Análisis y propuesta de una medida de flexibilidad para sistemas productivos del tipo Job Shop Flexible en entornos multiagentes.
Loading...
Date
2023
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de Concepción
Abstract
Los problemas relacionados con la planificación de la producción han experimentado cambios significativos en los últimos tiempos. Uno de los desafíos más frecuentes tanto en la literatura como en la industria es la secuenciación del Job Shop Flexible. Este problema se enmarca dentro de la programación y se considera una extensión del clásico problema del Job Shop. La adaptación de este proceso representa una mejora significativa para los fabricantes actuales, ya que los avances tecnológicos permiten que una máquina pueda manejar más de una operación. La creciente cantidad de posibles soluciones ha llevado a que este problema se clasifique como NP-hard en términos de complejidad computacional. En consecuencia, presenta un desafío en términos de la flexibilidad de los procesos, en lo que respecta a la asignación de operaciones a un conjunto de máquinas disponibles. En este estudio se propone una medida de desempeño diseñada para cuantificar la flexibilidad en sistemas productivos de tipo Job Shop Flexible. Esta medida se desarrolla a partir de diversas perspectivas de flexibilidad mencionadas en diferentes artículos científicos y se emplea para evaluar un problema encontrado en la literatura, a través de dos modelos; un modelo tradicional y un modelo Product Driven System (PDS) al someterse a eventos que alteran la planificación. Los resultados obtenidos indican que el modelo PDS supera al modelo tradicional en términos de flexibilidad, en promedio, existe una diferencia porcentual de un 3,8%. Esto se traduce en un menor impacto de las perturbaciones debido a la capacidad de los productos para tomar decisiones autónomas y optimizar sus acciones en función de las nuevas condiciones. En términos del tiempo de finalización del sistema, el modelo tradicional presenta un aumento promedio de un 34% luego de sufrir una perturbación, mientras que el modelo PDS un aumento de un 25%. Dada la consistencia de estos resultados con la literatura y la realidad de los modelos inteligentes, se respalda la validez y aplicabilidad de la medida de flexibilidad para evaluar la respuesta de los sistemas ante eventos inesperados, tanto internos como externos. Esto respalda la idea de que las organizaciones que adoptan una arquitectura anárquica en su proceso de fabricación logran un rendimiento superior y una mayor competitividad en el mercado actual, permitiéndoles satisfacer de manera más eficiente, rápida y ágil la demanda comercial.
The problems related to production planning have undergone significant changes in recent times. One of the most frequent challenges in both literature and industry is the sequencing of the Flexible Job Shop. This problem falls within the realm of scheduling and is considered an extension of the classic Job Shop problem. Adapting this process represents a significant improvement for current manufacturers, as technological advances enable a machine to handle more than one operation. The increasing number of possible solutions has led to the classification of this problem as NP-hard in terms of computational complexity. Consequently, it poses a challenge in terms of process flexibility, particularly concerning the assignment of operations to a set of available machines. This study proposes a performance measure designed to quantify flexibility in Flexible Job Shop production systems. This measure is developed from various flexibility perspectives mentioned in different scientific articles and is used to evaluate a problem found in the literature through two models: a traditional model and a Product Driven System (PDS) model subjected to events that disrupt the schedule. The results indicate that the PDS model outperforms the traditional model in terms of flexibility. On average, there is a percentage difference of 3.8%. This translates to a lower impact of disturbances due to the products' ability to make autonomous decisions and optimize their actions based on new conditions. In terms of system completion time, the traditional model shows an average increase of 34% after a disruption, while the PDS model shows an increase of 25%. Given the consistency of these results with the literature and the reality of intelligent models, the validity and applicability of the flexibility measure to assess system response to unexpected events, both internal and external, are supported. This reinforces the idea that organizations adopting an anarchic architecture in their manufacturing process achieve superior performance and greater competitiveness in the current market, enabling them to more efficiently, quickly, and agilely meet commercial d
The problems related to production planning have undergone significant changes in recent times. One of the most frequent challenges in both literature and industry is the sequencing of the Flexible Job Shop. This problem falls within the realm of scheduling and is considered an extension of the classic Job Shop problem. Adapting this process represents a significant improvement for current manufacturers, as technological advances enable a machine to handle more than one operation. The increasing number of possible solutions has led to the classification of this problem as NP-hard in terms of computational complexity. Consequently, it poses a challenge in terms of process flexibility, particularly concerning the assignment of operations to a set of available machines. This study proposes a performance measure designed to quantify flexibility in Flexible Job Shop production systems. This measure is developed from various flexibility perspectives mentioned in different scientific articles and is used to evaluate a problem found in the literature through two models: a traditional model and a Product Driven System (PDS) model subjected to events that disrupt the schedule. The results indicate that the PDS model outperforms the traditional model in terms of flexibility. On average, there is a percentage difference of 3.8%. This translates to a lower impact of disturbances due to the products' ability to make autonomous decisions and optimize their actions based on new conditions. In terms of system completion time, the traditional model shows an average increase of 34% after a disruption, while the PDS model shows an increase of 25%. Given the consistency of these results with the literature and the reality of intelligent models, the validity and applicability of the flexibility measure to assess system response to unexpected events, both internal and external, are supported. This reinforces the idea that organizations adopting an anarchic architecture in their manufacturing process achieve superior performance and greater competitiveness in the current market, enabling them to more efficiently, quickly, and agilely meet commercial d
Description
Tesis presentada para optar al grado de Magíster en Ingeniería Industrial.