Implementation of a Mathematical Approach to Rip Saw Arbor Design and Scheduling

Abstract

The following discussion will detail efforts to improve the margin profile of a hard wood flooring plant through reduction of wood waste using mathematical optimization techniques. This investigation focuses on the first step in the flooring manufacturing process: the rip saw. This step is one of the largest sources of waste in the process. Furthermore, it is highly automated and software driven. Recent technological implementations at the rip saw and the wealth of data recorded during the process, make it an ideal target for optimization.

We first describe the details and challenges of the manufacturing process and the ways that a flooring company, AHF products, deals with these challenges. We then introduce a mathematical formulation for a column generation-based algorithm for rip saw arbor design and scheduling from a 1996 paper from Y. Fathi et al. We tweak the algorithm to accommodate the unique nuances of this particular plant and situation and we incorporate it into a wholistic arbor design and scheduling model. Finally, we discuss the challenges of implementing this complex model and ensuring that it is used after my internship.

Along the way, we will discuss the challenges faced throughout research and modeling. I will share the other discoveries that appeared during development. Most significantly, we will discuss the importance not just yield alone but also demand adherence when considering rip saw set up. We will also share the many other projects and benefits that come from tracking data on the manufacturing process and making that data easily available to stakeholders.