Articles

PMGHWO: A particle-guided adaptive grey wolf optimizer for large-scale traveling salesman problems

10.11591/ijece.v16i5.pp2819-2835

The traveling salesman problem (TSP) remains challenging for metaheuristic algorithms, especially as the number of cities increases. This study proposes the particle-guided adaptive grey wolf optimizer (PMGHWO), which combines grey wolf optimizer (GWO)-based leadership guidance with crossover, particle-guided perturbation, and adaptive local refinement. PMGHWO was evaluated on ten TSPLIB instances and compared with grey wolf optimizer (GWO), genetic algorithms (GA), particle swarm optimization (PSO), whale optimization algorithm (WOA), and Harris Hawks optimization (HHO) under the same experimental conditions. The proposed method achieved shorter average tours on all tested instances, with reductions of 18.75% compared with GWO, 19.77% with HHO, 23.30% with GA, 39.08% with WOA, and 57.51% with PSO. Friedman and Wilcoxon tests confirmed significant differences between PMGHWO and the compared methods. The ablation study showed that adaptive refinement had the strongest effect on solution quality, while the influence of particle-guided perturbation and crossover varied across the benchmark instances. Although PMGHWO required more computation than the original GWO, its runtime remained competitive with several of the other methods. Overall, the results show that PMGHWO is a promising approach for TSP optimization and warrants further evaluation on larger and more complex combinatorial problems.

Volume: 16

Issue: 5

Page: 2819-2835

Publish at: 2026-10-01

Cognitive cyber-physical systems: integrating intelligence, adaptation, and real-time decision-making

10.11591/ijece.v16i5.pp2265-2271

Tole Sutikno

Cognitive cyber-physical systems (CCPS) represent a critical inflection point in the evolution of intelligent engineering systems, where computation, learning, and physical interaction are no longer separated but fused within a continuous adaptive loop. This editorial positions CCPS as the intermediate layer between foundational computational intelligence and fully autonomous socio-cognitive systems, extending the broader IJECE 2026 intellectual trajectory. In contrast to conventional cyber-physical systems governed by predefined control logic, CCPS embed learning-driven intelligence directly into sensing, communication, and control processes, enabling continuous real-time adaptation in uncertain and dynamically evolving environments. Intelligence in this paradigm is realized as embodied intelligence through continuous interaction across computational and physical layers rather than remaining confined to localized algorithmic modules. Tight coupling among perception, decision-making, and actuation enables continuous refinement of internal representations through environmental feedback. Multi-scale learning mechanisms further support simultaneous fast-response adaptation and long-horizon behavioral optimization. As a result, classical feedback architectures are transformed into predictive cognitive loops capable of anticipatory, context-aware, and autonomous operation under uncertainty. CCPS therefore provide the engineering realization of embodied intelligence, establishing the foundational bridge between computational intelligence and the emergence of future socio-cognitive systems.

Volume: 16

Issue: 5

Page: 2265-2271

Publish at: 2026-10-01

Robust resource allocation in multi-cell UE-specific RIS-assisted D2D relay networks under imperfect CSI

10.11591/ijece.v16i5.pp2575-2594

Kayode Popoola , Ayodeji Ajani , Stuart Nicholson , Muheeb Ahmed , Srilatha Narayangari Pamuri , Ibrahim Bala Alhassan

Device-to-device (D2D) communication enhances spectral efficiency but remains constrained by limited transmission range, underlay interference, and the half-duplex overhead of conventional relays. User equipment-specific reconfigurable intelligent surfaces (UE-RIS) offer a promising alternative by enabling passive beamforming to strengthen D2D links without additional spectrum consumption. However, existing studies typically assume perfect channel state information (CSI) and single-cell operation, limiting their applicability to practical deployments. This paper proposes a robust multi-cell resource allocation (RMRA) framework for UE-RIS-assisted D2D relay networks under imperfect CSI. A hybrid uncertainty model is adopted, combining statistical Gauss-Markov CSI errors for intra-cell links with bounded norm-ball uncertainty for inter-cell links. The joint optimisation of resource reuse, transmit power allocation, and RIS phase configuration is formulated as a stochastic mixed-integer nonlinear program that maximises network spectral efficiency while satisfying outage and quality-of-service constraints. To efficiently solve the problem, a three-stage algorithm is proposed comprising distance-pruned Hungarian assignment, robust power control using Bernstein-type inequality and S-procedure based semidefinite programming, and soft actor-critic (SAC) based passive beamforming. Simulation results show that RMRA achieves a 94% D2D access rate at light load and over 75% at full load, improves sum spectral efficiency by 34.7% and 70.2% over AF relaying and direct D2D, respectively, attains 118.5 bits/s/Hz/W energy efficiency, and maintains 30.2 bits/s/Hz under severe CSI uncertainty.

