A novel failure mode and effect analysis model for machine tool risk analysis

A generic energy prediction model of machine tools using deep learning algorithms

CNC milling is a highly complex machining process highly valued in various industries, including the automotive and aerospace industries. With the increasing competition, manufacturers are aiming to keep maintenance costs low while ensuring high levels of manufacturing equipment reliability. It is also highly important for them to maximize the service life of each cutting tool by avoiding premature replacements while minimizing the risks of scrap due to tool breakage. This calls for the need for a condition-based maintenance approach and more powerful, flexible and robust tool condition monitoring (TCM) techniques with minimal reliance on subjective diagnosis based on the expert knowledge. This paper discusses the technical aspects of recent developments in state-of-the-art TCM techniques and current challenges which limit the viability of TCM in real-life industrial applications. The technical challenges in modern TCM were split into two major groups of problems: (1) challenges in data processing and (2) issues regarding tool wear model performance. Current methodologies to overcome issues in each of the sections in this paper are discussed and, where possible, compared to highlight their respective advantages and disadvantages. Finally, this paper concludes with a discussion on possible trends in TCM development and interesting avenues for future research.

Technical data-driven tool condition monitoring challenges for CNC milling: a review

Enhancing the reliability of machine tools and preventing the potential failures in the manufacturing process is one of the most important tasks for the development of industry. The failure mode and effects analysis (FMEA) is the well-known and widely utilized approach to identify and evaluate potential failures for preventing risk in various enterprises. It is a group-oriented technique ordinarily conducted by a group of experts from related fields. Obviously, an effective approach should be developed, which is used to integrate risk evaluation information from multiexperts by considering group and individual risk attitudes. This article proposes a hybrid FMEA framework for addressing machine tool risk analysis problem by integrating cloud model, Choquet integral, and gained and lost dominance score (GLDS) method. In this framework, an improved Shapley cloud-Choquet weighting averaging operator is defined to fuse random and uncertain risk information by considering the correlations among experts. An extended GLDS method with developed Choquet integral based on distance measure of normal clouds is presented to prioritize the risk priority of each failures, in which the group and individual risk attitude and risk interactions are considered simultaneously. Finally, a real risk analysis of machine tool in a machine tool industry is introduced to illustrate the application and feasibility of the proposed approach, and comparison and sensitivity studies are also conducted to validate the effectiveness of the hybrid framework.

Hung Wei Chiu

Papers

Prediction of machining accuracy and surface quality for CNC machine tools using data driven approach