Metaheuristic-Guided Deep Learning for Smart Warehouse Anomaly Detection

View Sample PDF

Author(s): Usharani Bhimavarapu (Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, India)
Copyright: 2026
Pages: 26
Source title: Innovations in Green and Energy-Efficient Warehousing
Source Author(s)/Editor(s): Mohamed Amine Frikha (King Faisal University, Saudi Arabia), Mohieddine Rahmouni (King Faisal University, Saudi Arabia)and Ben Othman Soufiane (King Faisal University, Saudi Arabia)
DOI: 10.4018/979-8-3373-3176-8.ch014

Keywords: Energy / Engineering Science Reference / Nuclear & Energy Engineering / Science & Engineering

Purchase

View Metaheuristic-Guided Deep Learning for Smart Warehouse Anomaly Detection on the publisher's website for pricing and purchasing information.

Abstract

This study presents an intelligent approach to enhancing warehouse operations through smart warehousing detection using a Bi-Stacked Long Short-Term Memory (Bi-Stacked LSTM) model. Data were collected from a case study company's internal database, including sales records, inventory data, product categories, and pricing information over the past two years. After rigorous preprocessing involving data cleaning, integration, and transformation, relevant features were selected using the Ant Colony Optimization (ACO) algorithm to reduce dimensionality and improve model performance. The Bi-Stacked LSTM model was trained on the selected features to detect operational anomalies and performance patterns across time-series data. The bidirectional and stacked architecture of the LSTM allowed the model to learn both past and future contextual dependencies effectively.

The IRMA Community

Research IRM

Metaheuristic-Guided Deep Learning for Smart Warehouse Anomaly Detection

Purchase

Abstract

Related Content

IRMA Sponsors