Volume: 16

Issue: 5

Page: 2575-2594

Publish at: 2026-10-01

Contour-guided convexity defect geometry for precise ROI extraction in dorsal hand vein recognition

10.11591/ijece.v16i5.pp2622-2640

Habib Kadem , Merah Mostefa

Dorsal hand vein biometrics depend strongly on accurate region-of-interest (ROI) extraction, yet conventional convexity-defect-based methods remain vulnerable to contour noise, skin texture, and shadow-induced artifacts. This paper introduces contour-guided convexity defect geometry (CG-CDG), a geometry-driven ROI extraction framework in which a candidate convexity defect is accepted as an anatomical valley only if it jointly satisfies two independent criteria: chord-contour topological consistency, ensuring anatomical relevance, and a minimum enclosed-area threshold, ensuring physical plausibility. Unlike prior convexity-defect-based methods, which accept a candidate valley on the basis of a single depth- or angle-based criterion evaluated in isolation, CG-CDG applies this joint plausibility test before defining the region of interest, a square ROI centered on the hand centroid and aligned with the dominant valley orientation. Experiments on a public dorsal hand vein database of 1,024 images from 138 subjects show that, combined with HOG descriptors and City Block distance, CG-CDG achieves an equal error rate (EER) of 0.0266, with a 95% bootstrap confidence interval of [0.0241, 0.0289] over the matching scores (B = 1,000 resamples), the narrowest interval among the four ROI extraction methods compared under the primary evaluation setting. A post-hoc Wilcoxon signed-rank test with Bonferroni correction confirms that this improvement is statistically significant over all competing methods (p < 0.001). On this dataset, geometric validation substantially reduces intra-class variance and yields statistically validated performance without requiring any training data, making CG-CDG a strong candidate both as a standalone, training-free biometric solution and as an anatomically grounded preprocessing stage for hybrid deep-learning pipelines.

Volume: 16

Issue: 5

Page: 2622-2640

Publish at: 2026-10-01

Temporal deep representation learning based remaining useful lifecyle prediction of electric vehicle lithium-ion batteries

10.11591/ijece.v16i5.pp2365-2378

Pradish Vaidya S. , Kevin Denzil V. , Monish S. , S. Cloudin , S. Saradha

In recent years, rapid growth of renewable energy systems and electric vehicle technologies has intensified the focus on lithium-ion batteries as critical energy storage components. Consequently, evaluating their performance and lifespan has become a significant concern in both scientific research and industrial applications. The accurate remaining useful life (RUL) prediction of lithium-ion batteries are major importance to increase energy management efficacy and increase the battery lifetime. Recently, the quick advancement of machine learning (ML) and artificial intelligence (AI) knowledge, data-driven approaches for forecasting the lithium-ion battery duration are involved extensive attention. Numerous researchers are employing deep learning (DL) approaches to make several techniques for predicting RUL, namely recurrent neural network (RNN) and convolutional neural network (CNN). This paper introduces a temporal deep representation learning based remaining useful Lifecyle prediction (TRDL-RULP) approach for electric vehicle lithium-ion batteries. The main goal of the TRDL-RULP framework is to support intelligent battery management methods and contribute to increasing the operational safety and lifecycle management of electric vehicle energy reserves systems. Initially, raw battery degradation data are processed using data normalization to eliminate scale variations and improve model stability. To enhance feature relevance and reduce dimensionality, a snake optimization-based feature selection strategy is employed to enable the extraction of the most informative degradation indicators. For RUL prediction, a long short-term memory autoencoder has been exploited to capture nonlinear temporal dependencies and extract robust latent representations. Finally, the tuna swarm optimization algorithm can be applied for optimal optimize parameters to improve convergence speed and predictive performance. The simulation study of the proposed TRDL-RULP algorithm is conducted using a benchmark lithium-ion battery degradation dataset obtained from the Kaggle repository. Extensive comparative results demonstrate the superior performance of the TRDL-RULP approach over recent methodologies.

