A New Hybrid Approach for Product Management in E-Commerce DOI Creative Commons
Hacire Oya Yüregir, Metin Özşahin, Serap Akcan

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(13), P. 5735 - 5735

Published: July 1, 2024

Nowadays, due to the developments in technology and effects of pandemic, people have largely switched e-commerce instead traditional face-to-face commerce. In this sector, product variety reaches tens thousands, which has made it difficult manage make quick decisions on inventory, promotion, pricing, logistics. Therefore, is thought that obtaining accurate fast forecasting for future will provide significant benefits such companies every respect. This study was built proposal creating a cluster-based–genetic algorithm hybrid model including genetic (GA), cluster analysis, some models as new approach. study, unlike literature, an attempt create more successful many products at same time inside single forecasting. The proposed CBGA success compared separately both prediction method successes only algorithm-based by using real values from popular B2C company. As result, been observed than or algorithm.

Language: Английский

Novel Version of Horse Herd Optimization for Enhancing Electric Load Forecasting Capabilities of Neural Networks DOI
Manvi Mishra,

Priya Mahajan,

Rachana Garg

et al.

Arabian Journal for Science and Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 8, 2025

Language: Английский

Citations

0
Collaborative Forecasting of Multiple Energy Loads in Integrated Energy Systems Based on Feature Extraction and Deep Learning DOI Creative Commons
Zhe Wang, Jiali Duan, Fengzhang Luo

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(5), P. 1048 - 1048

Published: Feb. 21, 2025

Accurate load forecasting is crucial for the safe, stable, and economical operation of integrated energy systems. However, directly applying single models to predict coupled cooling, heating, electric loads under complex influencing factors often yields unsatisfactory results. This paper proposes a collaborative method based on feature extraction deep learning. First, complete ensemble empirical mode decomposition with adaptive noise algorithm decomposes data, dynamic time warping-based k-medoids clustering reconstructs subsequences aligned system components. Second, correlation analysis identifies key model input. Then, multi-task parallel learning framework combining regression convolutional neural network long short-term memory networks developed reconstructed subsequences. Case studies demonstrate that proposed achieves mean absolute percentage errors (MAPE) 2.24%, 2.75%, 1.69% electricity, heating summer workdays, accuracy (MA) values 97.76%, 97.25%, 98.31%, respectively. For winter MAPE are 2.92%, 1.66%, 2.87%, MA 97.08%, 98.34%, 97.13%. Compared traditional single-task models, weighted (WMA) improves by 2.01% 2.33% in winter, respectively, validating its superiority. provides high-precision tool planning

Language: Английский

Citations

0
Internet of Things Applications for Energy Management in Buildings Using Artificial Intelligence—A Case Study DOI Creative Commons
Izabela Rojek, Dariusz Mikołajewski, Adam Mroziński

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(7), P. 1706 - 1706

Published: March 28, 2025

IoT applications for building energy management, enhanced by artificial intelligence (AI), have the potential to transform how is consumed, monitored, and optimized, especially in distributed systems. By using sensors smart meters, buildings can collect real-time data on usage patterns, occupancy, temperature, lighting conditions.AI algorithms then analyze this identify inefficiencies, predict demand, suggest or automate adjustments optimize use. Integrating renewable sources, such as solar panels wind turbines, into systems uses IoT-based monitoring ensure maximum efficiency generation These also enable dynamic pricing load balancing, allowing participate grids storing selling excess energy.AI-based predictive maintenance ensures that systems, inverters batteries, operate efficiently, minimizing downtime. The case studies show AI are driving sustainable development reducing consumption carbon footprints residential, commercial, industrial buildings. Blockchain further secure transactions increasing trust, sustainability, scalability. combination of IoT, AI, sources line with global trends, promoting decentralized greener study highlights adopting management offers not only environmental benefits but economic benefits, cost savings independence. best achieved accuracy was 0.8179 (RMSE 0.01). overall effectiveness rating 9/10; thus, AI-based solutions a feasible, cost-effective, approach office management.

Language: Английский

Citations

0
COMPARATIVE ANALYSIS OF VARIOUS FORECAST MODELS OF ELECTRICITY CONSUMPTION IN SMART BUILDINGS DOI Creative Commons
Akylbek Tokhmetov, Kenzhegali Nurgaliyev, Liliya Tanchenko

et al.

Scientific Journal of Astana IT University, Journal Year: 2025, Volume and Issue: unknown

Published: March 30, 2025

The rapidly growing field of smart building technology depends heavily on accurate electricity consumption forecasting. By anticipating energy demands, managers can optimize resource allocation, minimize waste, and enhance overall efficiency. This study provides a comprehensive comparative analysis various models used to forecast in buildings, highlighting their strengths, limitations, suitability for different use cases. investigation focuses three major categories forecasting models: statistical methods, machine learning techniques, hybrid approaches. Statistical models, such as the Moving Average Method, leverage historical data patterns predict future trends. These enable analysts utilize predictive analytics, simulating real-world environments helping them make more informed decisions. offers detailed comparison several applied Internet Things (IoT) data, with particular emphasis buildings. Among short-term examined are gradient-enhanced regressors (XGBoost), random forest (RF), long memory networks (LSTM). performance these was evaluated based prediction errors identify most one. Time series, learning, considered analyzed. focus is accuracy forecasts applicability conditions, taking into account factors climate change obtained from sensors. shows that combining time series provide best over horizons. It also highlights importance integrating user behavior using IoT technologies improve model accuracy. results be create energy-efficient control systems buildings consumption.

