Mohammad Hafiz Ismail
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Mohammad Hafiz Ismail has a total of 59 co-authors. Here are the top co-authors by number of shared publications:
- Tajul Rosli Razak
Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA, Shah Alam, Malaysia
6 shared publications - Nurul Ain Mohd Zaki
Universiti Teknologi MARA
5 shared publications - Shukor Sanim Mohd Fauzi
Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA, Perlis Branch, Arau Campus, Arau, Perlis
5 shared publications - Arifah Fasha Rosmani
Universiti Teknologi MARA
4 shared publications - Aznoora Osman
Universiti Teknologi MARA (UiTM) Perlis
4 shared publications - Hasila Jarimi
The University of Nottingham
4 shared publications
Latest Publications
- Open Access
- 17 February 2026
- Published by Caravel Press in Artificial Intelligence for Sustainable Cities
- Vol. 1 (1) , 49-70
An AI-Based Hybrid Prophet–LSTM Model for Forecasting and Financial Optimisation in Sustainable Energy Grids
Accurate and reliable forecasting is critical for sustainable energy grid planning, as fluctuating demand, market volatility, and policy uncertainty pose significant challenges to both operational and financial stability. This paper proposes a novel AI-based hybrid forecasting framework that integrates Facebook Prophet, for interpretable long-term trend and seasonality decomposition, with Long Short-Term Memory (LSTM) networks, for modelling nonlinear short-term residual dynamics. This design explicitly addresses the dual requirement of transparency and high predictive accuracy in complex, non-stationary energy systems. The framework is evaluated using 18 years of historical financial data from Tenaga Nasional Berhad (TNB), Malaysia’s largest electricity utility, as a representative large-scale grid operation case study. Performance is benchmarked against ARIMA, standalone Prophet, and standalone LSTM models using RMSE, MAE, MAPE, and SMAPE, with statistical significance assessed via the Diebold–Mariano test and robustness examined under varying forecast horizons and noise perturbations. Results show that the proposed hybrid Prophet–LSTM model achieves up to 15% lower RMSE and MAPE than the best-performing baseline while maintaining stable performance under adverse conditions. The findings demonstrate that the proposed framework provides a robust, interpretable, and modular decision-support tool for utility operators, energy planners, and policymakers, enabling improved financial optimisation, tariff planning, and operational resilience in sustainable energy grid systems.
- 17 February 2026
- Published by Caravel Press in Artificial Intelligence for Sustainable Cities
- Vol. 1 (1) , 49-70
Open AccessAn AI-Based Hybrid Prophet–LSTM Model for Forecasting and Financial Optimisation in Sustainable Energy Grids
Accurate and reliable forecasting is critical for sustainable energy grid planning, as fluctuating demand, market volatility, and policy uncertainty pose significant challenges to both operational and financial stability. This paper proposes a novel AI-based hybrid forecasting framework that integrates Facebook Prophet, for interpretable long-term trend and seasonality decomposition, with Long Short-Term Memory (LSTM) networks, for modelling nonlinear short-term residual dynamics. This design explicitly addresses the dual requirement of transparency and high predictive accuracy in complex, non-stationary energy systems. The framework is evaluated using 18 years of historical financial data from Tenaga Nasional Berhad (TNB), Malaysia’s largest electricity utility, as a representative large-scale grid operation case study. Performance is benchmarked against ARIMA, standalone Prophet, and standalone LSTM models using RMSE, MAE, MAPE, and SMAPE, with statistical significance assessed via the Diebold–Mariano test and robustness examined under varying forecast horizons and noise perturbations. Results show that the proposed hybrid Prophet–LSTM model achieves up to 15% lower RMSE and MAPE than the best-performing baseline while maintaining stable performance under adverse conditions. The findings demonstrate that the proposed framework provides a robust, interpretable, and modular decision-support tool for utility operators, energy planners, and policymakers, enabling improved financial optimisation, tariff planning, and operational resilience in sustainable energy grid systems.