Volume: 16

Issue: 5

Page: 2365-2378

Publish at: 2026-10-01

Incremental pattern mining with novelty-based pruning for dynamic analysis and adaptive e-learning recommendations

10.11591/ijece.v16i5.pp2607-2621

Emad Mahmoud Alazazi , Ahmed Sultan Alhegami

Traditional pattern mining algorithms become increasingly inefficient in dynamic e learning environments, where learner interactions accumulate continuously. When new data arrive, these algorithms typically discard previously discovered patterns and recompute everything from scratch—a costly process that this work seeks to avoid. The proposed incremental intelligent maximal frequent itemsets (IIMFIs) framework integrates three complementary mechanisms: incremental maintenance, maximal pattern extraction, and novelty based pruning. Rather than restarting computation with each new batch, IIMFIs stores and selectively updates previously discovered patterns. The novelty threshold \theta controls pruning of highly similar candidates, further reducing the search space. Experiments on the OULAD dataset split chronologically into four batches—demonstrate several advantages. First, IIMFIs compresses the pattern set considerably: at the lowest support threshold, it reduces 2,597 patterns generated by classical algorithms to only 342, achieving a compression factor of 7.6×. Second, incremental updates achieve a 4.2× speedup over full restart and run 28% faster than restarting FP Growth. Third, activating novelty pruning (θ=0.7) improves recommendation precision by 15.5% while maintaining diversity above 0.95. Notably, F1@10 scores remain identical to those obtained from full recomputation, confirming that compactness does not compromise quality. These results demonstrate that IIMFIs provides an efficient and scalable solution for adaptive e-learning environments.

Volume: 16

Issue: 5

Page: 2607-2621

Publish at: 2026-10-01

A comparative analysis of hybrid FFNN-LSTM and FFNN-RNN architectures for short term electricity load forecasting

10.11591/ijece.v16i5.pp2347-2356

Temitope Akinyede , Josephine Adenike Akinyede , Paul Kehinde Olulope , Emmanuel Taiwo Fasina , Temitope Adewale Olominu

Short-term accurate forecasting of electricity demand is crucial for power-system operation and energy scheduling and the equilibrium between electricity generation and consumption. The predicting of electric power consumption is however difficult because electricity-demand profiles are non-linear and time varying. In this paper, we consider two hybrid deep learning architectures feedforward neural network long short-term memory (FFNN-LSTM) and feedforward neural network recurrent neural network (FFNN-RNN) for multi-horizon electricity-load forecasting. The architectures we proposed combine the ability of FFNN to represent data non-linearly with the sequential modelling capability of LSTM and RNN. The authors assess the models based on historical electricity-load observations from Ado-Ekiti, at three different forecasting horizons of 24 hours, 72 hours, and 168 hours. Root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) assessed predictive performance. The findings reveal that FFNN-LSTM achieves a consistently lower RMSE across the evaluated horizons, resulting in greater effectiveness to confine relatively large forecasting errors. On the other hand, the separate LSTM and RNN models achieve lower MAE and MAPE in various scenarios, suggesting stronger short-term reaction to electricity demand. The results, therefore, show a trade-off between forecast stability and sensitivity to rapid load changes. In general, the performance of hybrid architecture is more stable compared to the standalone recurrent models which are more responsive to short-term changes when we look at most of the forecasting horizons. According to the research paper, the forecasting architectures can now be selected for smart grid applications.

