initial point that over satisfies the inequality constraints. In this study, the genetic algorithm is used to validate the SUMT nonlinear programming approach. The numerical example is provided to illustrate the solution methods. Finally, future research and conclusion recommendations come in the last section of paper.