MOSS—Multi-Modal Best Subset Modeling in Smart Manufacturing

Abstract

Smart manufacturing, which integrates a multi-sensing system with physical manufacturing processes, has been widely adopted in the industry to support online and real-time decision making to improve manufacturing quality. A multi-sensing system for each specific manufacturing process can efficiently collect the in situ process variables from different sensor modalities to reflect the process variations in real-time. However, in practice, we usually do not have enough budget to equip too many sensors in each manufacturing process due to the cost consideration. Moreover, it is also important to better interpret the relationship between the sensing modalities and the quality variables based on the model. Therefore, it is necessary to model the quality-process relationship by selecting the most relevant sensor modalities with the specific quality measurement from the multi-modal sensing system in smart manufacturing. In this research, we adopted the concept of best subset variable selection and proposed a new model called Multi-mOdal beSt Subset modeling (MOSS). The proposed MOSS can effectively select the important sensor modalities and improve the modeling accuracy in quality-process modeling via functional norms that characterize the overall effects of individual modalities. The significance of sensor modalities can be used to determine the sensor placement strategy in smart manufacturing. Moreover, the selected modalities can better interpret the quality-process model by identifying the most correlated root cause of quality variations. The merits of the proposed model are illustrated by both simulations and a real case study in an additive manufacturing (i.e., fused deposition modeling) process.

Multi-Attribute Subset Selection enables prediction of representative phenotypes across microbial populations

“…In contrast, for example, each principal component in PCA is a linear combination of all available attributes, and it would be less straightforward to choose the most descriptive attributes out of those. As opposed to other methods that broadly fall into the category of learning problems (such as Multi-Modal Best Subset (MOSS) modeling, which is focused on product quality optimization based on sensors in additive manufacturing 62 ), our approach is specifically helpful for analyzing biological phenotype matrices as it is designed to find, given a certain number of attributes ( p ), a selection of predictor conditions (attributes) which optimally predicts all the remaining attributes.…”

Prediction of representative phenotypes using Multi-Attribute Subset Selection

“…In contrast, for example, each principal component in PCA is a linear combination of all available attributes and it would be less straightforward to choose the most descriptive attributes out of those. As opposed to other methods that broadly fall into the category of learning problems (such as Multi-Modal Best Subset (MOSS) modeling, which is focused on product quality optimization based on sensors in additive manufacturing (Wang, Du, and Jin 2021)), our approach is specifically helpful for analyzing biological phenotype matrices as it is designed to find, given a certain number of attributes ( p ), a selection of predictor conditions (attributes) which optimally predicts all the remaining attributes.…”

A Systematic Literature Review: Industry 4.0 Based Monitoring and Control Systems in Additive Manufacturing

“…With its open-source software and hardware, the proposed system additionally aims to help manufacturers to cope with major disruptions (such as the global pandemic). Wang et al [51] developed "Multi-modal best subset" model to increase cost-effectiveness in smart manufacturing systems by choosing the correct sensors and deciding on sensor locations. As a case study, they installed an infrared sensor, accelerometers and thermocouples on an FDM-type 3D printer, and then successfully found the most relevant sensory data to monitor a quality variable.…”