- Open Access
- 12 December 2025
- Published by Universitas Negeri Semarang in Jurnal Pendidikan Fisika Indonesia
- Vol. 21 (2) , 209-217
Teaching and Learning Hooke’s Law Using Physical, Interactive Software and Blended Approach: An Overview
At present, with technological advancement and changes in students’ preferences, there are both challenges and opportunities to enhance the teaching and learning of subjects in science, technology, engineering, and mathematics (STEM). In this article, we explore reported works in established databases on the topic of Hooke’s Law, to see the approaches used in teaching and learning of the topic. Searches were performed in Web of Science, Scopus and Google Scholar databases, after which manual filtering for relevant articles was carried out by reading and browsing through the abstracts. Based on the types of works, the articles were separated into three parts: physical, interactive software, and blended. In total, 21 relevant articles were reviewed. The data obtained can be separated into three categories: physical, interactive software, and blended. Of the three, blended approaches were used the least. To further explore the topics discussed in the articles related to Hooke’s Law, a bibliometric study was performed. From the results, we found that, with the advantages offered by the blended approach, and none using artificial intelligence, there is still room for improvement to enhance the teaching of Hooke’s Law, specifically, and generally, on introductory physics subjects.
- 12 December 2025
- Published by Universitas Negeri Semarang in Jurnal Pendidikan Fisika Indonesia
- Vol. 21 (2) , 209-217
Open AccessTeaching and Learning Hooke’s Law Using Physical, Interactive Software and Blended Approach: An Overview
At present, with technological advancement and changes in students’ preferences, there are both challenges and opportunities to enhance the teaching and learning of subjects in science, technology, engineering, and mathematics (STEM). In this article, we explore reported works in established databases on the topic of Hooke’s Law, to see the approaches used in teaching and learning of the topic. Searches were performed in Web of Science, Scopus and Google Scholar databases, after which manual filtering for relevant articles was carried out by reading and browsing through the abstracts. Based on the types of works, the articles were separated into three parts: physical, interactive software, and blended. In total, 21 relevant articles were reviewed. The data obtained can be separated into three categories: physical, interactive software, and blended. Of the three, blended approaches were used the least. To further explore the topics discussed in the articles related to Hooke’s Law, a bibliometric study was performed. From the results, we found that, with the advantages offered by the blended approach, and none using artificial intelligence, there is still room for improvement to enhance the teaching of Hooke’s Law, specifically, and generally, on introductory physics subjects.
- Open Access
- 15 November 2025
- Published by Atlantis Press SARL
- p. 368-375
Meeting Teachers’ and Students’ Needs: Developing a Contextual Teaching-Based Digital Assessment Model to Enhance Higher-Order Thinking Skills (HOTS)
- 15 November 2025
- Published by Atlantis Press SARL
- p. 368-375
Open AccessMeeting Teachers’ and Students’ Needs: Developing a Contextual Teaching-Based Digital Assessment Model to Enhance Higher-Order Thinking Skills (HOTS)
- 11 September 2025
- Published by Elsevier in Energy and Buildings
- Vol. 348
Data-driven thermal comfort modeling: Comparing AI-based predictions with PMV-PPD models
- 11 September 2025
- Published by Elsevier in Energy and Buildings
- Vol. 348
Data-driven thermal comfort modeling: Comparing AI-based predictions with PMV-PPD models
- Open Access
- 1 August 2025
- Published by Caravel Press in Research and Reviews in Sustainability
- Vol. 1 (1) , 1-22
Artificial Intelligence in Renewable Energy: A Systematic Review of Trends in Solar, Wind, and Smart Grid Applications
Artificial Intelligence (AI) revolutionizes the renewable energy sector by enabling advanced forecasting, real-time optimization, and autonomous system control. As global efforts toward decarbonization intensify, AI applications have rapidly expanded across various renewable energy domains. However, the literature remains fragmented, lacking a focused synthesis of evolving AI techniques and their domain-specific implementations. This study addresses this gap by systematically reviewing AI applications in three critical energy sectors: Solar Energy, Wind Energy, and Energy Storage & Smart Grids. Using the PRISMA methodology, peer-reviewed articles published between 2015 and 2025 were extracted from two authoritative databases—IEEE Xplore and ScienceDirect. The selected studies were classified based on AI methods, including machine learning, deep learning, reinforcement learning, fuzzy logic, and emerging paradigms such as explainable AI (XAI), generative AI, graph neural networks (GNNs), and physics-informed neural networks (PINNs). Key contributions of this review include a cross-source comparative analysis, domain-specific trend mapping over a decade, and the identification of gaps in methodological transparency. Findings reveal increasing use of hybrid models, growing interest in interpretable and physically grounded techniques, and persistent underreporting of AI methodologies in the literature. This review provides actionable insights and research directions toward developing intelligent, explainable, sustainable energy systems.
