Establishment of an IoT-based smart factory and data analysis model for the quality management of SMEs die-casting companies in Korea

Abstract

In this research, an Internet of things–based smart factory was established for a die-casting company that produces automobile parts, and the effect of casting parameters on quality was analyzed using data collected from the system. Most of the die-casting industry in Korea consists of small- and medium-sized enterprises with inferior finances and skeptical views about the establishment of a smart factory. In response, the Korean government is providing various types of support to spread the implementation of smart factories for small- and medium-sized enterprises. Although small- and medium-sized enterprises have become more active in establishing smart factories according to the government policies, the effect of smart factories requires real-time monitoring. A monitoring system has been built but the data collected are not being utilized properly. Therefore, it is necessary to establish a system suitable for the die-casting environment and data analysis purposes and to utilize it to enable the analysis of data. To this end, we established to smart factory that provides data based on the Internet of things. Among the data collected, casting parameter data were analyzed through a data mining technique to establish a relationship between casting parameters and the quality of production. It is expected that a method of systematic implementation will be provided to die-casting companies that want to build smart factories in the future and that a plan for managing casting parameter by-product will be established. In addition, algorithms that can solve the problem of multi-collinearity among the casting parameters and aid in the development of new products are needed to detect optimum casting parameters.

Machine Learning: Supervised Algorithms to Determine the Defect in High‐Precision Foundry Operation

“…It is used to take previous inputs to predict 2 Journal of Nanomaterials the feature aspects. It is again divided into two subcategories [15,16] (A) Regression: this method generally gives the output as {correct, wrong}, {positive, negative}, {yes, no}, {true, false} etc., picture as an input source [17] (B) Classification: this method will give the output as a new category or classification as an input of given historical datasets [18] (2) Unsupervised learning: it will predict the output by using inputs as previous data or historical evidence…”

Research on Risk Analysis and Countermeasures of Micro Store Operation Model

“…Foreign scholars have studied the composition of the business model and the operating mechanism, as follows (Kumar et al, 2012;Lambert et al, 2019;Lyu et al, 2020;Park et al, 2019).…”

Factors Analysis and Prediction in Die-casting Process for Defects Reduction

“…mold quality, mold temperature, liquid aluminum temperature, casting pressure, pouring rate, runner inadequate, low/high speed range inadequate, etc. (Sangwoo Park et al 2019).…”