Company X, a leader in the Life Sciences supply chain, specializes in producing high-quality medical and laboratory equipment. Given the dynamic market influenced by trends, seasonal variations, economic shifts, and global events such as pandemics, traditional forecasting methods often fall short in accuracy. This work seeks to enhance demand forecasting for highly demanded products by implementing advanced time series models and developing a comprehensive planning model that integrates these methods. Utilizing the Cross-Industry Standard Process for Data Mining (CRISP-DM) framework, the work systematically approaches the data-driven challenges of demand and supply management. Each product time series is first statistically classified to determine the presence of seasonality, guiding the selection of appropriate forecasting models. For non-seasonal products, ARIMA, Holt’s Linear, ARIMA-SVR, and Facebook Prophet were evaluated, while seasonal series employed SARIMAX, Holt-Winters, SARIMAX-SVR, and Prophet models. The results indicate that the ARIMA-SVR and SARIMAX-SVR models consistently out-perform others, reducing forecasting errors by approximately 35.3
Modern deep learning optimization faces persistent challenges in computational efficiency and convergence stability due to complex loss landscapes and heterogeneous parameter dynamics across architectures like CNNs, SNNs, and Transformers. While existing adaptive optimizers (e.g., AdamW, Lion) provide partial solutions, they inadequately address inter-parameter dependencies and neural adaptation mechanisms. Inspired by biological principles of synaptic plasticity and cortical organization, we propose PANDA (Parameter Adaptation through Neural Dynamics Alignment), a novel optimization framework that enhances convergence stability and model generalization through three key innovations: 1) Gradient similarity graphs modeling parameter interdependencies; 2) Dynamic importance quantification enabling context-aware parameter updates; and 3) Region-based grouping balancing local-global optimization dynamics. Comprehensive evaluations demonstrate PANDA's superiority over state-of-the-art methods, achieving 73.3
This paper proposes an unpaired shadow removal method based on the CycleGAN framework, in which the generator is structurally enhanced by introducing skip connections. Existing CycleGAN-based methods commonly rely on plain encoder-decoder generators originally tailored for style transfer, which often fail to preserve pixel-level spatial alignment and low-level structural details. This limitation leads to incomplete shadow removal, texture degradation, and inconsistent illumination in the restored images. To overcome these issues, the proposed generator introduces skip connections that inject low-level features directly from the encoder to the decoder. This architectural design helps preserve structural integrity and improve visual coherence across the entire image. By leveraging both global and local information, the network achieves a more accurate reconstruction of shadow-free images. The method is evaluated on the public ISTD dataset. Quantitative results show that it consistently outperforms existing unpaired models including CycG, Mask-CycG, and DC-CycG, measured by root mean square error (RMSE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM). Visual comparisons further confirm fewer artifacts, more complete shadow removal, and more consistent illumination in the restored regions. Feature map analysis further confirms that the skip-connected architecture captures richer and more diverse semantic features, facilitating better decoding. These findings highlight the practical and theoretical value of enhancing CycleGAN-based generator with skip connections, offering an effective solution for high-quality unpaired shadow removal.
This study evaluates the sediment control performance of two sewage pump stations (PS) configurations, PS-1 and PS-2, in Langat Kajang, Malaysia, under varying rainfall conditions. The primary focus is on the removal efficiencies of Total Suspended Solids (TSS) and Volatile Suspended Solids (VSS). Over 100 days (24 July 2023–31 October 2023), operational data were analyzed using Principal Component Analysis (PCA) and Agglomerative Hierarchical Clustering (AHC). PCA results revealed that the first two principal components accounted for 77.83
Accurate river water level prediction is vital for early flood warning and effective water resource management. The traditional methods of river water level prediction based on Autoregressive Integrated Moving Average (ARIMA) and linear regression struggled to capture the nonlinear and dynamic nature of hydrological systems. Therefore, this study evaluates the performance of deep learning models, namely Long Short-Term Memory (LSTM), Basic LSTM, Stacked LSTM, Bidirectional LSTM, and LSTM with Dropout, alongside classical models such as Linear Regression, Moving Average, and ARIMA across five river monitoring stations. Model performance was assessed using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Mean Squared Error (MSE), and R-squared (R2) metrics on test data. The results consistently showed that the LSTM with Dropout model outperformed all other approaches in terms of predictive accuracy and generalization, achieving the lowest RMSE and MAE values and the highest R2 scores at four out of five stations. The standard LSTM model achieved the best performance at one station but was generally outperformed by its dropout-regularized counterpart. In contrast, classical models exhibited significantly lower performance, particularly in capturing temporal dependencies, with R2 values often below zero. These findings validated the effectiveness and robustness of LSTM-based deep learning, particularly with dropout regularization, in modeling and forecasting river water level dynamics across diverse hydrological conditions. The study supports the integration of advanced machine learning models in river monitoring systems to enhance flood preparedness and mitigation strategies.