- 1 August 2025
- Published by Caravel Press in Research and Reviews in Sustainability
- Vol. 1 (1) , 1-22
Open AccessArtificial Intelligence in Renewable Energy: A Systematic Review of Trends in Solar, Wind, and Smart Grid Applications
Artificial Intelligence (AI) revolutionizes the renewable energy sector by enabling advanced forecasting, real-time optimization, and autonomous system control. As global efforts toward decarbonization intensify, AI applications have rapidly expanded across various renewable energy domains. However, the literature remains fragmented, lacking a focused synthesis of evolving AI techniques and their domain-specific implementations. This study addresses this gap by systematically reviewing AI applications in three critical energy sectors: Solar Energy, Wind Energy, and Energy Storage & Smart Grids. Using the PRISMA methodology, peer-reviewed articles published between 2015 and 2025 were extracted from two authoritative databases—IEEE Xplore and ScienceDirect. The selected studies were classified based on AI methods, including machine learning, deep learning, reinforcement learning, fuzzy logic, and emerging paradigms such as explainable AI (XAI), generative AI, graph neural networks (GNNs), and physics-informed neural networks (PINNs). Key contributions of this review include a cross-source comparative analysis, domain-specific trend mapping over a decade, and the identification of gaps in methodological transparency. Findings reveal increasing use of hybrid models, growing interest in interpretable and physically grounded techniques, and persistent underreporting of AI methodologies in the literature. This review provides actionable insights and research directions toward developing intelligent, explainable, sustainable energy systems.
- Open Access
- 16 May 2025
- Published by Caravel Press in Energy Catalyst
- Vol. 1, 35-53
Machine Learning-Based Predictive Models for Indoor Air Quality and Thermal Comfort: Bridging Sensor Data and Human Perception in Healthcare Facilities
Poor indoor air quality (IAQ) and inadequate thermal comfort continue to challenge healthcare facilities in Malaysia, especially in rural areas where poor ventilation and high humidity are prevalent. These environmental stressors negatively impact patient outcomes, staff well-being, and operational efficiency. Existing solutions often overlook the integration of subjective human perception with objective sensor data, resulting in limited adaptability and responsiveness. This study introduces a datadriven, human-centric framework that combines environmental sensor measurements with user-reported comfort feedback to develop predictive models for IAQ and thermal comfort. A comprehensive machine learning pipeline was implemented using four algorithms—Random Forest (RF), XGBoost, Artificial Neural Network (ANN), and Support Vector Machine (SVM)—to model both continuous IAQ indicators and categorical thermal preferences. Experimental results show that Random Forest achieved the best overall performance, with the lowest root mean squared error (RMSE = 14.35) in regression and the highest classification accuracy (87.5%) in predicting thermal preference. Statistical validation confirmed that Random Forest and XGBoost performed similarly in regression, while Random Forest and SVM showed no significant differences in classification accuracy. These findings validate Random Forest as a robust and consistent model across both tasks. This study contributes a validated, AI-enhanced framework for intelligent environmental monitoring in healthcare settings, emphasizing the integration of subjective and objective data streams. The approach supports personalized, data-driven interventions and offers practical insights for facility managers, clinicians, and policymakers aiming to optimize indoor conditions. Future work will focus on scaling the dataset across diverse facilities and climates, enabling real-time deployment, and incorporating explainable AI techniques to enhance model transparency and stakeholder trust.
- 16 May 2025
- Published by Caravel Press in Energy Catalyst
- Vol. 1, 35-53
Open AccessMachine Learning-Based Predictive Models for Indoor Air Quality and Thermal Comfort: Bridging Sensor Data and Human Perception in Healthcare Facilities
Poor indoor air quality (IAQ) and inadequate thermal comfort continue to challenge healthcare facilities in Malaysia, especially in rural areas where poor ventilation and high humidity are prevalent. These environmental stressors negatively impact patient outcomes, staff well-being, and operational efficiency. Existing solutions often overlook the integration of subjective human perception with objective sensor data, resulting in limited adaptability and responsiveness. This study introduces a datadriven, human-centric framework that combines environmental sensor measurements with user-reported comfort feedback to develop predictive models for IAQ and thermal comfort. A comprehensive machine learning pipeline was implemented using four algorithms—Random Forest (RF), XGBoost, Artificial Neural Network (ANN), and Support Vector Machine (SVM)—to model both continuous IAQ indicators and categorical thermal preferences. Experimental results show that Random Forest achieved the best overall performance, with the lowest root mean squared error (RMSE = 14.35) in regression and the highest classification accuracy (87.5%) in predicting thermal preference. Statistical validation confirmed that Random Forest and XGBoost performed similarly in regression, while Random Forest and SVM showed no significant differences in classification accuracy. These findings validate Random Forest as a robust and consistent model across both tasks. This study contributes a validated, AI-enhanced framework for intelligent environmental monitoring in healthcare settings, emphasizing the integration of subjective and objective data streams. The approach supports personalized, data-driven interventions and offers practical insights for facility managers, clinicians, and policymakers aiming to optimize indoor conditions. Future work will focus on scaling the dataset across diverse facilities and climates, enabling real-time deployment, and incorporating explainable AI techniques to enhance model transparency and stakeholder trust.