Volume: 16

Issue: 5

Page: 2347-2356

Publish at: 2026-10-01

Functional safety approach for thermal runaway in battery management systems

10.11591/ijece.v16i5.pp2405-2416

Nouhaila Belmajdoub , Rachid Lajouad , Abdelmounime El Magri , Soukaina Boudoudouh

Forecasting demand for autonomous power storage and supply systems for electric vehicles (EV) has become a key area of research. The rapid popularization of EVs is heavily based on the reliability and safety of rechargeable energy storage systems (ESSs), especially lithium-ion batteries (LIBs). Although these batteries offer high energy density and efficiency, they remain vulnerable to hazardous events such as thermal runaway, leading to severe consequences, including fire and explosion. This paper presents a methodology to fulfill the functional safety of BMS thermal runaway according to ISO 26262 standard. The approach integrates hazard analysis and risk assessment (HARA), the formulation of safety goals (SG), and the specification of functional and technical safety requirements (FSR and TSR). To strengthen assurance, safety contracts based on battery chemistry and specifications are established and linked to structured safety cases using Goal Structuring Notation (GSN). The proposed methodology is validated through simulations in MATLAB Simulink, demonstrating how dynamic safety assurance can detect deviations, trigger control actions, and update safety cases to prevent unsafe states. The results show that the framework effectively minimizes the risks related with over-current, over-voltage, and over-temperature conditions, thus contributing to regulatory compliance and increased confidence in the safety of EVs.

Volume: 16

Issue: 5

Page: 2405-2416

Publish at: 2026-10-01

Design science research in developing a religious chatbot based on Bulugh al-Maram

10.11591/ijece.v16i5.pp2782-2794

Aris Tjahyanto , Irmasari Hafidz , Faizal Johan Atletiko

Chatbots have recently gained significant popularity. For instance, ChatGPT has become a preferred tool for many individuals seeking instant answers without relying on human responses. This immediacy sets chatbots apart from books, which require users to search for information manually. This time-consuming process does not align with millennials' preference for convenience and efficiency. Studying hadith independently using the Bulugh al-Maram book demands considerable time and effort. The limited use of natural language processing technologies in religious chatbots restricts their ability to handle complex inquiries effectively. A chatbot capable of answering hadith-related questions could greatly assist the public in studying hadith texts by providing direct responses without extensive searching. This chatbot was designed for web browsers, utilizing deep learning as its core technology. This research led to the development of a chatbot prototype for learning hadith from Bulugh al-Maram. Built using the design science research (DSR) methodology, the prototype achieves an intent recognition rate (IRR) of 86.82%. However, its capabilities are below the BERT model, demonstrating a strong ability to accurately interpret user questions and statements.

Volume: 16

Issue: 5

Page: 2782-2794

Publish at: 2026-10-01

Performance and quality analysis of brain MRI image transmission over free-space optical communication systems under severe atmospheric conditions

10.11591/ijece.v16i5.pp2526-2536

Entidhar Mhawes Zghair , Seham Hashem , Ali Hammadi

Reliable transfer of brain magnetic resonance imaging (MRI) data over atmospheric free-space optical (FSO) links is a key enabler of telemedicine, yet conventional FSO studies judge link quality by communication metrics such as the bit error rate (BER) alone, which cannot guarantee the structural and contrast fidelity that diagnosis demands. This study proposes a quality-aware FSO transmission framework for brain MRI in which link performance is assessed jointly through BER, peak signal-to-noise ratio (PSNR), and the structural similarity index (SSIM). A physical-layer FSO channel is modelled in OptiSystem 20 and co-simulated with MATLAB R2023b, which performs image serialization, reconstruction, and quality analysis. Thirty axial T2-weighted slices (256×256, 8-bit) from the public IXI dataset are transmitted at 1550 nm over clear-air, rain, and fog channels at 500, 1000, and 2000 m. Adopting conservative diagnostic thresholds of PSNR ≥ 30 dB and SSIM ≥ 0.85, the link is diagnostically usable in clear air at all tested distances (PSNR = 42.1 dB, SSIM = 0.98 at 500 m) and in rain up to 2000 m (PSNR ≥ 31.2 dB), whereas fog degrades quality below the thresholds at every distance, reaching PSNR = 22.7 dB and SSIM = 0.68 at 2000 m. A concatenated forward-error-correction (FEC) scheme is then shown to restore diagnostic quality under fog up to 1000 m, extending the usable fog range, while 2000 m remains infeasible and motivates hybrid FSO/RF operation. The framework provides quantitative deployment limits for FSO-based medical image transport.

