Search

The rising throughput of log-yards imposes new constraints on existing equipment and increases the complexity of delivering an optimal and uninterrupted supply of pulpwood to pulp mills. To find ways of addressing these problems by reducing log cycle times, this work uses a discrete-event mathematics model to simulate operations at a log-yard and study the impact of three different log-yard inventory strategies and two alternative machine systems for log transportation between main log-yard and buffer storage. The yard's existing inventory strategy of last load in and first out limits access to older logs at the main storage site. By allocating space for 89,000 m3 and 99,000 m3 of pulpwood at the buffer storage it is possible to keep the log cycle time at the main storage to a maximum of 12 and 6 months. Additionally, the use of an alternative log transportation machine system comprising a material handler with a trailer increased the work time capacity utilization relative to the yard's current machine system of two shuttle trucks and a material handler for transporting logs between the main and buffer storage areas. Compared to the currently-used last in first out inventory strategy and purposely emptying the main storage area once or twice per year did reduce the total work time of both machine systems by 14% and 30%. Consequently, the volume delivered from the buffer to the log-yard decreased on average by 17% and 37% when emptying the main storage area once and twice per year. Even with reduced work time when emptying the main storage area, both machine systems could fulfil given work load for transporting logs from the buffer storage to the main log-yard without interrupting operations of the log-yard. [ABSTRACT FROM AUTHOR]

This article deals with the topic of modelling the log-yard of one of our industry partners. To this end, our framework is based on discrete-events modelling (DEM), as consequence that many stages of the process run as a sequence of events. The sequence starts when trucks, trains or ships arrive loaded with logs to the log-yard. A machine unloads these logs and accumulates them in different storage areas. Consequently, a machine transports logs from these areas to the pulp mill, thus finishing the process. As using probability density functions is the core concept of DEM, the necessary process data to build these PDFs have been partly provided by the company. Other necessary data have been acquired through time studies, and by defining operational requirements. The company data tell when trucks, trains, or ships arrive to the log-yard, and the amount of volume they carry. The objective is to develop the necessary formulations, model calibration techniques, and software, such that computer simulations reproduce the quantities observed in these data. To this end, this work suggests two alternatives to analyse the data itself. These two alternatives lead to two different models: (1) The first being a hybrid model, in the sense that it involves the events in the process, and the logic decisions taken by machine operators for handling the incoming load, and (2) the second containing only the main mathematical essence of the process. After running 100 simulations, both mathematical models show that the simulated values for input and output, in terms of transport units and their volume, differ only by less than 3% compared to company data. The first model has also shown the ability to replicate the decision making that a machine operator undergoes for driving the logs to the storage areas, and from there to the mill. Therefore, the framework adopted provides the necessary mathematical tools and data analysis to model the log-yard and obtain highly reliable results via simulations. [ABSTRACT FROM AUTHOR]