- Open Access
- 7 October 2024
- Published by Akademia Baru Publishing in Journal of Advanced Research in Applied Sciences and Engineering Technology
- Vol. 55 (1) , 44-62
An Integrated Approach: Watershed Segmentation with Local Maxima and Minima Algorithms for Tree Crown Delineation of Mango (Mangifera indica) using UAV Multispectral Imagery
Precision forestry and agriculture rely on precise tree crown delineation as a fundamental component for effective forest management activities. The selection of appropriate techniques and algorithms for tree crown delineation significantly influences the accuracy and reliability of the outcomes. This study was conducted in a mango plantation at Universiti Teknologi MARA, Arau, Perlis, employing Unmanned Aerial Vehicle (UAV) multispectral imaging. The focus is to assess the accuracy of tree crown delineation using two different algorithms, namely watershed segmentation and local maxima and minima. The conventional method was used for reference parameter derivation and tree positioning, while object-based image analysis was employed for processing the multispectral images. Additionally, a manual digitization approach was utilized to conduct an accuracy assessment of the resulting tree crown delineation. The study findings indicated that the watershed algorithm exhibited superior accuracy, achieving 83.3% total 1:1 match and 73% goodness of fit, compared to the local maxima and minima algorithm, which yielded 81.7% and 71%, respectively. These results hold significant implications for forest management planning and remote sensing-based forest quantification estimation, enabling the development of efficient and effective forest management strategies for the future.
- 7 October 2024
- Published by Akademia Baru Publishing in Journal of Advanced Research in Applied Sciences and Engineering Technology
- Vol. 55 (1) , 44-62
Open AccessAn Integrated Approach: Watershed Segmentation with Local Maxima and Minima Algorithms for Tree Crown Delineation of Mango (Mangifera indica) using UAV Multispectral Imagery
Precision forestry and agriculture rely on precise tree crown delineation as a fundamental component for effective forest management activities. The selection of appropriate techniques and algorithms for tree crown delineation significantly influences the accuracy and reliability of the outcomes. This study was conducted in a mango plantation at Universiti Teknologi MARA, Arau, Perlis, employing Unmanned Aerial Vehicle (UAV) multispectral imaging. The focus is to assess the accuracy of tree crown delineation using two different algorithms, namely watershed segmentation and local maxima and minima. The conventional method was used for reference parameter derivation and tree positioning, while object-based image analysis was employed for processing the multispectral images. Additionally, a manual digitization approach was utilized to conduct an accuracy assessment of the resulting tree crown delineation. The study findings indicated that the watershed algorithm exhibited superior accuracy, achieving 83.3% total 1:1 match and 73% goodness of fit, compared to the local maxima and minima algorithm, which yielded 81.7% and 71%, respectively. These results hold significant implications for forest management planning and remote sensing-based forest quantification estimation, enabling the development of efficient and effective forest management strategies for the future.