Language: Английский

Citations

0
Optimizing Energy Consumption in IoT Sensors Through Deep Learning-Based Power Management DOI

David Samuel Azariya S.,

V. Mohanraj,

V. Sathiyamoorthi

et al.

Advances in systems analysis, software engineering, and high performance computing book series, Journal Year: 2025, Volume and Issue: unknown, P. 151 - 168

Published: Feb. 7, 2025

The rapid growth of internet things (IoT) devices necessitates efficient power management to curb escalating energy consumption. This chapter proposes a novel solution by employing deep learning techniques optimize use in IoT sensors. authors review existing sensor consumption challenges and conventional limitations. Drawing on learning's successes, they develop an architecture trained curated data. Practical implications span industries, scalability, generalizability diverse setups. Economic insights highlight potential cost savings benefits. In conclusion, the innovative learning-based approach addresses challenges, offering promising that optimizes usage could reshape device efficiency. work opens avenues for hybrid strategies, merging with other techniques, further advancing systems.

Language: Английский

Citations

0
A Novel AI Analytics Driven Forecasting Paradigm for Strategic Inventory Management Through Fuzzy Multi-criteria Decision Model in Supply Chains DOI

Michael Sandra,

Samayan Narayanamoorthy,

Krishnan Suvitha

et al.

International Journal of Fuzzy Systems, Journal Year: 2025, Volume and Issue: unknown

Published: April 19, 2025

Language: Английский

Citations

0
A Hybrid Mud Ring Algorithm and Continual Spatio-Temporal Graph Convolutional Networks for Short-Term Residential Load Forecasting in Smart Grids DOI

B. Suresh Babu,

Rajagopalan Thiruvengadathan,

P. Santhosh Srinivasan

et al.

Smart Grids and Sustainable Energy, Journal Year: 2025, Volume and Issue: 10(2)

Published: April 23, 2025

Language: Английский

Citations

0
Explainable knowledge graph embeddings for industrial process monitoring & control DOI
Michael Weyns, Thibault Blyau, Bram Steenwinckel

et al.

Information Fusion, Journal Year: 2025, Volume and Issue: unknown, P. 103242 - 103242

Published: May 1, 2025

Language: Английский

Citations

0
An overview of Artificial Intelligence applications to electrical power systems and DC microgrids DOI Creative Commons
M. Rajitha,

A. Raghu Ram

E3S Web of Conferences, Journal Year: 2024, Volume and Issue: 547, P. 01002 - 01002

Published: Jan. 1, 2024

Microgrids are composed of distributed energy resources such as storage devices, photovoltaic (PV) systems, backup generators, and wind conversion systems. Because renewable sources intermittent, modern power networks must overcome the stochastic problem increasing penetration energy, which necessitates precise demand forecasting to deliver best possible supply. Technologies based on artificial intelligence (AI) have become a viable means implementing optimizing microgrid management. Owing sporadic nature sources, offers range solutions growth in sensor data compute capacity create sustainable dependable power. Artificial techniques continue evolve DC with aim perfect voltage profile, minimum distribution losses, optimal schedule power, planning controlling grid parameters lowering unit price. AI methods can improve Micro performance by monitoring reducing computational processing time. This paper comprehensive summary some most recent research used grids electrical system networks.

Language: Английский

Citations

3
Advanced Integration of Forecasting Models for Sustainable Load Prediction in Large-Scale Power Systems DOI Open Access
Jiansong Tang, Ryosuke Saga, Hanbo Cai

et al.

Sustainability, Journal Year: 2024, Volume and Issue: 16(4), P. 1710 - 1710

Published: Feb. 19, 2024

In the burgeoning field of sustainable energy, this research introduces a novel approach to accurate medium- and long-term load forecasting in large-scale power systems, critical component for optimizing energy distribution reducing environmental impacts. This study breaks new ground by integrating Causal Convolutional Neural Networks (Causal CNN) Variational Autoencoders (VAE), among other advanced models, surpassing conventional methodologies domain. Methodologically, these cutting-edge models is harnessed assimilate analyze wide array influential factors, including economic trends, demographic shifts, natural phenomena. enables more nuanced comprehensive understanding dynamics, essential forecasting. The results demonstrate remarkable improvement accuracy, with 15% increase precision over traditional models. Additionally, robustness under varying conditions showcases significant advancement predicting loads reliably. conclusion, findings not only contribute substantially but also highlight pivotal role innovative promoting practices. work establishes foundational framework future addressing immediate challenges exploring potential avenues system management.

Language: Английский

Citations

2