Volume: 16

Issue: 5

Page: 2526-2536

Publish at: 2026-10-01

Information architecture debt: Why legacy platform schema decisions constrain enterprise AI capability

10.11591/ijece.v16i5.pp2473-2482

Mihir Shah

Enterprise platforms accumulate a specific category of technical debt this paper terms information architecture debt. Schema decisions optimized for transactional efficiency in earlier computing eras produce structural constraints that limit what artificial intelligence can accomplish on those platforms, largely independent of which models are selected or how they are orchestrated. This paper positions information architecture debt as a category distinct from code debt and infrastructure debt, and it argues that the distinction matters because remediation locality and coordination cost differ sharply. A five-indicator diagnostic framework is proposed, covering duplicate canonical entities, broken semantic chains, provenance gaps, enforcement asymmetry, and consumer assumption divergence. The framework supports a capability ceiling hypothesis: substrate defects impose an upper bound on AI outcomes that no model choice appears able to exceed. A sequenced remediation approach follows, prioritizing entity resolution, semantic alignment, and provenance instrumentation by AI-capability impact rather than ease of fix. Grounded in practitioner experience structuring enterprise information architecture across multi-language, multi-system platforms, the framework offers leaders diagnostic and sequencing tools, and it establishes a conceptual foundation for later quantitative and sector-specific refinement.

Volume: 16

Issue: 5

Page: 2473-2482

Publish at: 2026-10-01

Uncovering hidden predictors of teacher knowledge and attitudes in child sexual abuse prevention

10.11591/ijere.v15i5.38744

Tetti Solehati , Helmy Hazmi , Rachelya Nurfirdausi Islamah , Yanti Hermayanti , Cecep Eli Kosasih , Mira Trisyani Koeryaman , Muhammad Yusuf

Teachers play a critical role in preventing child sexual abuse (CSA) in school settings. However, empirical evidence on how sociodemographic characteristics and information exposure shape teachers’ knowledge and attitudes remains limited, particularly in low- and middle-income contexts. This study aimed to examine the associations between teachers’ sociodemographic characteristics and exposure to CSA-related information with their knowledge and attitudes toward prevention. A quantitative cross-sectional study was conducted from March to December 2024 involving 56 teachers from two districts in West Java, Indonesia, using total population sampling. Data were collected using structured questionnaires and analyzed using descriptive statistics and chi-square tests, with Phi (φ) and Cramér’s V used to assess the strength of associations. Teachers’ knowledge was significantly associated with gender and information sources, including newspapers, Instagram, and headmasters (p

Volume: 15

Issue: 5

Page: 4026-4037

Publish at: 2026-10-01

Primary science teachers’ perceptions of digital tools: a qualitative study of challenges and pedagogical implications

10.11591/ijere.v15i5.37255

Wenlin Yang , Sirinapa Kijkuakul

This qualitative study investigates primary school science teachers’ attitudes, experiences, and challenges in integrating digital tools into science instruction. Drawing on semi-structured interviews with 25 teachers from urban and rural schools in Meizhou City, Guangdong Province, China, thematic analysis explored how simulations, virtual labs, and multimedia platforms are employed, their perceived benefits, and systemic barriers to effective use. Findings indicate that digital tools support visualization of abstract concepts, supplement limited physical resources, and enhance student engagement through interactive experimentation. However, challenges such as misalignment with curriculum goals, insufficient technical support, and concerns about over-reliance on digital methods potentially diminishing hands-on learning were identified. Full-time and higher-grade teachers showed more consistent integration than part-time or lower-grade counterparts. The findings indicate a need to balance digital approaches with experiential learning, while reaffirming the continued importance of tactile, hands-on activities in science education. For educators and policymakers, the findings highlight priority areas for action, including professional development (PD) aligned with curricular goals, improved availability of digital resources across school contexts, and institutional mechanisms that guide teachers’ use of digital tools in practice.

