Utilização de uma abordagem híbrida System Dynamics- Factory Physics para a investigação do efeito de ações de melhoria contínua na redução do lead time

Abstract

Nowadays, with the competitive environment dominating the world scene, companies need to be alert to continuous improvement opportunities. In this scene lead time reduction arises as one of the greatest goal to be accomplished. Within this context, the main goal of this research is to study the effect of continuous improvement in shop floor variables (such as variability, setup time, defect rate, among others) on lead time reduction for two productive environments: one with single machine and another flow shop environment. This is done using a quantitative model, which uses a hybrid System Dynamics factory Physics approach. This model was created by Godinho Filho and Uzsoy (2008a, 2008b, 2009). In this research this single machine model is extended by means of adding more workstations in order to create flow shop environment. Besides to the main goal, this research has the following specifics goals: (1) Check if the results found by Godinho Filho e Uzsoy (2008a, 2008b, 2009) are the same for scenes with different characteristics of productive system. (2) Determine if the positive effect in lead time of a continuous improvement action in the shop floor variables is sensitive to characteristics of productive system. If so, how is the relationship between the improvement action and the productive system. (3)Compare the results related with the two previous objectives for the single machine and flow shop cases. The results found in this research proved that conclusions made by Godinho Filho e Uzsoy (2008a, 2008b, 2009) are robusts for different characteristics of productive system in the single machine and flow shop environments studied in this research. In simulated scenes for these environments, the continuous improvements actions in processing time variability and setup time are the most effective in lead time reduction. Furthermore, the positive effect made by the improvement action in lead time is proved sensitive to almost the same characteristics of productive system for single machine and flow shop environments. For these environments, improvement actions in time to failures and repair time are the most sensitive to different characteristics and the most appropriate in productive systems with negative characteristics (e.g. high setup times, high processing time variability, among others). Results generated by quantitative model in this research were analyzed using design of experiments to obtain conclusions.