- Open Access
- 28 June 2024
- Published by Universitas Airlangga in Journal of Information Systems Engineering and Business Intelligence
- Vol. 10 (2) , 232-249
A Systematic Literature Review on Leaf Disease Recognition Using Computer Vision and Deep Learning Approach
Background: Plant diseases affect agricultural output, quality and profitability, making them serious obstacles for agriculture. It is essential to detect diseases early in order to reduce losses while retaining sustainable practices. Plant disease detection has benefited greatly from the use of computer vision and deep learning in recent years because of their outstanding precision and computing capability. Objective: In this paper, we intend to investigate the role of deep learning in computer vision for plant disease detection while looking into how these techniques address complex disease identification problems. A variety of deep learning architectures were reviewed, and the contribution of frameworks such as Tensorflow, Keras, Caffe and PyTorch to the researchers' model construction was studied as well. Additionally, the usage of open repositories such as PlantVillage and Kaggle along with the customized datasets were discussed. Methods: We gathered the most recent developments in deep learning techniques for leaf disease detection through a systematic literature review of research papers published over the past decade, using reputable academic databases like Scopus and Web of Science, following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) method for guidance. Results: This study finds that researchers consistently enhance existing deep learning architectures to improve prediction accuracy in plant disease detection, often by introducing novel architectures and employing transfer learning methods. Frameworks like TensorFlow, Keras, Caffe, and PyTorch are widely favored for their efficiency in development. Additionally, most studies opt for public datasets such as PlantVillage, Kaggle, and ImageNet, which offer an abundance of labelled data for training and testing deep learning models. Conclusion: While no singular ‘best' model emerges, the adaptability of deep learning and computer vision demonstrates the dynamic nature of plant disease recognition area, and this paper provides a comprehensive overview of deep learning's transformative impact on plant disease recognition by bringing together information from different studies. Keywords: Deep learning, Computer vision, Plant disease, Systematic literature review
- 28 June 2024
- Published by Universitas Airlangga in Journal of Information Systems Engineering and Business Intelligence
- Vol. 10 (2) , 232-249
Open AccessA Systematic Literature Review on Leaf Disease Recognition Using Computer Vision and Deep Learning Approach
Background: Plant diseases affect agricultural output, quality and profitability, making them serious obstacles for agriculture. It is essential to detect diseases early in order to reduce losses while retaining sustainable practices. Plant disease detection has benefited greatly from the use of computer vision and deep learning in recent years because of their outstanding precision and computing capability. Objective: In this paper, we intend to investigate the role of deep learning in computer vision for plant disease detection while looking into how these techniques address complex disease identification problems. A variety of deep learning architectures were reviewed, and the contribution of frameworks such as Tensorflow, Keras, Caffe and PyTorch to the researchers' model construction was studied as well. Additionally, the usage of open repositories such as PlantVillage and Kaggle along with the customized datasets were discussed. Methods: We gathered the most recent developments in deep learning techniques for leaf disease detection through a systematic literature review of research papers published over the past decade, using reputable academic databases like Scopus and Web of Science, following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) method for guidance. Results: This study finds that researchers consistently enhance existing deep learning architectures to improve prediction accuracy in plant disease detection, often by introducing novel architectures and employing transfer learning methods. Frameworks like TensorFlow, Keras, Caffe, and PyTorch are widely favored for their efficiency in development. Additionally, most studies opt for public datasets such as PlantVillage, Kaggle, and ImageNet, which offer an abundance of labelled data for training and testing deep learning models. Conclusion: While no singular ‘best' model emerges, the adaptability of deep learning and computer vision demonstrates the dynamic nature of plant disease recognition area, and this paper provides a comprehensive overview of deep learning's transformative impact on plant disease recognition by bringing together information from different studies. Keywords: Deep learning, Computer vision, Plant disease, Systematic literature review
- Open Access
- 1 June 2024
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 13 (2) , 1398-1407
Python scikit-fuzzy: developing a fuzzy expert system for diabetes diagnosis
Nowadays, improvements in diabetes detection that provide patients with vital information are needed. This is due to the fact that Diabetes mellitus has generated a worldwide epidemic that costs society and people. Also, patients tend to misread symptoms, and clinicians who collect insufficient data may produce erroneous outcomes. Therefore, this study aims to demonstrate that a programme that integrates expert advice such as decisions, recommendations, or solutions is an excellent method for reducing the incidence of diabetes. Specifically, this study intends to implement a fuzzy expert system that can detect and report the early stages of diabetes as a viable approach. Furthermore, since this programme is available to everyone, people may easily self-diagnose themselves if they have a blood glucose monitoring device. However, developing the fuzzy expert system for real-world situations, such as diabetes patients, using any programming tools is not straightforward. Therefore, this study will provide a comprehensive approach to constructing a fuzzy expert system using the popular programming language Python.
- 1 June 2024
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 13 (2) , 1398-1407
Open AccessPython scikit-fuzzy: developing a fuzzy expert system for diabetes diagnosis
Nowadays, improvements in diabetes detection that provide patients with vital information are needed. This is due to the fact that Diabetes mellitus has generated a worldwide epidemic that costs society and people. Also, patients tend to misread symptoms, and clinicians who collect insufficient data may produce erroneous outcomes. Therefore, this study aims to demonstrate that a programme that integrates expert advice such as decisions, recommendations, or solutions is an excellent method for reducing the incidence of diabetes. Specifically, this study intends to implement a fuzzy expert system that can detect and report the early stages of diabetes as a viable approach. Furthermore, since this programme is available to everyone, people may easily self-diagnose themselves if they have a blood glucose monitoring device. However, developing the fuzzy expert system for real-world situations, such as diabetes patients, using any programming tools is not straightforward. Therefore, this study will provide a comprehensive approach to constructing a fuzzy expert system using the popular programming language Python.