Volume: 15

Issue: 5

Page: 4573-4587

Publish at: 2026-10-01

Predictive and fault-tolerant virtual machine migration for energy-efficient cloud data centers

10.12928/telkomnika.v24i5.27706

Rakshan; Sri Venkateswara College of Engineering G. K. , P.; Sri Venkateswara College of Engineering Vinothiyalakshmi

Virtual machine (VM) migration is a key mechanism for improving energy efficiency, service continuity and reliability in cloud data centers. However, conventional migration strategies are largely reactive and fail to account for workload fluctuations and potential failures, often resulting in inefficient resource utilization and increased service-level agreement (SLA) violations. This paper proposes the predictive VM migration manager (PVMM), a unified framework that integrates workload forecasting and failure prediction for proactive migration control. PVMM combines a gated recurrent unit (GRU)-based model for short-term resource prediction with a machine-learning-based failure predictor to identify high-risk VMs. An adaptive dynamic threshold based on energy consumption (ADT-EC) detects host overload conditions, while an improved energy-aware best-fit (IEABF) heuristic selects optimal migration targets. Experimental results demonstrate that PVMM reduces unnecessary migration, improves SLA compliance and enhances overall energy efficiency and system reliability. These findings highlight the effectiveness of predictive, multi-objective migration strategies for next-generation cloud data center management.

Volume: 24

Issue: 5

Page: 1513-1525

Publish at: 2026-10-01

An impact of activation functions on CNN-Bi-LSTM SoC estimation for Li-Ion batteries

10.11591/ijece.v16i5.pp2357-2364

Monica K. M. , Abhay A. Deshpande

Estimation of state of charge (SoC) in Li-ion batteries has been accomplished by many methods over couple of years. Every research in this domain majorly focusses on improving accuracy of estimation and some in exploring new algorithms. In this regard, we worked in analyzing the accuracy of estimation of SoC of sophisticatedly accurate techniques. A data-driven SoC estimation using deep neural network architectures (CNN) were collaborated with different activation functions like global pooling, exponential linear unit, Leaky rectified linear unit on convolutional neural network (CNN) and bidirectional-long short-term memory (Bi-LSTM) models. The performance of all these models was evaluated using quantitative error analysis and qualitative signal tracking studies to compare the estimated and actual SoC trajectory. This work consolidates the performances of different models and helps realize how best a fusion model works on the best accuracy in estimation.

Volume: 16

Issue: 5

Page: 2357-2364

Publish at: 2026-10-01

Access the latest knowledge in applied science, electrical engineering, computer science and information technology, education, and health.

Filters article

Years

Source Title

Authors

31,291 Article Results

PMGHWO: A particle-guided adaptive grey wolf optimizer for large-scale traveling salesman problems

Cognitive cyber-physical systems: integrating intelligence, adaptation, and real-time decision-making

Robust resource allocation in multi-cell UE-specific RIS-assisted D2D relay networks under imperfect CSI

Contour-guided convexity defect geometry for precise ROI extraction in dorsal hand vein recognition

Temporal deep representation learning based remaining useful lifecyle prediction of electric vehicle lithium-ion batteries

Incremental pattern mining with novelty-based pruning for dynamic analysis and adaptive e-learning recommendations

A comparative analysis of hybrid FFNN-LSTM and FFNN-RNN architectures for short term electricity load forecasting

Functional safety approach for thermal runaway in battery management systems

Design science research in developing a religious chatbot based on Bulugh al-Maram

Performance and quality analysis of brain MRI image transmission over free-space optical communication systems under severe atmospheric conditions

Information architecture debt: Why legacy platform schema decisions constrain enterprise AI capability

Uncovering hidden predictors of teacher knowledge and attitudes in child sexual abuse prevention

Primary science teachers’ perceptions of digital tools: a qualitative study of challenges and pedagogical implications

Predictive and fault-tolerant virtual machine migration for energy-efficient cloud data centers

An impact of activation functions on CNN-Bi-LSTM SoC estimation for Li-Ion batteries

Discover Our Library