Hybrid LSTM-ARMA Demand-Forecasting Model Based on Error Compensation for Integrated Circuit Tray Manufacturing

Wang, Chien-Chih; Chang, Hsin-Tzu; Chien, Chun-Hua

doi:10.3390/math10132158

Cited by 12 publications

(4 citation statements)

References 25 publications

(24 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Drawing inspiration from this research, we incorporated considerations of deviations between forecasted and actual demand in our model construction to mitigate decision biases.Furthermore, Kleywegt et al [8] employed a sample average approximation method, replacing the assumption of demand distribution with empirical data. They discussed the convergence speed, stopping rules, and computational complexity of this approach.Wang et al [9] introduced an error compensation mechanism for demand forecasting and assessed the necessity of compensation using individual and moving range (I-MR) control charts. This evaluation aimed to leverage predictive models in addressing current bottleneck issues.…”

Section: Demand Forcastingmentioning

confidence: 99%

Supply Chain Inventory Management from the Perspective of “Cloud Supply Chain”—A Data Driven Approach

Tan,

Gu,

et al. 2024

Mathematics

View full text Add to dashboard Cite

This study systematically investigates the pivotal role of inventory management within the framework of “cloud supply chain” operations, emphasizing the efficacy of leveraging machine learning methodologies for inventory allocation with the dual objectives of cost reduction and heightened customer satisfaction. Employing a rigorous data-driven approach, the research endeavors to address inventory allocation challenges inherent in the complex dynamics of a “cloud supply chain” through the implementation of a two-stage model. Initially, machine learning is harnessed for demand forecasting, subsequently refined through the empirical distribution of forecast errors, culminating in the optimization of inventory allocation across various service levels.The empirical evaluation draws upon data derived from a reputable home appliance logistics company in China, revealing that, under conditions of ample data, the application of data-driven methods for inventory allocation surpasses the performance of traditional methods across diverse supply chain structures. Specifically, there is an improvement in accuracy by approximately 13% in an independent structure and about 16% in a dependent structure. This study transcends the constraints associated with examining a singular node, adopting an innovative research perspective that intricately explores the interplay among multiple nodes while elucidating the nuanced considerations germane to supply chain structure. Furthermore, it underscores the methodological significance of relying on extensive, large-scale data. The investigation brings to light the substantial impact of supply chain structure on safety stock allocation. In the context of a market characterized by highly uncertain demand, the strategic adaptation of the supply chain structure emerges as a proactive measure to avert potential disruptions in the supply chain.

show abstract

Section: Demand Forcastingmentioning

confidence: 99%

Supply Chain Inventory Management from the Perspective of “Cloud Supply Chain”—A Data Driven Approach

Tan,

Gu,

et al. 2024

Mathematics

View full text Add to dashboard Cite

show abstract

“…Therefore, any model can use a time series control chart if the residuals fulfill the white noise's assumption (Qiu et al, 2020). A control chart to be used in this condition is the IMR chart, and its calculation is shown in Table 1 (Wang et al, 2022).…”

Section: 𝑈𝐶𝐿 = 𝜇 𝑤 + 𝑘𝜎 𝑤mentioning

confidence: 99%

Control Chart for Correcting the ARIMA Time Series Model of GDP Growth Cases

Imro'ah,

Huda,

Utami

et al. 2024

JTAM

View full text Add to dashboard Cite

The essential prerequisite for attending the G20 conference is a country's GDP because G20 members can significantly boost the economy and preserve the nation's financial stability. Time series data can be thought of as a country's Gross Domestic Product (GDP) at a particular point in time. In this research, the GDP numbers from five Southeast Asian nations that are attending the G20 fulfilling are used. The total was 47 observations made yearly, which extended from 1975 to 2001. A time series analysis was performed on the data gathered. The correctness of time series models is also evaluated using control charts based on this research. The control chart is constructed using the time series model's residuals as observations. After applying the IMR control chart for verification, the results revealed that the residuals, specifically the models for GDP in Malaysia, Singapore, and Thailand, are out of control. The white noise assumption is fulfilled by the time series model obtained for Brunei and Indonesia's GDP, but the residuals are out of control. Whether controlled residuals are used depends on the accuracy with which the time series model predicts the future. If the amount of residuals is under control, then the time series model produced is accurate and good enough for prediction. After using the IMR control chart to verify the residuals, the results indicate that the residuals, namely the models for GDP in Malaysia, Singapore, and Thailand, are not under control. The assumption of white noise is proved correct by the time series model obtained for the GDP of Brunei Darussalam and Indonesia. With that being said, the residuals are entirely out of control. The model must improve its ability to forecast various future periods. It is a consequence of the unmanageable residuals that the model contains. Even if the best available model has been obtained based on the criteria that have been defined, it is anticipated that the research findings will improve the theories that have previously been developed and raise knowledge regarding the usefulness of testing the time series model. In addition to all of that, it is intended that the research will produce a summary of cases of an increase in GDP from five Southeast Asian countries participating in the G20 conference.

show abstract

“…MAE gives equal weight to all errors; it may not penalise large errors as heavily as MSE. This can be a disadvantage when large errors need to be minimised or when they are particularly costly [41][42][43].…”

Section: Error Metricsmentioning

confidence: 99%

Prediction of Wind Power with Machine Learning Models

Karaman

2023

Applied Sciences

View full text Add to dashboard Cite

Wind power is a vital power grid component, and wind power forecasting represents a challenging task. In this study, a series of multiobjective predictive models were created utilising a range of cutting-edge machine learning (ML) methodologies, namely, artificial neural networks (ANNs), recurrent neural networks (RNNs), convolutional neural networks, and long short-term memory (LSTM) networks. In this study, two independent data sets were combined and used to predict wind power. The first data set contained internal values such as wind speed (m/s), wind direction (°), theoretical power (kW), and active power (kW). The second data set was external values that contained the meteorological data set, which can affect the wind power forecast. The k-nearest neighbours (kNN) algorithm completed the missing data in the data set. The results showed that the LSTM, RNN, CNN, and ANN algorithms were powerful in forecasting wind power. Furthermore, the performance of these models was evaluated by incorporating statistical indicators of performance deviation to demonstrate the efficacy of the employed methodology effectively. Moreover, the performance of these models was evaluated by incorporating statistical indicators of performance deviation, including the coefficient of determination (R2), root mean square error (RMSE), mean absolute error (MAE), and mean square error (MSE) metrics to effectively demonstrate the efficacy of the employed methodology. When the metrics are examined, it can be said that ANN, RNN, CNN, and LSTM methods effectively forecast wind power. However, it can be said that the LSTM model is more successful in estimating the wind power with an R2 value of 0.9574, MAE of 0.0209, MSE of 0.0038, and RMSE of 0.0614.

show abstract

Hybrid LSTM-ARMA Demand-Forecasting Model Based on Error Compensation for Integrated Circuit Tray Manufacturing

Cited by 12 publications

References 25 publications

Supply Chain Inventory Management from the Perspective of “Cloud Supply Chain”—A Data Driven Approach

Supply Chain Inventory Management from the Perspective of “Cloud Supply Chain”—A Data Driven Approach

Control Chart for Correcting the ARIMA Time Series Model of GDP Growth Cases

Prediction of Wind Power with Machine Learning Models

Contact Info

Product

Resources

About