Application of a production planning model based on linear programming and machine learning techniques

Abstract

The absence of efficient optimization methods combined with Artificial Intelligence concepts has led to inefficiencies and high costs in the production planning of organizations. Thus, this study aims to optimize production planning in an electronic equipment company, using Linear Programming and Machine Learning to support assertive and efficient decisions. The methodological process comprises seven stages: Literature review; Collection and analysis of production data; Application of Machine Learning methods for modelling; Selection of the best model; Development and application of the Linear Programming model; Analysis of results; Validation with stakeholders. The approach resulted in optimized production planning, capable of reducing operating costs and assisting in the daily decision-making of the organization. The Machine Learning forecasting technique achieved an average error of 9%, demonstrating its accuracy in forecasting future demand. This study evidences a robust and promising approach to improve efficiency and effectiveness in production planning operations. In this context, the union between Operations Research and Machine Learning emerges as a response to existing gaps and a driving direction for continuously optimizing these crucial processes.