http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Development of A Novel Framework for Liquid Bulk Cargo Volume Analysis
Suhyeon Kim,Wonho Sohn,Dongcheol Lim,Junghye Lee 대한산업공학회 2019 대한산업공학회 추계학술대회논문집 Vol.2019 No.11
Port cargo volume analysis is a challenging task for researchers because of non-stationary and highly volatile data affected by external factors. Nevertheless, it is important to establish an analysis system for the port cargo volume as the analysis of the port cargo volume can provide information on the establishment of strategies for port planning and management. In this paper, we propose a new framework to analyze port cargo volume, which consists of three parts: item segmentation, exploratory data analysis, and time series forecasting specifically for liquid bulk cargo volume. We firstly create an item dictionary containing main keywords to characterize each item and then categorize items based on the dictionary. Next, we perform an exploratory data analysis to understand the volume characteristics of each subcategorized item. Lastly, we use representation learning- and deep learning-based time series techniques to forecast their port volume and compare the results with existing statistical models. Experimental results for the three steps show the usefulness of our novel framework in several aspects including forecasting accuracy. It is believed that our proposed method will be a helpful system for stakeholders in port logistics to have insights and to make better decisions.