Optimization & Operations Research
Niousha Zeidyahyaee; Sajjad Shokouhyar; Alireza Motameni
Abstract
This study develops a mathematically informed optimization framework for decision-making in reverse supply chain management, with an application to Apple’s MacBook product line. The proposed framework integrates Failure Mode and Effects Analysis (FMEA) with deep learning, based sentiment analysis ...
Read More
This study develops a mathematically informed optimization framework for decision-making in reverse supply chain management, with an application to Apple’s MacBook product line. The proposed framework integrates Failure Mode and Effects Analysis (FMEA) with deep learning, based sentiment analysis in a multi-stage structure designed to quantify risk factors and predict consumer-driven outcomes. The dataset consists of 91 days of Twitter user feedback on Apple notebooks, processed using supervised learning algorithms to extract sentiment scores and thematic indicators of product performance. The analysis identifies “power and battery” and “storage” as the most critical components contributing to user dissatisfaction and elevated risk severity. These data-driven insights are incorporated into an optimization model that supports decisions on product recycling, refurbishment, and reuse. The hybrid framework enhances decision stability and accuracy compared with conventional reverse logistics models, while improving operational efficiency and environmental performance. The results demonstrate the model’s suitability as a scalable, machine-learning-supported optimization tool for reverse supply chain systems.
Machine Learning & Data Science in Optimization
Saeed Mirzajani; Majid Roohi
Abstract
The prediction of chaotic time series is essential for understanding highly nonlinear and sensitive systems, with the Lorenz system serving as a standard benchmark due to its intricate and non-periodic dynamics. Classical forecasting approaches often struggle to capture such irregularities, ...
Read More
The prediction of chaotic time series is essential for understanding highly nonlinear and sensitive systems, with the Lorenz system serving as a standard benchmark due to its intricate and non-periodic dynamics. Classical forecasting approaches often struggle to capture such irregularities, motivating a shift toward deep learning–based strategies. In this study, we develop two hybrid models—Feedback Long Short-Term Memory (FB-LSTM) and Feedback Variational Stacked LSTM (FBVS-LSTM), specifically designed for multivariate prediction of the Lorenz system. By embedding feedback structures into LSTM networks, the proposed methods deliver enhanced short-term prediction performance without substantial computational costs. Comparative simulations indicate that our frameworks surpass traditional RNNs and baseline LSTM models, achieving prediction accuracies up to 94%. These findings indicate that feedback-enhanced architectures offer effective and practical tools for forecasting chaotic systems, with potential applications in both scientific research and engineering practice.
)
)