Integration of discrete-event dynamics and machining dynamics for machine tool: modeling, analysis and algorithms

TL;DR

This paper integrates discrete-event and machining dynamics to optimize machine tool operation, using a learning-based cost function and mathematical models to improve efficiency and reduce costs in practical production settings.

Contribution

It introduces a novel integration of discrete-event and machining dynamics with a learning-based cost function for machine tool optimization.

Findings

Proposed a learning-based cost function for tool life prediction.

Developed mathematical models considering cost, makespan, and due date.

Numerical results demonstrate effectiveness and practical potential.

Abstract

Machining dynamics research lays a solid foundation for machining operations by providing stable combinations of spindle speed and depth of cut. Furthermore, machine learning has been applied to predict tool life as a function of cutting speed. However, the existing research does not consider the discrete-event dynamics in machine shop, i.e., the machine tool needs to process a series of parts in queue under various practical production requirements. This paper addresses the integration of discrete-event dynamics and machining dynamics to achieve cost savings in machining. We first propose a learning-based cost function for the studied integrated optimization problem of machine tool. The proposed cost function utilizes the predicted tool life under different stable cutting speeds for further optimizing speed selection of machine tool to deal with the discrete-event dynamics in machine shop. Then, according to the practical production requirements, we develop several mathematical optimization models for the related integrated optimization problems with the consideration of cost, makespan and due date. Numerical results show the effectiveness of our proposed methods and also the potential to be used in practice.