• 무역학회지
• /
• 제47권6호
• /
• pp.291-305
• /
• 2022

In this study, the volume of Pohang Port was predicted. All cargo of Pohang port, iron ore, steel, and bituminous coals were selected as prediction targets. SARIMA, Prophet, and Neural Prophet were used as analysis methods. The predictive power of each model was verified, and a predictive model with high performance was used to predict the volume of goods in Pohang port. As a result of the analysis, it was found that Neural Prophet showed the highest performance in all predictive power. As a result of predicting the future volume of goods until August 2027 using Neural Prophet, it was found that the volume of all items in Pohang port was decreasing. In particular, it was analyzed that the decline in steel cargo was steep. In order to increase the volume of cargo at Pohang port, it is necessary to diversify the cargo handled at Pohang port and check the policy of increasing the volume of cargo.

• 한국컴퓨터정보학회논문지
• /
• 제28권7호
• /
• pp.29-37
• /
• 2023

기후변화로 인해 농작물 생산과 유통에 관한 관심이 증대되고 있고, 빅데이터와 AI를 활용한 생산량 예측을 통해 농가의 출하량 조절과 유통단계의 조절에 활용하는 시도가 이루어지고 있다. 농산물 반입량 예측은 가격에 영향을 미칠 뿐 아니라 농가의 출하량과 유통회사의 유통량 조절을 할 수 있으므로 마케팅 전략을 수립하는데 중요한 정보이다. 본 연구에서는 농업 통계 정보 시스템에서 공개한 도매시장 참외 반입량 데이터를 기반으로 미래의 반입량을 예측하는 인공지능 예측 모델을 생성하고 정확도를 평가한다. 우리는 Neural Prophet 기법과 Ensembled Neural Prophet 모델 그리고 GRU 모델 등 세 가지 모델을 사용하여 예측 모델을 생성한다. 모델의 성능은 MAE와 RMSE라는 두 가지 주요 지표를 비교하여 평가한 결과 Ensembled Neural Prophet 모델이 가장 정확하게 예측하였으며, GRU 모델도 앙상블 모델과 유사한 성능을 보여주고 있다. 본 연구에서 개발된 모형은 웹에 publish 되어 현장에서 1년 6개월 동안 사용하고 있으며, 가까운 미래의 참외 생산량을 예측하고, 마케팅 및 유통전략을 수립하는 데 활용되고 있다.

• 산업경영시스템학회지
• /
• 제46권3호
• /
• pp.123-138
• /
• 2023

The Fourth Industrial Revolution and sensor technology have led to increased utilization of sensor data. In our modern society, data complexity is rising, and the extraction of valuable information has become crucial with the rapid changes in information technology (IT). Recurrent neural networks (RNN) and long short-term memory (LSTM) models have shown remarkable performance in natural language processing (NLP) and time series prediction. Consequently, there is a strong expectation that models excelling in NLP will also excel in time series prediction. However, current research on Transformer models for time series prediction remains limited. Traditional RNN and LSTM models have demonstrated superior performance compared to Transformers in big data analysis. Nevertheless, with continuous advancements in Transformer models, such as GPT-2 (Generative Pre-trained Transformer 2) and ProphetNet, they have gained attention in the field of time series prediction. This study aims to evaluate the classification performance and interval prediction of remaining useful life (RUL) using an advanced Transformer model. The performance of each model will be utilized to establish a health index (HI) for cutting blades, enabling real-time monitoring of machine health. The results are expected to provide valuable insights for machine monitoring, evaluation, and management, confirming the effectiveness of advanced Transformer models in time series analysis when applied in industrial settings.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권3호
• /
• pp.528-550
• /
• 2024

Warehousing demand prediction is an essential part of the supply chain, providing a fundamental basis for product manufacturing, replenishment, warehouse planning, etc. Existing forecasting methods cannot produce accurate forecasts since warehouse demand is affected by external factors such as holidays and seasons. Some aspects, such as consumer psychology and producer reputation, are challenging to quantify. The data can fluctuate widely or do not show obvious trend cycles. We introduce a new model for warehouse demand prediction called MAGRU, which stands for Multi-layer Attention with GRU. In the model, firstly, we perform the embedding operation on the input sequence to quantify the external influences; after that, we implement an encoder using GRU and the attention mechanism. The hidden state of GRU captures essential time series. In the decoder, we use attention again to select the key hidden states among all-time slices as the data to be fed into the GRU network. Experimental results show that this model has higher accuracy than RNN, LSTM, GRU, Prophet, XGboost, and DARNN. Using mean absolute error (MAE) and symmetric mean absolute percentage error(SMAPE) to evaluate the experimental results, MAGRU's MAE, RMSE, and SMAPE decreased by 7.65%, 10.03%, and 8.87% over GRU-LSTM, the current best model for solving this type of problem.