Scheduling and production planning for a sawmill: A comparison of a mathematical model versus a heuristic
Sergio Maturana, Enzo Pizani
Abstract
Sergio Maturana, Enzo Pizani
Abstract
Production planning and scheduling for sawmills poses many challenges. The solution approaches most commonly used to tackle this problem in the forest industry are: simulation, mathematical programming models, and a combination of both. In this paper we compare the results of using a mathematical model versus a heuristic in scheduling and production planning of a sawmill. We developed a mathematical model that takes into account most of the complexities involved in scheduling and production planning of a sawmill, for a one-month planning horizon. One of the main difficulties is that the supply of hardwood lumber pieces can have different diameters, which are not known in advanced. The cutting patterns that can be used depend on the diameter of the lumber piece to be cut. The cutting pattern, on the other hand, determines the products that are produced. Sometimes suboptimal cutting patterns must be used in order to meet the production plan. We compare the quality of the solutions generated by the mathematical model and the heuristic. Several experiments were conducted using data that represents situations faced by a Chilean sawmill.
OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Production planning and scheduling for sawmills poses many challenges. The solution approaches most commonly used to tackle this problem in the forest industry are: simulation, mathematical programming models, and a combination of both. In this paper we compare the results of using a mathematical model versus a heuristic in scheduling and production planning of a sawmill. We developed a mathematical model that takes into account most of the complexities involved in scheduling and production planning of a sawmill, for a one-month planning horizon. One of the main difficulties is that the supply of hardwood lumber pieces can have different diameters, which are not known in advanced. The cutting patterns that can be used depend on the diameter of the lumber piece to be cut. The cutting pattern, on the other hand, determines the products that are produced. Sometimes suboptimal cutting patterns must be used in order to meet the production plan. We compare the quality of the solutions generated by the mathematical model and the heuristic. Several experiments were conducted using data that represents situations faced by a Chilean sawmill.
Key concepts: Production planning, Scheduling (production processes), Time horizon, Computer science, Heuristic, Automated planning and scheduling, Industrial engineering, Mathematical optimization