An Efficient Approach for Uncertain Event Detection in RFID Complex Event Processing

The globalization of manufacturing has increased the risk of counterfeiting as the demand grows, the production flow increases, and the availability expands. The intensifying counterfeit issues causing a worriment to companies and putting lives at risk. Companies have ploughed a large amount of money into defensive measures, but their efforts have not slowed counterfeiters. In such complex manufacturing processes, decision-making and real-time reactions to uncertain situations throughout the production process are one way to exploit the challenges. Detecting uncertain conditions such as counterfeit and missing items in the manufacturing environment requires a specialized set of technologies to deal with a flow of continuously created data. In this paper, we propose an uncertain detection algorithm (UDA), an approach to detect uncertain events such as counterfeit and missing items in the RFID distributed system for a manufacturing environment. The proposed method is based on the hashing and thread pool technique to solve high memory consumption, long processing time and low event throughput in the current detection approaches. The experimental results show that the execution time of the proposed method is averagely reduced 22% in different tests, and our proposed method has better performance in processing time based on RFID event streams.

Effect of Customized Cosmetics Selection Attributes and Safety Perception on Intention to Purchase

Purpose: This study investigated how choice attributes and safety perception of adult female consumers influence purchase intention.Methods: Data were collected from women aged 20 years and older and analyzed using SPSS WIN 25.0 program. Analyses included frequency, descriptive statistics, cross-tabulation, one-way ANOVA, factor, reliability, principal component, and regression.Results: Among the attributes of personalized cosmetic selection, quality was highly significant for women aged 40 and older. As regards the safety perception of customized cosmetics, the safety of the manufacturing environment was significantly important for women in their 30s and that of the expiration date, and the ingredients were significantly higher for women aged 40 and above. The significance of safety was high in the case of married women and those in the income group above 4 million won. Quality has the greatest positive impact on safety of the manufacturing environment and shelf life, and perceptual value has the greatest positive impact on universal safety. Perceptual value and universal safety had the greatest positive influence on purchasing intention.Conclusion: Through this study, we increase the safety perception of customized cosmetics to consumers who choose customized cosmetics and help them select customized cosmetics according to their selection attributes. For related companies, we not only hope to develop customized cosmetics with proven safety but also build and develop appropriate marketing strategies.

Towards low-cost machine learning solutions for manufacturing SMEs

Machine learning (ML) is increasingly used to enhance production systems and meet the requirements of a rapidly evolving manufacturing environment. Compared to larger companies, however, small- and medium-sized enterprises (SMEs) lack in terms of resources, available data and skills, which impedes the potential adoption of analytics solutions. This paper proposes a preliminary yet general approach to identify low-cost analytics solutions for manufacturing SMEs, with particular emphasis on ML. The initial studies seem to suggest that, contrarily to what is usually thought at first glance, SMEs seldom need digital solutions that use advanced ML algorithms which require extensive data preparation, laborious parameter tuning and a comprehensive understanding of the underlying problem. If an analytics solution does require learning capabilities, a ‘simple solution’, which we will characterise in this paper, should be sufficient.