Study of Multi-Stage Production Planning Based on Integer Planning

Weining Feng,Lingxuan Weng
DOI: https://doi.org/10.1145/3582935.3583036
2022-11-04
Abstract:Production planning at minimum cost while meeting external product demand is one of the most important ways to improve a company's core competitiveness. This paper takes a factory producing a pipe-cleaning robot as an example: with minimum production costs as the optimization objective, a single-objective planning model with minimum total cost as the objective is developed, using the daily assembly and inventory quantities of the product WPCR and other components as decision variables. In order to calculate the total production cost, a 0-1 variable was introduced into the model to determine whether or not to produce on that day, and a limit on component man-hours, the daily demand for the WPCR and the stipulation that no product is in stock on Mondays and Sundays were introduced as constraints to construct the final integer planning model. In this paper, a single-objective-based optimization model is developed to solve the problem of cyclical product production in a factory, providing an optimal production solution for each planning stage. This paper uses a simulated annealing algorithm with the introduction of memory variables to solve the single-objective optimization problem, making the solution process faster and avoiding falling into a local optimum. ARIMA predictions are also used to obtain confidence intervals at a certain confidence level and are combined with p-quantile tests to ensure the reliability of future production planning.
Engineering,Business,Computer Science
What problem does this paper attempt to address?