Una optimización integrada de la producción de madera aserrada y el mantenimiento de los equipos críticos.
Loading...
Date
2025
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de Concepción
Abstract
Esta investigación propone un modelo de optimización integrada para la planificación de producción y mantenimiento en aserraderos, considerando las etapas secuenciales de aserrío, almacenamiento en tránsito y secado en cámaras. A diferencia de los enfoques tradicionales que tratan estos procesos por separado, se incorpora la evolución dinámica de la confiabilidad de los equipos críticos, representada mediante distribuciones de Weibull y priorizada con el método estadístico Jack-Knife. El modelo propuesto es formulado como un problema de programación lineal entera mixta, que minimiza el makespan global del sistema bajo restricciones operativas, tanto de mantenimiento como de secuenciación de productos. La solución es validada a través de dos instancias de prueba, una con cuatro productos y otra con nueve productos y diferentes cantidades de cada producto a producir. Los resultados muestran que el enfoque integrado permite reducir el tiempo total de producción, mejorar la disponibilidad de los equipos y disminuir los costos operativos frente a un modelo que no considera el mantenimiento en la etapa de optimización. Además, al incorporar la probabilidad de falla y simular condiciones reales, se observa que el modelo basado en confiabilidad es más robusto y mejora el desempeño en makespan y costos comparado con el modelo sin considerar el mantenimiento desde un principio. Esta propuesta contribuye al desarrollo de herramientas avanzadas para la gestión de activos en la industria forestal, fortaleciendo la sostenibilidad operativa y la toma de decisiones estratégicas.
This research proposes an integrated optimization model for production and maintenance planning in sawmills, considering the sequential stages of sawing, in transit storage, and kiln drying. Unlike traditional approaches that address these processes separately, the model incorporates the dynamic evolution of equipment reliability, represented through Weibull distributions and prioritized using the Jack-Knife statistical method. The proposed model is formulated as a Mixed Integer Linear Programming problem, aiming to minimize the system’s global makespan under operational constraints related to maintenance activities and product sequencing. The solution is validated through two test instances with varying numbers and types of products to be manufactured. Results show that the integrated approach reduces total production time, improves equipment availability, and lowers operational costs compared to a model that does not consider maintenance during optimization. Furthermore, by incorporating failure probability and simulating real world conditions, it is observed that the reliability-based model is more robust and efficient in makespan and cost, compared to the non-reliability-based model. This work contributes to the development of advanced tools for asset management in the forestry industry, enhancing operational sustainability and supporting strategic decision-making.
This research proposes an integrated optimization model for production and maintenance planning in sawmills, considering the sequential stages of sawing, in transit storage, and kiln drying. Unlike traditional approaches that address these processes separately, the model incorporates the dynamic evolution of equipment reliability, represented through Weibull distributions and prioritized using the Jack-Knife statistical method. The proposed model is formulated as a Mixed Integer Linear Programming problem, aiming to minimize the system’s global makespan under operational constraints related to maintenance activities and product sequencing. The solution is validated through two test instances with varying numbers and types of products to be manufactured. Results show that the integrated approach reduces total production time, improves equipment availability, and lowers operational costs compared to a model that does not consider maintenance during optimization. Furthermore, by incorporating failure probability and simulating real world conditions, it is observed that the reliability-based model is more robust and efficient in makespan and cost, compared to the non-reliability-based model. This work contributes to the development of advanced tools for asset management in the forestry industry, enhancing operational sustainability and supporting strategic decision-making.
Description
Tesis presentada para optar al grado de Magíster en Ingeniería Industrial.
Keywords
Mantenibilidad (Ingeniería), Aserraderos, Programación lineal