Research On Harmonious Production Planning For Multiple Product Line

On the basis of analyzing the research status of production planning and investigation for manufacture-enterprise production-scene and studying thoroughly programming theory, a relational structure of up-down stream parts with limited buffer is established, which has a certain universality. A harmonious production planning problem for multiple-product-line is presented by regarding it as an object of research. The paper studies the problem of production planning from three aways which are production planning modeling, optimization algorithm and algorithm realizing. By analyzing the relational structure of up-down stream parts and the workshops production process. the paper offers several production planning model to befit varied production programming demands, such as satisfying the clients' demand,decreasing the cost of production,harmonizing the synchronization among the production-lines, avoiding the overflow of the buffer and increasing the benefit. For the optimization algorithm applied to the production planning programming, because the objective function of the models is nonlinear, the gradient of which is piecewise constant, the normal nonlinear programming algorithm can not be used in here. On account of the linear of the constraints, the paper first convert the nonlinear objective function to linear objective function. So the nonlinear programming problem which is difficult to be solved, can be transformed into a linear programming problem,Then using the amend-simplex algorithm and two-stage relation/adding-power average prediction approach to solve it. The paper uses OOP methods for designing production planning simulation models. The simulation models library and algorithm library can be setup by vc6.0.At the end of this paper, Simulation results of several examples are given to illustrate the efficiency of the proposed model and algorithm. In addition, the paper finally gives the limitation and more work of the theory and algorithm.