Deep reinforcement learning in smart manufacturing: A review and prospects

CIRP journal of manufacturing science and technology, Journal Year: 2022, Volume and Issue: 40, P. 75 - 101

Published: Dec. 2, 2022

Language: Английский

---

A Systematic Study on Reinforcement Learning Based Applications

Energies, Journal Year: 2023, Volume and Issue: 16(3), P. 1512 - 1512

Published: Feb. 3, 2023

We have analyzed 127 publications for this review paper, which discuss applications of Reinforcement Learning (RL) in marketing, robotics, gaming, automated cars, natural language processing (NLP), internet things security, recommendation systems, finance, and energy management. The optimization use is critical today’s environment. mainly focus on the RL application Traditional rule-based systems a set predefined rules. As result, they may become rigid unable to adjust changing situations or unforeseen events. can overcome these drawbacks. learns by exploring environment randomly based experience, it continues expand its knowledge. Many researchers are working RL-based management (EMS). utilized such as optimizing smart buildings, hybrid automobiles, grids, managing renewable resources. contributes achieving net zero carbon emissions sustainable In context technology, be optimize regulation building heating, ventilation, air conditioning (HVAC) reduce consumption while maintaining comfortable atmosphere. EMS accomplished teaching an agent make judgments sensor data, temperature occupancy, modify HVAC system settings. has proven beneficial lowering usage buildings active research area buildings. used electric vehicles (HEVs) learning optimal control policy maximize battery life fuel efficiency. acquired remarkable position gaming applications. majority security-related operate simulated recommender provide good suggestions accuracy diversity. This article assists novice comprehending foundations reinforcement

Language: Английский

---

Machine learning scopes on microgrid predictive maintenance: Potential frameworks, challenges, and prospects

Renewable and Sustainable Energy Reviews, Journal Year: 2023, Volume and Issue: 190, P. 114088 - 114088

Published: Nov. 16, 2023

Language: Английский

---

Reinforcement and deep reinforcement learning-based solutions for machine maintenance planning, scheduling policies, and optimization

Journal of Manufacturing Systems, Journal Year: 2023, Volume and Issue: 70, P. 244 - 263

Published: Aug. 5, 2023

Language: Английский

---

Paradigm Shift for Predictive Maintenance and Condition Monitoring from Industry 4.0 to Industry 5.0: A Systematic Review, Challenges and Case Study

Results in Engineering, Journal Year: 2024, Volume and Issue: unknown, P. 102935 - 102935

Published: Sept. 1, 2024

Language: Английский

---

A Review of Deep Reinforcement Learning Approaches for Smart Manufacturing in Industry 4.0 and 5.0 Framework

Applied Sciences, Journal Year: 2022, Volume and Issue: 12(23), P. 12377 - 12377

Published: Dec. 3, 2022

In this review, the industry’s current issues regarding intelligent manufacture are presented. This work presents status and potential for I4.0 I5.0’s revolutionary technologies. AI and, in particular, DRL algorithms, which a perfect response to unpredictability volatility of modern demand, studied detail. Through introduction RL concepts development those with ANNs towards DRL, variety these kinds algorithms highlighted. Moreover, because data based, their modification meet requirements industry operations is also included. addition, review covers inclusion new concepts, such as digital twins, an absent environment model how it can improve performance application even more. highlights that applicability demonstrated across all manufacturing operations, outperforming conventional methodologies most notably, enhancing process’s resilience adaptability. It stated there still considerable be carried out both academia fully leverage promise disruptive tools, begin deployment industry, take step closer I5.0 industrial revolution.

Language: Английский

---

Q-learning driven multi-population memetic algorithm for distributed three-stage assembly hybrid flow shop scheduling with flexible preventive maintenance

Expert Systems with Applications, Journal Year: 2023, Volume and Issue: 232, P. 120837 - 120837

Published: June 19, 2023

Language: Английский

---

Machine Learning in Manufacturing towards Industry 4.0: From ‘For Now’ to ‘Four-Know’

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(3), P. 1903 - 1903

Published: Feb. 1, 2023

While attracting increasing research attention in science and technology, Machine Learning (ML) is playing a critical role the digitalization of manufacturing operations towards Industry 4.0. Recently, ML has been applied several fields production engineering to solve variety tasks with different levels complexity performance. However, spite enormous number use cases, there no guidance or standard for developing solutions from ideation deployment. This paper aims address this problem by proposing an application roadmap industry based on state-of-the-art published topic. First, presents two dimensions formulating tasks, namely, ’Four-Know’ (Know-what, Know-why, Know-when, Know-how) ’Four-Level’ (Product, Process, Machine, System). These are used analyze development trends manufacturing. Then, provides implementation pipeline starting very early stages solution summarizes available methods, including supervised learning semi-supervised unsupervised reinforcement along their typical applications. Finally, discusses current challenges during applications outline possible directions future developments.

Language: Английский

---

Dynamic maintenance scheduling approach under uncertainty: Comparison between reinforcement learning, genetic algorithm simheuristic, dispatching rules

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: 248, P. 123404 - 123404

Published: Feb. 8, 2024

Maintenance planning and scheduling are an essential part of manufacturing companies to prevent machine breakdowns increase uptime, along with production efficiency. One the biggest challenges is effectively address uncertainty (e.g., unexpected failures, variable time repair). Multiple approaches have been used solve maintenance problem, including dispatching rules (DR), metaheuristics simheuristics, or most recently reinforcement learning (RL). However, best our knowledge, no study has ever studied what extent these techniques effective when faced different levels uncertainty. To overcome this gap in research, paper presents approach by analyzing impact categorized uncertainty, specifically high low, on failure distribution repair. Upon formalization experiments conducted performed simulated scenarios degrees also considering a real-life use case. The results indicate that rescheduling based genetic algorithm (GA) simheuristic outperforms RL DR terms total but not mean repair configured re-optimization frequencies (i.e., hourly re-optimization), rapidly underperforms frequency decreases. Furthermore, demonstrates GA-simheuristic highly computationally demanding compared rule-based policies.

Language: Английский

---

Strategies for overcoming data scarcity, imbalance, and feature selection challenges in machine learning models for predictive maintenance

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: April 26, 2024

Predictive maintenance harnesses statistical analysis to preemptively identify equipment and system faults, facilitating cost- effective preventive measures. Machine learning algorithms enable comprehensive of historical data, revealing emerging patterns accurate predictions impending failures. Common hurdles in applying ML PdM include data scarcity, imbalance due few failure instances, the temporal dependence nature data. This study proposes an ML-based approach that adapts these through generation synthetic feature extraction, creation horizons. The employs Generative Adversarial Networks generate LSTM layers extract features. trained on generated achieved high accuracies: ANN (88.98%), Random Forest (74.15%), Decision Tree (73.82%), KNN (74.02%), XGBoost (73.93%).

Language: Английский

---